| clarkb | notmorgan: maybe? not sure if ubuntu or python proper | 00:00 |
|---|---|---|
| notmorgan | clarkb: (on xenial that is) | 00:00 |
| notmorgan | i'd be surprised this "changed" randomely between 2.7.x and 2.7.x+ | 00:00 |
| ianw | clarkb: yeah, that does *not* work of fedora, for example | 00:00 |
| notmorgan | or it's a backport? | 00:00 |
| clarkb | ok sounds like it msotly doesn't work anywhere and I should just write a patch to fix it for cerberus | 00:01 |
| notmorgan | the syntax looks very odd fwiw | 00:01 |
| mordred | I can't import _uuid_generate_random on fedora at all | 00:02 |
| mordred | regardless of syntax | 00:02 |
| notmorgan | mordred: is that actually what is intended? | 00:02 |
| clarkb | mordred: from uuid import uuid4 as _uuid_generate_random | 00:02 |
| clarkb | mordred: ^ that doesnt' work? | 00:02 |
| mordred | oh. WOW | 00:02 |
| notmorgan | mordred: it seems like it's naming uuid4 as _uuid..yea | 00:02 |
| notmorgan | the syntax is really unintuative | 00:03 |
| mordred | really? | 00:03 |
| clarkb | because that from as foo works for me if I go one at a time | 00:03 |
| mordred | the person wants to import uuid4 as two different names? | 00:03 |
| notmorgan | mordred: yeah | 00:03 |
| clarkb | its just together that it breaks | 00:03 |
| notmorgan | mordred: from uuid import UUID, uuid4 as _uuid4 \n _uuid_generate_random = _uuid4 | 00:03 |
| clarkb | anyways I think I am sufficiently happy to not try to understand python here and accept it | 00:03 |
| *** _sarob has quit IRC | 00:03 | |
| notmorgan | :P | 00:03 |
| mordred | yah | 00:03 |
| *** Sukhdev has quit IRC | 00:03 | |
| mordred | I think whoever is doing that is a bug | 00:04 |
| mordred | regardless of whether it worked somewhere | 00:04 |
| notmorgan | mordred: people can be bugs now? | 00:04 |
| notmorgan | :P | 00:04 |
| clarkb | ok python person elsewhere set me straight | 00:05 |
| ianw | mordred: i don't think so ... it just seems that ubuntu trusty has a "_uuid_generate_random" function, while others don't | 00:05 |
| clarkb | _uuid_generate_random was a private thing in older stdlib | 00:05 |
| clarkb | its gone now | 00:05 |
| notmorgan | OH | 00:05 |
| notmorgan | got it | 00:05 |
| clarkb | ianw: ya | 00:05 |
| notmorgan | yah that makes more sense. | 00:05 |
| clarkb | I completely misparsed the as stuff so ya tahts definitely a bug | 00:06 |
| notmorgan | also... importing internal interfaces = *facepalm* | 00:06 |
| clarkb | don't use private interfaces in stdlib | 00:06 |
| * clarkb ignores | 00:06 | |
| notmorgan | yes! | 00:06 |
| clarkb | my xenial list is getting very small | 00:06 |
| notmorgan | mordred, jeblair: finalizing gear stuff now, should we warn if py3 and bytes is passed for name/unique vs string? | 00:06 |
| ianw | interesting differences -> http://paste.openstack.org/show/532253/ | 00:06 |
| notmorgan | or error (explode) | 00:06 |
| notmorgan | since technically we say it needs to be a string. | 00:07 |
| mordred | ok. so - the thing I thought it was doing was the thing it was doing | 00:07 |
| notmorgan | mordred: yes. | 00:07 |
| clarkb | mordred: ya I just didn't aprse it well enough | 00:07 |
| mordred | so the people wrote buggy code by importing a private function | 00:07 |
| notmorgan | i parsed it the same as clarkb cause... don't use private interfaces unless you expect it to break horribly | 00:07 |
| clarkb | and also their unittests don't work because of dep version conflicts | 00:07 |
| notmorgan | and then still don't do it | 00:07 |
| clarkb | notmorgan: the uuid4 as _uuid4 threw me off | 00:07 |
| mordred | clarkb: is this one of our libs? or a dep of ours? | 00:07 |
| clarkb | oh they are giving things their own names ok | 00:08 |
| notmorgan | clarkb: also - this is an argument for one import per line | 00:08 |
| clarkb | mordred: its cerberus a project that uses zuul's python-jobs template which is what I sourced my test all the things list from | 00:08 |
| clarkb | notmorgan: ++ | 00:08 |
| notmorgan | because that would have been immediately clear | 00:08 |
| mordred | clarkb: cool. so we can fix it | 00:08 |
| *** gyee has quit IRC | 00:08 | |
| clarkb | mordred: well I think I am just going to send them an email | 00:08 |
| clarkb | since their tests don't work etiher | 00:08 |
| mordred | clarkb: or, rather, file a bug and tell them to fix it | 00:08 |
| mordred | yah | 00:08 |
| clarkb | ya that | 00:08 |
| jeblair | notmorgan: maybe blowup? | 00:08 |
| notmorgan | jeblair: i'm good with that. | 00:08 |
| mordred | "we're going to xenial, you don't work, you fix" | 00:09 |
| jeblair | also, we should have -1d the name cerberus | 00:09 |
| *** yamamoto has joined #openstack-infra | 00:09 | |
| jeblair | "sorry, there's already a kerberos" | 00:09 |
| *** anteaya has quit IRC | 00:10 | |
| notmorgan | jeblair: ok i'll make some kind of nice error happen in this case just for the sake of not giving an opaque "XXX doesn't have attribute encode" | 00:10 |
| mordred | "XXX doesn't have attribute encode" | 00:11 |
| mordred | "XXX doesn't have attribute encode" is the worst error | 00:11 |
| notmorgan | right? | 00:11 |
| mordred | it's almost as bad as the one about not supporting the buffer protocol or whatever | 00:11 |
| notmorgan | mordred: actually - AttributeError: 'bytes' object has no attribute 'encode' | 00:11 |
| clarkb | it last got code about 9 months ago too | 00:11 |
| *** Julien-zte has joined #openstack-infra | 00:11 | |
| notmorgan | but yuck. | 00:12 |
| * clarkb files a bug anyways and if anyone cares they can address it | 00:12 | |
| clarkb | https://bugs.launchpad.net/cerberus/+filebug uhm ok | 00:13 |
| mordred | ++ | 00:13 |
| clarkb | maybe we need to just retire it | 00:13 |
| *** Jeffrey4l has quit IRC | 00:13 | |
| *** thorst has joined #openstack-infra | 00:13 | |
| *** akshai has joined #openstack-infra | 00:14 | |
| clarkb | I am taking it off of my list now, if I can't file a bug then my ability to care is suddenly very low | 00:14 |
| *** yamamoto has quit IRC | 00:14 | |
| mordred | clarkb: so ... back to the glean thing ... | 00:14 |
| jeblair | dhellmann touched it last -- by hot potato rules, he owns it! | 00:14 |
| mordred | as it relates to people being able to use our images locally | 00:15 |
| mordred | clarkb: don't they need to config-drive anyway to get their ssh key into the image? | 00:15 |
| clarkb | mordred: oh right! I don't ebacuse I get mine baked in | 00:15 |
| mordred | right | 00:15 |
| clarkb | mordred: excellent point | 00:15 |
| mordred | so I tihnk we need to write up some good and clear docs on the config-drive thing | 00:16 |
| clarkb | ya you either have to config drive or make a new image with that as the base | 00:16 |
| mordred | like even "here's a script that will make a config drive with your ssh key and hostname" or whatever | 00:16 |
| clarkb | ya I think writing config drive docs is what we want to do | 00:16 |
| mordred | ++ | 00:16 |
| clarkb | a script that was make-config-drive.sh --hostname foo --pubkey ssh-rsa.... and wrote a .iso would be good | 00:17 |
| clarkb | and if you want something more complicated you can figure it out | 00:17 |
| clarkb | I will try to work on this | 00:17 |
| *** thorst has quit IRC | 00:18 | |
| *** thorst has joined #openstack-infra | 00:18 | |
| clarkb | mordred: you TC'd during poppy things, who is a good person to ask about their broken tests? | 00:19 |
| mordred | clarkb: shrug? | 00:21 |
| clarkb | ok | 00:21 |
| clarkb | they have a similar issue in that imports fail but this import is from poppy.manager.default import analytics | 00:21 |
| clarkb | I think maybe its circular? | 00:21 |
| clarkb | time to do manual testing without stevedore in he way | 00:21 |
| clarkb | aha! stevedore was masking the error | 00:22 |
| clarkb | no more sslv3 in xenial python | 00:22 |
| *** thorst has quit IRC | 00:22 | |
| clarkb | ianw: do you know if fedora is keeping sslv3 in python stdlib? I THink it was debian that removed it first then ubuntu inherited it not sure if upstream did too in which case fedora would get it | 00:23 |
| *** garyk has joined #openstack-infra | 00:23 | |
| ianw | not sure immediately sorry | 00:23 |
| *** hurgleburgler has quit IRC | 00:26 | |
| *** Swami has quit IRC | 00:26 | |
| *** matt-borland has quit IRC | 00:28 | |
| clarkb | https://bugs.launchpad.net/poppy/+bug/1602894 | 00:28 |
| openstack | Launchpad bug 1602894 in Poppy "ssl.PROTOCOL_SSLv3 no longer available in moder pythons" [Undecided,New] | 00:28 |
| clarkb | ianw: looks like upstream python won't expose it if the underlying libs don't expose it so likely same behavior on fedora | 00:28 |
| *** sputnik13 has joined #openstack-infra | 00:30 | |
| *** yuanying has quit IRC | 00:30 | |
| *** jamesmcarthur has quit IRC | 00:30 | |
| *** garyk1 has joined #openstack-infra | 00:30 | |
| *** garyk has quit IRC | 00:30 | |
| *** sputnik13 has quit IRC | 00:30 | |
| *** adrian_otto has quit IRC | 00:30 | |
| *** david-lyle has quit IRC | 00:31 | |
| *** rdx_matt has joined #openstack-infra | 00:35 | |
| *** rdx_matt has quit IRC | 00:36 | |
| *** david-lyle_ has quit IRC | 00:36 | |
| *** david-lyle_ has joined #openstack-infra | 00:36 | |
| *** jamesmcarthur has joined #openstack-infra | 00:37 | |
| jroll | mordred | like even "here's a script that will make a config drive with your ssh key and hostname" or whatever <- fwiw, ironic/bifrost has or has talked about having this | 00:37 |
| clarkb | ya bifrost has something to make it I think it may be a playbook | 00:38 |
| clarkb | my only concern with that is double checking that bifrost makes correct config drives now | 00:38 |
| clarkb | because it didn't in the past | 00:38 |
| jroll | :| | 00:38 |
| *** cody-somerville has quit IRC | 00:38 | |
| clarkb | thats where network_data vs network_info in glean comes from | 00:39 |
| *** jamesmcarthur has quit IRC | 00:39 | |
| clarkb | nova is one bifrost the otehr and I can never remember which is canonical so I made a change just to add a comment to glean to clarify | 00:39 |
| *** jcoufal has joined #openstack-infra | 00:42 | |
| SpamapS | clarkb: hah yeah it did them very wrong :-P | 00:42 |
| SpamapS | and yes, bifrost makes them with a playbook | 00:43 |
| SpamapS | https://github.com/openstack/bifrost/tree/master/playbooks/roles/bifrost-configdrives-dynamic | 00:43 |
| mordred | SpamapS: it should be making them correctly now, right? | 00:44 |
| SpamapS | mordred: it is | 00:44 |
| SpamapS | and glean supports both | 00:44 |
| mordred | \o/ | 00:44 |
| SpamapS | falls back to the old broken name just in case | 00:44 |
| ianw | are we at the point that we have ed25519 keys everywhere? adding a new key, should i just go straight to that? | 00:45 |
| clarkb | ianw: we still ahev precise which doesn't iirc | 00:45 |
| clarkb | but I could be wrong about that | 00:45 |
| clarkb | we are almost done with precise though | 00:46 |
| clarkb | Gerrit is also not able iirc but thats easy to swap out and not root related | 00:47 |
| SpamapS | clarkb: even w/ bouncy castle? | 00:49 |
| openstackgerrit | James Slagle proposed openstack-infra/tripleo-ci: Add mulitnode CI job support to tripleo-ci https://review.openstack.org/324777 | 00:51 |
| openstackgerrit | James Slagle proposed openstack-infra/tripleo-ci: Add mulitnode CI job support to tripleo-ci https://review.openstack.org/324777 | 00:52 |
| *** SumitNaiksatam has quit IRC | 00:53 | |
| clarkb | SpamapS: last I checked yup | 00:54 |
| *** karthikp_ has joined #openstack-infra | 00:54 | |
| *** tonytan4ever has joined #openstack-infra | 00:54 | |
| clarkb | I also haven't tried very hard to switch because random things keep not liking it | 00:54 |
| *** edtubill has joined #openstack-infra | 00:54 | |
| *** yamahata has quit IRC | 00:55 | |
| *** esker has joined #openstack-infra | 00:55 | |
| *** Julien-zte has quit IRC | 00:56 | |
| *** karthikp_ has quit IRC | 00:57 | |
| *** karthikp_ has joined #openstack-infra | 00:57 | |
| *** karthikp_ has quit IRC | 00:58 | |
| clarkb | I held a trusty node to test this zaqar thing on because i Have a hunch it is broken there too and some sort of special test ordering gets it to pass | 00:58 |
| clarkb | I used --reason too :) | 00:58 |
| openstackgerrit | Ian Wienand proposed openstack-infra/system-config: Add ianw key and to infra-root https://review.openstack.org/341888 | 00:58 |
| *** karthikp_ has joined #openstack-infra | 00:59 | |
| *** megm has quit IRC | 00:59 | |
| pleia2 | ianw \o/ | 00:59 |
| *** jamesmcarthur has joined #openstack-infra | 00:59 | |
| *** karthikp_ has quit IRC | 00:59 | |
| *** tonytan4ever has quit IRC | 00:59 | |
| *** karthikp_ has joined #openstack-infra | 01:00 | |
| *** anteaya has joined #openstack-infra | 01:01 | |
| notmorgan | jeblair: hm. making it explode on non-strings is proving difficult since we pass values in the ServerJob around. i'm leaning towards accepting bytes or str in py3 for ease of code maintenance. | 01:02 |
| notmorgan | jeblair: rahter than overriding .name/unique specially in ServerJob over Job etc. | 01:03 |
| *** Julien-zte has joined #openstack-infra | 01:03 | |
| *** akshai has quit IRC | 01:03 | |
| *** thorst has joined #openstack-infra | 01:04 | |
| *** jamesmcarthur has quit IRC | 01:04 | |
| *** jamesmcarthur has joined #openstack-infra | 01:04 | |
| *** zz_dimtruck is now known as dimtruck | 01:04 | |
| *** rdx_matt has joined #openstack-infra | 01:04 | |
| *** megm has joined #openstack-infra | 01:04 | |
| *** zhurong has joined #openstack-infra | 01:04 | |
| SpamapS | notmorgan: ugh | 01:06 |
| SpamapS | notmorgan: in general, it should be strings for everything except I/O | 01:06 |
| SpamapS | but .. I've slammed into the problem where there's just too much bytes-centric behavior in the API | 01:06 |
| notmorgan | SpamapS: yeahhhh | 01:06 |
| SpamapS | still, with Gear | 01:07 |
| notmorgan | SpamapS: so i am just trying to make job.name/job.unique auto encode/decode for consumption | 01:07 |
| SpamapS | I'd expect a pretty clear dividing line | 01:07 |
| SpamapS | notmorgan: those should just always be strings | 01:07 |
| notmorgan | SpamapS: so it can easily be referenced. | 01:07 |
| SpamapS | when would you want it to be bytes? | 01:07 |
| *** Apoorva_ has joined #openstack-infra | 01:07 | |
| *** thorst has quit IRC | 01:08 | |
| * SpamapS asked, as he walks away from computer to go do not-computer for a while :-P | 01:08 | |
| notmorgan | SpamapS: except they *cant* be internally since we need to pack them into submitjob | 01:08 |
| clarkb | the only thing I can come up with is if you want to use some esoteric encoding | 01:08 |
| clarkb | or if you are talking to an existing gearman setup that does | 01:09 |
| notmorgan | SpamapS: it's a lot of encode/decode, internally it saves headaches to avoid needing to encode/decode for self.functions etc | 01:09 |
| SpamapS | notmorgan: they should only need to be encoded when they're sent over the wire or hashed | 01:09 |
| notmorgan | we can use the encoded values | 01:09 |
| notmorgan | i mean, i can add a lot more encode/decode | 01:09 |
| notmorgan | i'm just looking at it from a standpoint of "generally folks aren't consuming this part" | 01:09 |
| notmorgan | the interface devs are working with is what i'm trying to taget to avoid needing to do .encode/decode when instantiating the job/referencing the job when you get things back. | 01:10 |
| *** gouthamr has joined #openstack-infra | 01:10 | |
| *** yamamoto has joined #openstack-infra | 01:11 | |
| notmorgan | as i've said before, python3 is at least consistently horrible at bytes vs str | 01:11 |
| *** Apoorva has quit IRC | 01:11 | |
| *** rdx_matt has quit IRC | 01:11 | |
| notmorgan | unlike python 2 which is inconsistent and horrible | 01:11 |
| *** piet has quit IRC | 01:11 | |
| *** Apoorva_ has quit IRC | 01:11 | |
| *** mriedem has joined #openstack-infra | 01:14 | |
| greghaynes | this is in gearman? Why does gearman as a server need to care about strings at all | 01:15 |
| notmorgan | greghaynes: this is the job interface | 01:15 |
| notmorgan | greghaynes: in gear library | 01:15 |
| greghaynes | oh, so its the api for a client? | 01:15 |
| notmorgan | yeah. | 01:15 |
| greghaynes | ah | 01:15 |
| *** shashank_hegde has quit IRC | 01:15 | |
| greghaynes | gotcha | 01:15 |
| greghaynes | yep, thats tricky ; | 01:16 |
| notmorgan | the server interally also uses the structures | 01:16 |
| *** piet has joined #openstack-infra | 01:16 | |
| greghaynes | eek | 01:16 |
| greghaynes | thats scary | 01:16 |
| *** jesusaur has quit IRC | 01:16 | |
| greghaynes | as a server it really really shouldnt care about strings unless it has to IMO | 01:16 |
| notmorgan | it's not just packet pushing | 01:17 |
| *** yamamoto has quit IRC | 01:17 | |
| greghaynes | oh? | 01:17 |
| notmorgan | it handles CanDo and CantDo etc | 01:17 |
| clarkb | it has to introspect | 01:17 |
| notmorgan | yeah | 01:17 |
| greghaynes | sure, but why does that require a decode? | 01:17 |
| notmorgan | it doesn't. | 01:17 |
| greghaynes | is it translitterating or somethig | 01:17 |
| greghaynes | yea | 01:17 |
| greghaynes | so... | 01:17 |
| notmorgan | the point is the internal structures are the same | 01:17 |
| notmorgan | it's an intheritence ick | 01:17 |
| clarkb | the way gear is built the client and worker and server are all basically the same | 01:17 |
| notmorgan | yep | 01:17 |
| greghaynes | yea, that seems like a great way to make more problems | 01:17 |
| clarkb | I think its server and client that are identical and worker is a little specialer | 01:17 |
| notmorgan | well. in py2 and if we expect the cleint to decode the values off the object, it's simple | 01:18 |
| clarkb | so when this was py3'd the first time the rule was basically use bytes always to keep it simple | 01:18 |
| notmorgan | or we accept a litle sloppier data | 01:18 |
| greghaynes | youre basically defining an implicit on wire encoding in to the protocol which doesnt require it | 01:18 |
| clarkb | greghaynes: yes thats why I mentioned you'd want it for esoteric encoding | 01:18 |
| clarkb | like gregfmt | 01:19 |
| notmorgan | LOL | 01:19 |
| notmorgan | this is a python bytes vs str headache with trying to be "nice" to people consuming it | 01:19 |
| notmorgan | if we leave this as a low-level interface | 01:19 |
| notmorgan | and expect the devs to eat the bytes [cause... it's bytes] | 01:19 |
| greghaynes | it sounds like its actually a client library dictating server behavior issue | 01:20 |
| notmorgan | then the headache is that there is a lot of job.name.decode() all over zuul | 01:20 |
| greghaynes | because the server shouldnt care about encoding if it doesnt need to | 01:20 |
| notmorgan | for example | 01:20 |
| notmorgan | well, sockets() require the data to be in bytes. | 01:20 |
| notmorgan | at the base level. | 01:20 |
| notmorgan | we have to conver it at some point. | 01:20 |
| greghaynes | as as client, not as a server | 01:21 |
| notmorgan | we just explicitly try to go utf-8 encode (if you already encoded it. we just let the value sit) | 01:21 |
| notmorgan | right, but the server needs to consume some of these values, which since they're wire format need to be bytes | 01:21 |
| notmorgan | the issue is we just reused data structures (classes) and splitting this up is a lot of code.. | 01:21 |
| greghaynes | yea, requiring an encoding for a server because of a code design seems :( | 01:22 |
| greghaynes | those bugs are really annoying to deal with | 01:22 |
| notmorgan | yar. | 01:22 |
| notmorgan | the *easiest* solution is accept either bytes() or str() for string and/or Name | 01:23 |
| *** jamesmcarthur has quit IRC | 01:23 | |
| notmorgan | erm... unqiue and name | 01:23 |
| notmorgan | though i did find a bug in the code | 01:23 |
| *** jamesmcarthur has joined #openstack-infra | 01:23 | |
| notmorgan | we accept Name as None() which we shouldn't | 01:23 |
| notmorgan | it will cause bad exceptions | 01:23 |
| greghaynes | i'd really just do bytes everywher and if you want a shiney client that can handle string encode/decode wrap it | 01:23 |
| greghaynes | then the server has pretty straightforward semantics WRT encoding | 01:24 |
| *** thorst has joined #openstack-infra | 01:24 | |
| notmorgan | right, i am looking at simply adding @property for job.name and job.unique that make it easier to consume it. | 01:24 |
| notmorgan | so consumers of the lib like zuul don't have to decode/encode all over | 01:25 |
| notmorgan | because it's ... well here | 01:25 |
| *** thorst has quit IRC | 01:25 | |
| greghaynes | gotcha | 01:25 |
| notmorgan | greghaynes: look here | 01:25 |
| notmorgan | https://review.openstack.org/#/c/325661/ | 01:25 |
| *** thorst has joined #openstack-infra | 01:25 | |
| notmorgan | for why this is a "headache" | 01:25 |
| notmorgan | some of it can't be helped | 01:26 |
| greghaynes | yea, I get it, you just really dont want to push that headache in to the server - youll still have the headache and itll be harder to diagnose | 01:26 |
| notmorgan | like job.data, we *cant* know what that is in gear.lib | 01:26 |
| notmorgan | but zuul knows it is json | 01:26 |
| greghaynes | you can because youre the client | 01:26 |
| notmorgan | but job.name and job.unique can be much simpler | 01:26 |
| notmorgan | the other option is wrapping this in zuul | 01:27 |
| greghaynes | seems like there should be a subclassed job which is only used as a client | 01:27 |
| notmorgan | but we then need to strip the wrapper before we send it to submitJob() | 01:27 |
| *** piet has quit IRC | 01:28 | |
| notmorgan | it's ... just kinda ugly. | 01:28 |
| notmorgan | everywhere | 01:28 |
| notmorgan | strings vs bytes makes me wish this wasn't in python | 01:28 |
| greghaynes | yep, you really have to draw a clear line of where the encoding happens and then make sure it only ever happens there | 01:28 |
| *** jesusaur has joined #openstack-infra | 01:29 | |
| notmorgan | maybe i'll just make zuul have a stub class that has .__getattr__ on it that encodes/decodes name/unique and then pulls the underlying job out when sending to submitJob | 01:30 |
| notmorgan | putting this in gear is feeling like *not* worth it | 01:30 |
| *** thorst has quit IRC | 01:30 | |
| greghaynes | Yea, or just a similar thing that is StringJob or w/e | 01:30 |
| greghaynes | if you did put it in gear | 01:30 |
| greghaynes | I definitely wouldnt put it anywhere in the bowels of gear because of what youre hitting | 01:31 |
| *** gouthamr_ has joined #openstack-infra | 01:32 | |
| *** dimtruck is now known as zz_dimtruck | 01:33 | |
| openstackgerrit | Ian Wienand proposed openstack-infra/puppet-user: Allow to specify different key types https://review.openstack.org/341897 | 01:33 |
| notmorgan | I'm going to just do this in zuul. | 01:34 |
| notmorgan | diving into gear here is going to make for more and more headaches | 01:34 |
| *** gouthamr has quit IRC | 01:35 | |
| greghaynes | Yea, thats a lot more in line what what ive run in to - if the server can be just bytes then clients are on their own to implement encoding | 01:35 |
| notmorgan | greghaynes: i think i'd rather write rust or C++ here :P | 01:35 |
| greghaynes | heh, those issues exist there too! | 01:36 |
| notmorgan | greghaynes: /me throws salt on python. | 01:36 |
| *** yuanying has joined #openstack-infra | 01:36 | |
| greghaynes | the typing does help a bit | 01:36 |
| clarkb | though rust says you use uf8 iirc | 01:36 |
| clarkb | rather than generic unicdoe which helps a bit | 01:36 |
| greghaynes | it has to have a bytes type | 01:36 |
| notmorgan | clarkb: which is why i like it | 01:36 |
| clarkb | it does too | 01:36 |
| greghaynes | you cant interface with the world without bytes | 01:36 |
| *** EricGonczer_ has joined #openstack-infra | 01:36 | |
| openstackgerrit | Ian Wienand proposed openstack-infra/system-config: Add ianw key and to infra-root https://review.openstack.org/341888 | 01:37 |
| notmorgan | clarkb: it becomes a more direct rather than asuming this bundle of "well...it could have lots of encoding values" | 01:37 |
| *** garyk1 has quit IRC | 01:37 | |
| clarkb | I wonder how they deal with windows | 01:37 |
| *** edtubill has quit IRC | 01:38 | |
| *** jamesmcarthur has quit IRC | 01:38 | |
| clarkb | if there is a secret utf16 handler in places | 01:38 |
| notmorgan | possibly | 01:38 |
| greghaynes | This is really an issue because its IPC though - even if it was in rust we couldnt assume the other end was also rust | 01:38 |
| greghaynes | yea, because issues like that ;) | 01:38 |
| * notmorgan seriously looks at the X99 chipset board, would be nice to drive a second monitor w/o breaking my VFIO passthrough stuff. | 01:38 | |
| clarkb | yup | 01:38 |
| notmorgan | this is also because Zuul doesn't talk py3 | 01:39 |
| notmorgan | if zuul was all py3 [what i am working on] | 01:40 |
| notmorgan | this probably wouldn't have been so ugly | 01:40 |
| notmorgan | since i'm stripping py2 support on zuulv3 branch, this will be easier. less "this coul;d look like X or Y or Z" | 01:40 |
| *** SumitNaiksatam has joined #openstack-infra | 01:41 | |
| notmorgan | greghaynes: totally unrelated, i have a fixed X1C 3rd gen now. | 01:41 |
| notmorgan | :P | 01:41 |
| greghaynes | haha | 01:41 |
| notmorgan | lenovo fixed it | 01:41 |
| greghaynes | mine is still alive somehow | 01:41 |
| notmorgan | it was so destroyed | 01:41 |
| greghaynes | what happened? | 01:41 |
| * clarkb just thinks notmorgan and nibz must be hard on their laptops | 01:42 | |
| notmorgan | fell off a desk, shattered the screen, tore the screen off [broke the screen casing] | 01:42 |
| clarkb | yup confirmed :P | 01:42 |
| *** jamesmcarthur has joined #openstack-infra | 01:42 | |
| greghaynes | wow | 01:42 |
| notmorgan | knocked it off a desk in the hotel when diving for my phone when i had a migrane and trying to turn off the alarm | 01:42 |
| notmorgan | didn't realize i had broken the laptop until ~6hrs later when i had recovered | 01:42 |
| notmorgan | the other (4th gen x1c) just has a faulty SIM slot, (factory faulty) | 01:43 |
| notmorgan | so i will send it off for repair soon. but otherwise that one is in perfect condition. | 01:43 |
| greghaynes | hrm, I really wanted to experiment with a sim | 01:44 |
| greghaynes | I know nibalizer has one | 01:44 |
| notmorgan | it is REALLY ugly to get working on the 4thgen x1c's | 01:45 |
| clarkb | its chaeper to have a single dvice with one | 01:45 |
| clarkb | so I just use my phone | 01:45 |
| greghaynes | hrm | 01:45 |
| notmorgan | the 7455 sierrawireless is painful | 01:45 |
| clarkb | also it just works usb0 ftw | 01:45 |
| notmorgan | needs a bleeding edge libQMI and libibmi | 01:45 |
| *** jamesmcarthur has quit IRC | 01:45 | |
| notmorgan | but since i have GoogleFi it's easier to add a data-sim | 01:46 |
| notmorgan | no cost, etc. | 01:46 |
| notmorgan | and don't burn phone battery to tether | 01:46 |
| notmorgan | (or consume a ton of laptop battery since USB-C pulls a lot down, esp. when tethering) | 01:46 |
| *** jamesmcarthur has joined #openstack-infra | 01:47 | |
| clarkb | notmorgan: if you plug into not usb c its fine | 01:47 |
| clarkb | assuming you don't have a bad cable | 01:47 |
| notmorgan | clarkb: i find the USB-A->C also pull a lot of power | 01:47 |
| notmorgan | or don't keep the phone charged. | 01:48 |
| clarkb | notmorgan: then you likely have a bad cable | 01:48 |
| notmorgan | nah it's the high-amp port on the laptop | 01:48 |
| greghaynes | didnt the google person reviewing those fry all his gear so he cant test anymore? | 01:48 |
| clarkb | there was a whole thing about this where cables with the wrong resistor in them were drawing far too much power through decives that werent supposed to | 01:48 |
| notmorgan | and i got the usb-a to c from google | 01:48 |
| clarkb | notmorgan: it doesn't matter a proper cable shopuld prevent it on a high amp or .5amp port | 01:48 |
| greghaynes | ah, the google one was good | 01:48 |
| notmorgan | it's the phone burning battery and pulling down power to stay charged. | 01:48 |
| clarkb | greghaynes: I think there was multiple pices of hardware fried yes | 01:49 |
| greghaynes | gotta stop the pokemon go app when plugged in | 01:49 |
| notmorgan | it burns a lot of battery tethering. | 01:49 |
| clarkb | notmorgan: huh I can tether all day on x240 with a full charge in laptop and phone when I start | 01:49 |
| notmorgan | clarkb: which phone? | 01:49 |
| clarkb | 5x | 01:49 |
| notmorgan | clarkb: hmm. yeah my 5x burns a lot of battery tethering. | 01:49 |
| clarkb | my display is still my biggest power draw regardless so optimizing that helps the most | 01:50 |
| notmorgan | on Fi or tmobile, wondering if it's the magic provider switching code in the baseband | 01:50 |
| clarkb | tmobile | 01:50 |
| notmorgan | or $other$ provider? | 01:50 |
| notmorgan | yeah maybe if i locked it to tmo it would be better | 01:50 |
| *** jamesmcarthur has quit IRC | 01:50 | |
| clarkb | oh I also can't do any local "builds" | 01:50 |
| clarkb | do not recomment running tox when on battery | 01:51 |
| notmorgan | hehe, yeah i am exposing a VPN endpoint to my home workstation | 01:51 |
| notmorgan | and i do builds/tox runs on it | 01:51 |
| ianw | clarkb: http://logs.openstack.org/97/341897/1/check/gate-puppet-user-puppet-lint/b9f2bb2/console.html <- is this a xenial / ruby versions / puppet issue ? "can't modify frozen Symbol" | 01:52 |
| clarkb | ianw: I think thats puppet issue | 01:52 |
| Shrews | mordred: notmorgan: the merge of https://review.openstack.org/315713 fixed the shade gate, thanks to rcarrillocruz. as for the performance thing, i have not tried it again recently | 01:53 |
| notmorgan | Shrews: ++ ok | 01:53 |
| clarkb | ianw: I read that as when "compiling" that rb file it ran into an error and its puppet's gem | 01:53 |
| nibalizer | yah wwan in laptop is totally worth it -- game changer | 01:53 |
| clarkb | nibalizer: can you look at ianw's failure | 01:53 |
| clarkb | nibalizer: I am also curious why we use puppet 3.0.0 | 01:53 |
| clarkb | that seems very old which may not work on xenial | 01:54 |
| notmorgan | nibalizer: i can't wait to get it. which WWAN card you have? | 01:54 |
| nibalizer | notmorgan: the one that came? | 01:54 |
| nibalizer | sierra something | 01:54 |
| nibalizer | clarkb: yea lets fix that | 01:54 |
| *** watanabe_isao has quit IRC | 01:54 | |
| *** akshai has joined #openstack-infra | 01:55 | |
| *** jamesmcarthur has joined #openstack-infra | 01:55 | |
| notmorgan | nibalizer: lsusb? wondering what version you have. | 01:55 |
| *** watanabe_isao has joined #openstack-infra | 01:55 | |
| * jroll reads about laptops breaking, meanwhile has thrown his macbook onto concrete multiple times with only cosmetic damage | 01:55 | |
| notmorgan | jroll: with the screen open and hitting things on the screen part? | 01:56 |
| notmorgan | jroll: cause i've seen the x1c's fare well when it's dropped when closed [not mine] | 01:56 |
| jroll | notmorgan: no, it was closed for the concrete throws | 01:56 |
| notmorgan | jroll: yeah. that would have probably made the difference | 01:56 |
| jroll | I've likely dropped it with screen open but don't recall for sure | 01:56 |
| *** xarses has joined #openstack-infra | 01:57 | |
| jroll | notmorgan: I'm just surprised every time I whip my backpack onto my shoulder with a zipper open, laptop flies out, hits asphalt, still works :) | 01:57 |
| ianw | nibalizer: https://github.com/puppetlabs/puppet/commit/d02820a116d5e3ae0b129a0b4384ddaf76f0f83b <- related? fixed in 3.4.3 so it says ... | 01:57 |
| timrc | jroll: I'm glad to know I'm not the only one that happens too :P | 01:58 |
| jroll | hehe | 01:58 |
| jroll | timrc: I solved it with a messenger bag, but then fiancee did it the other day with her backpack and even hit my car >.> | 01:58 |
| openstackgerrit | Spencer Krum proposed openstack-infra/puppet-openstack_infra_spec_helper: Dependon ~> 3.8 for puppet. https://review.openstack.org/341901 | 01:59 |
| nibalizer | ianw: clarkb that should be the fix | 01:59 |
| nibalizer | I don't think you can depends-on though | 02:00 |
| timrc | jroll: Haha. | 02:00 |
| openstackgerrit | Ian Wienand proposed openstack-infra/puppet-user: Allow to specify different key types https://review.openstack.org/341897 | 02:01 |
| timrc | jroll: It's pretty incredible what kind of beating some laptops can take. | 02:01 |
| openstackgerrit | Spencer Krum proposed openstack-infra/project-config: Run basic tests in check and gate https://review.openstack.org/341902 | 02:01 |
| clarkb | huh I wonder if this zaqar fail is realted to mongodb | 02:01 |
| clarkb | that would be funny | 02:01 |
| *** sdake has joined #openstack-infra | 02:02 | |
| nibalizer | and that should make the tests acutally run not just hide in experimental | 02:02 |
| *** tonytan4ever has joined #openstack-infra | 02:02 | |
| *** yuanying has quit IRC | 02:03 | |
| *** yuanying has joined #openstack-infra | 02:03 | |
| *** dizquierdo has quit IRC | 02:04 | |
| ianw | nibalizer: thanks, fast service :) | 02:04 |
| *** yuanying has quit IRC | 02:04 | |
| nibalizer | yah well jesusaur actually figured it out | 02:05 |
| *** rdx_matt has joined #openstack-infra | 02:07 | |
| *** rdx_matt has quit IRC | 02:08 | |
| *** yuanying has joined #openstack-infra | 02:08 | |
| *** tonytan_brb has joined #openstack-infra | 02:10 | |
| *** rkukura has quit IRC | 02:12 | |
| *** yuanying has quit IRC | 02:12 | |
| *** yamamoto has joined #openstack-infra | 02:13 | |
| *** tonytan4ever has quit IRC | 02:13 | |
| *** amotoki has joined #openstack-infra | 02:13 | |
| *** yuanying has joined #openstack-infra | 02:14 | |
| *** sdake has quit IRC | 02:14 | |
| *** yamamoto has quit IRC | 02:18 | |
| *** akscram has quit IRC | 02:22 | |
| openstackgerrit | Merged openstack-infra/irc-meetings: Fix chair for oslo team meetings https://review.openstack.org/341131 | 02:22 |
| *** cody-somerville has joined #openstack-infra | 02:22 | |
| *** yuanying has quit IRC | 02:26 | |
| openstackgerrit | Matt Riedemann proposed openstack-infra/project-config: Revert "Run identity v3 only job in nova" https://review.openstack.org/341908 | 02:27 |
| mriedem | blarg!? | 02:27 |
| mriedem | pabelanger: ^ | 02:27 |
| mriedem | gd, merge conflict | 02:27 |
| *** akscram has joined #openstack-infra | 02:28 | |
| *** akshai has quit IRC | 02:30 | |
| pabelanger | mriedem: ohai | 02:33 |
| pabelanger | ianw: fedora-24 uploading | 02:33 |
| pabelanger | mriedem: shouldn't be an issue to revert | 02:33 |
| mriedem | pabelanger: i'll fix that merge conflict in a sec | 02:33 |
| *** ddieterly has joined #openstack-infra | 02:33 | |
| mriedem | all of a sudden i noticed a bunch of new non-voting jobs in nova's check queue and had to start digging | 02:34 |
| pabelanger | Ya, we're usually pretty good with getting sign off from a PTL or core on job changes, likely just slipped through by mistake | 02:35 |
| *** rdx_matt has joined #openstack-infra | 02:35 | |
| mriedem | np | 02:37 |
| openstackgerrit | Matt Riedemann proposed openstack-infra/project-config: Revert "Run identity v3 only job in nova" https://review.openstack.org/341908 | 02:37 |
| *** itisha has quit IRC | 02:40 | |
| *** thorst has joined #openstack-infra | 02:40 | |
| *** thorst has quit IRC | 02:40 | |
| *** gouthamr_ has quit IRC | 02:41 | |
| *** rdx_matt has quit IRC | 02:43 | |
| *** markvoelker has quit IRC | 02:45 | |
| *** roxanaghe has quit IRC | 02:46 | |
| *** jamesmcarthur has quit IRC | 02:46 | |
| *** roxanaghe has joined #openstack-infra | 02:46 | |
| *** akscram has quit IRC | 02:48 | |
| *** woodster_ has quit IRC | 02:49 | |
| *** akscram has joined #openstack-infra | 02:49 | |
| *** ddieterly has quit IRC | 02:50 | |
| *** roxanaghe has quit IRC | 02:51 | |
| *** akscram has quit IRC | 02:53 | |
| *** akscram has joined #openstack-infra | 02:54 | |
| *** shashank_hegde has joined #openstack-infra | 02:56 | |
| *** regXboi has joined #openstack-infra | 02:56 | |
| * regXboi pads in looking for an infra core or two | 02:57 | |
| openstackgerrit | Merged openstack-infra/project-config: Revert "Run identity v3 only job in nova" https://review.openstack.org/341908 | 02:57 |
| clarkb | regXboi: I am sort of here whats up? | 02:58 |
| *** sdake has joined #openstack-infra | 02:59 | |
| regXboi | clarkb: looking for another set of eyes on https://review.openstack.org/337011 | 02:59 |
| *** jamesmcarthur has joined #openstack-infra | 02:59 | |
| *** yanyanhu has joined #openstack-infra | 03:01 | |
| regXboi | hoping to get the +2/W+1 so that it gets into the gate and the folks that are working on adding the tests can actually start seeing them work/fail | 03:01 |
| clarkb | ok I do note the fix hasnt merged yet so they will likely fail | 03:03 |
| *** itisha has joined #openstack-infra | 03:03 | |
| regXboi | clarkb: which fix are we talking about here? | 03:03 |
| *** asettle has joined #openstack-infra | 03:03 | |
| clarkb | the one linked in that change | 03:04 |
| * regXboi goes to look | 03:04 | |
| clarkb | int the commit message | 03:04 |
| regXboi | ah... ok, the point is to get the -nv job in 337011 running so that that patch can be rechecked against it and verify that it actually is doing what it claims to be doing | 03:05 |
| regXboi | that patch = 335838 | 03:05 |
| *** akscram has quit IRC | 03:07 | |
| clarkb | ok | 03:08 |
| *** asettle has quit IRC | 03:08 | |
| regXboi | thanks clarkb | 03:09 |
| *** zhurong_ has joined #openstack-infra | 03:09 | |
| *** timello has quit IRC | 03:09 | |
| *** coolsvap_ is now known as coolsvap | 03:10 | |
| *** tphummel has quit IRC | 03:11 | |
| openstackgerrit | James Polley proposed openstack-dev/pbr: Fix handling of old git log output https://review.openstack.org/339392 | 03:12 |
| *** yamamoto has joined #openstack-infra | 03:12 | |
| *** shashank_hegde has quit IRC | 03:12 | |
| openstackgerrit | Merged openstack-infra/project-config: Enable Neutron VPNaaS API CI tests https://review.openstack.org/337011 | 03:13 |
| *** yuanying has joined #openstack-infra | 03:13 | |
| *** zhurong has quit IRC | 03:13 | |
| *** akscram has joined #openstack-infra | 03:13 | |
| *** yanyanhu has quit IRC | 03:14 | |
| *** mordred has quit IRC | 03:14 | |
| *** rajinir has quit IRC | 03:15 | |
| *** shashank_hegde has joined #openstack-infra | 03:17 | |
| *** esker has quit IRC | 03:17 | |
| *** yuanying has quit IRC | 03:18 | |
| ianw | pabelanger: yep, cool ... i added stackviz into my local tests too :) | 03:18 |
| *** timello has joined #openstack-infra | 03:18 | |
| *** yuanying has joined #openstack-infra | 03:18 | |
| dmsimard | stackviz all the things | 03:21 |
| *** regXboi has quit IRC | 03:23 | |
| *** yuanying has quit IRC | 03:23 | |
| *** esker has joined #openstack-infra | 03:24 | |
| *** jamesmcarthur has quit IRC | 03:27 | |
| *** treaki_ has joined #openstack-infra | 03:28 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/shade: Change operating to interacting with in README https://review.openstack.org/341927 | 03:29 |
| *** esker has quit IRC | 03:29 | |
| *** akscram has quit IRC | 03:31 | |
| *** Julien-zte has quit IRC | 03:31 | |
| *** treaki has quit IRC | 03:32 | |
| *** akscram has joined #openstack-infra | 03:33 | |
| coolsvap | how can I update status of any channel, i want to update status of #openstack-requirements | 03:36 |
| *** EricGonczer_ has quit IRC | 03:38 | |
| *** akscram has quit IRC | 03:38 | |
| *** rdx_matt has joined #openstack-infra | 03:39 | |
| *** markvoelker has joined #openstack-infra | 03:39 | |
| *** rdx_matt has quit IRC | 03:39 | |
| *** akscram has joined #openstack-infra | 03:40 | |
| *** thorst has joined #openstack-infra | 03:41 | |
| *** mriedem has quit IRC | 03:41 | |
| *** EricGonczer_ has joined #openstack-infra | 03:41 | |
| *** EricGonczer_ has quit IRC | 03:43 | |
| *** jamesmcarthur has joined #openstack-infra | 03:43 | |
| *** thorst has quit IRC | 03:46 | |
| *** akscram has quit IRC | 03:47 | |
| *** yamamoto has quit IRC | 03:47 | |
| *** yamahata has joined #openstack-infra | 03:48 | |
| *** jamesmcarthur has quit IRC | 03:50 | |
| *** yamamoto has joined #openstack-infra | 03:51 | |
| *** akscram has joined #openstack-infra | 03:54 | |
| *** watanabe_isao has quit IRC | 03:54 | |
| *** julim has quit IRC | 03:56 | |
| openstackgerrit | Ian Wienand proposed openstack/diskimage-builder: Revert "Revert "Pre-install pip/virtualenv packages"" https://review.openstack.org/341934 | 03:56 |
| *** julim has joined #openstack-infra | 03:57 | |
| *** julim has quit IRC | 03:57 | |
| *** yuanying has joined #openstack-infra | 03:59 | |
| *** akscram has quit IRC | 04:01 | |
| *** akscram has joined #openstack-infra | 04:04 | |
| *** sflanigan has quit IRC | 04:04 | |
| *** dchen has joined #openstack-infra | 04:04 | |
| *** ajmiller has joined #openstack-infra | 04:06 | |
| *** shashank_hegde has quit IRC | 04:07 | |
| *** rdx_matt has joined #openstack-infra | 04:07 | |
| *** adrian_otto has joined #openstack-infra | 04:09 | |
| *** harlowja has quit IRC | 04:13 | |
| *** Julien-zte has joined #openstack-infra | 04:13 | |
| *** akscram has quit IRC | 04:13 | |
| *** rdx_matt has quit IRC | 04:14 | |
| *** roxanaghe has joined #openstack-infra | 04:15 | |
| *** jamesmcarthur has joined #openstack-infra | 04:16 | |
| *** esker has joined #openstack-infra | 04:16 | |
| *** eric_lopez has joined #openstack-infra | 04:18 | |
| *** ramishra has quit IRC | 04:19 | |
| *** yamamoto has quit IRC | 04:20 | |
| *** jamesmcarthur has quit IRC | 04:21 | |
| *** esker has quit IRC | 04:21 | |
| *** ramishra has joined #openstack-infra | 04:21 | |
| *** elo has quit IRC | 04:21 | |
| *** sflanigan has joined #openstack-infra | 04:21 | |
| *** sflanigan has joined #openstack-infra | 04:21 | |
| *** akscram has joined #openstack-infra | 04:23 | |
| *** david-lyle_ has quit IRC | 04:26 | |
| *** links has joined #openstack-infra | 04:27 | |
| *** akscram has quit IRC | 04:28 | |
| *** akscram has joined #openstack-infra | 04:28 | |
| *** tphummel has joined #openstack-infra | 04:29 | |
| *** reed_ has joined #openstack-infra | 04:31 | |
| *** shashank_hegde has joined #openstack-infra | 04:31 | |
| *** roxanaghe has quit IRC | 04:32 | |
| *** roxanaghe has joined #openstack-infra | 04:32 | |
| *** armax has quit IRC | 04:33 | |
| *** bhavik has joined #openstack-infra | 04:34 | |
| *** coolsvap_ has joined #openstack-infra | 04:34 | |
| *** ajmiller has quit IRC | 04:35 | |
| *** ajmiller has joined #openstack-infra | 04:36 | |
| *** roxanaghe has quit IRC | 04:36 | |
| *** jcoufal has quit IRC | 04:37 | |
| *** coolsvap has quit IRC | 04:37 | |
| *** david-lyle_ has joined #openstack-infra | 04:42 | |
| *** psachin has joined #openstack-infra | 04:42 | |
| *** Nikolay_St has quit IRC | 04:42 | |
| *** tphummel has quit IRC | 04:44 | |
| *** bhavik has quit IRC | 04:45 | |
| *** reed_ has quit IRC | 04:45 | |
| openstackgerrit | Tony Breeds proposed openstack-infra/irc-meetings: Set meeting chairs to the canonical form https://review.openstack.org/251195 | 04:46 |
| *** shashank_hegde has quit IRC | 04:47 | |
| *** akscram has quit IRC | 04:48 | |
| *** PollyZ has joined #openstack-infra | 04:50 | |
| *** shashank_hegde has joined #openstack-infra | 04:50 | |
| *** pcrews has quit IRC | 04:51 | |
| *** pcrews has joined #openstack-infra | 04:53 | |
| *** yuanying has quit IRC | 04:58 | |
| *** akscram has joined #openstack-infra | 04:58 | |
| *** yamamoto has joined #openstack-infra | 05:03 | |
| *** ajmiller has quit IRC | 05:04 | |
| *** asettle has joined #openstack-infra | 05:04 | |
| *** winggundamth has quit IRC | 05:05 | |
| *** shashank_hegde has quit IRC | 05:05 | |
| openstackgerrit | Tony Breeds proposed openstack-infra/irc-meetings: Add IRC nick for nova-notification-meeting https://review.openstack.org/341950 | 05:06 |
| openstackgerrit | Tony Breeds proposed openstack-infra/irc-meetings: Add IRC nick for storyboard-meeting https://review.openstack.org/341951 | 05:06 |
| openstackgerrit | Tony Breeds proposed openstack-infra/irc-meetings: Add IRC nicks for trove-dbaas-team-meeting https://review.openstack.org/341952 | 05:06 |
| *** shashank_hegde has joined #openstack-infra | 05:06 | |
| *** yuanying has joined #openstack-infra | 05:08 | |
| *** nadya has joined #openstack-infra | 05:08 | |
| *** dchen has quit IRC | 05:09 | |
| *** arnewiebalck_ has joined #openstack-infra | 05:09 | |
| *** asettle has quit IRC | 05:09 | |
| openstackgerrit | Andreas Jaeger proposed openstack-infra/project-config: Add console-log to gate-generate-specs-site https://review.openstack.org/341953 | 05:10 |
| AJaeger | fungi, thanks for your findings. Let's fix the console-log directly - and discuss later once we both are around - how to test the post job. Patch for console-log ^ | 05:10 |
| *** arnewiebalck has quit IRC | 05:10 | |
| *** arnewiebalck_ is now known as arnewiebalck | 05:10 | |
| *** strigazi has quit IRC | 05:10 | |
| *** rdx_matt has joined #openstack-infra | 05:10 | |
| *** rdx_matt has quit IRC | 05:10 | |
| *** rdx_matt has joined #openstack-infra | 05:11 | |
| *** yamamoto has quit IRC | 05:12 | |
| *** yamamoto has joined #openstack-infra | 05:12 | |
| *** HenryG has quit IRC | 05:12 | |
| *** armax has joined #openstack-infra | 05:12 | |
| *** coolsvap_ is now known as coolsvap | 05:12 | |
| *** strigazi has joined #openstack-infra | 05:12 | |
| *** winggundamth has joined #openstack-infra | 05:13 | |
| *** piet has joined #openstack-infra | 05:14 | |
| *** HenryG has joined #openstack-infra | 05:15 | |
| *** rdx_matt has quit IRC | 05:15 | |
| *** shashank_hegde has quit IRC | 05:15 | |
| *** PollyZ_ has joined #openstack-infra | 05:16 | |
| *** shashank_hegde has joined #openstack-infra | 05:16 | |
| *** sflanigan has quit IRC | 05:18 | |
| *** PollyZ has quit IRC | 05:18 | |
| openstackgerrit | Andreas Jaeger proposed openstack-infra/project-config: Add console-log to gate-generate-specs-site https://review.openstack.org/341953 | 05:22 |
| *** jcoufal has joined #openstack-infra | 05:23 | |
| openstackgerrit | Merged openstack/diskimage-builder: Make Fedora 24 the default https://review.openstack.org/339681 | 05:26 |
| *** armax has quit IRC | 05:26 | |
| *** sdake has quit IRC | 05:26 | |
| *** pgadiya has joined #openstack-infra | 05:27 | |
| *** shashank_hegde has quit IRC | 05:27 | |
| *** Nikolay_St has joined #openstack-infra | 05:28 | |
| *** rkukura has joined #openstack-infra | 05:29 | |
| *** sflanigan has joined #openstack-infra | 05:30 | |
| *** javeriak has joined #openstack-infra | 05:30 | |
| *** kushal has quit IRC | 05:30 | |
| *** nadya has quit IRC | 05:31 | |
| *** sarob has joined #openstack-infra | 05:32 | |
| *** krtaylor has quit IRC | 05:33 | |
| *** roxanaghe has joined #openstack-infra | 05:36 | |
| *** sarob has quit IRC | 05:37 | |
| *** bhavik has joined #openstack-infra | 05:38 | |
| *** akscram has quit IRC | 05:40 | |
| *** roxanaghe has quit IRC | 05:41 | |
| *** bswartz has quit IRC | 05:41 | |
| *** rdx_matt has joined #openstack-infra | 05:41 | |
| *** bhavik has quit IRC | 05:42 | |
| *** hashar has joined #openstack-infra | 05:44 | |
| *** rdx_matt has quit IRC | 05:46 | |
| *** ccamacho has quit IRC | 05:46 | |
| *** javeriak has quit IRC | 05:50 | |
| *** markvoelker has quit IRC | 05:52 | |
| AJaeger | coolsvap: see http://docs.openstack.org/infra/system-config/irc.html | 05:52 |
| *** javeriak has joined #openstack-infra | 05:53 | |
| *** Goneri has joined #openstack-infra | 05:54 | |
| *** rdx_matt has joined #openstack-infra | 05:55 | |
| *** rdx_matt has quit IRC | 05:55 | |
| coolsvap | AJaeger yes tonyb pointed me to the POC for doing it | 05:55 |
| coolsvap | I think everyone cannot do that | 05:56 |
| *** rdx_matt has joined #openstack-infra | 05:56 | |
| coolsvap | http://git.openstack.org/cgit/openstack-infra/project-config/tree/accessbot/channels.yaml#n37 | 05:56 |
| *** yuanying has quit IRC | 05:56 | |
| openstackgerrit | Ananth Narayan S proposed openstack-infra/project-config: Add new project Plasma to openstack repo https://review.openstack.org/341966 | 05:57 |
| *** akscram has joined #openstack-infra | 05:57 | |
| *** jamesmcarthur has joined #openstack-infra | 05:57 | |
| *** shashank_hegde has joined #openstack-infra | 05:57 | |
| *** yuanying has joined #openstack-infra | 05:58 | |
| AJaeger | coolsvap: you might need to have operator privilidegs. The document shows how to query that. You can ask one of the admins to give one of the requirements team those permissions in that channel as well - but seems nobody is around right now. | 05:58 |
| *** karthikp_ has quit IRC | 05:58 | |
| coolsvap | AJaeger, yes i am waiting for that, i will ping someone once online | 05:59 |
| openstackgerrit | Ananth Narayan S proposed openstack-infra/project-config: Add new project Plasma to openstack repo https://review.openstack.org/341966 | 06:01 |
| *** rcernin has joined #openstack-infra | 06:01 | |
| *** jamesmca_ has joined #openstack-infra | 06:01 | |
| *** jamesmcarthur has quit IRC | 06:01 | |
| *** andreas_s has joined #openstack-infra | 06:02 | |
| *** ramishra_ has joined #openstack-infra | 06:06 | |
| *** esker has joined #openstack-infra | 06:06 | |
| *** ramishra has quit IRC | 06:07 | |
| *** akscram has quit IRC | 06:08 | |
| *** jamesmca_ has quit IRC | 06:10 | |
| *** Ravikiran_K has joined #openstack-infra | 06:10 | |
| *** esker has quit IRC | 06:10 | |
| *** piet has quit IRC | 06:13 | |
| *** jamesmcarthur has joined #openstack-infra | 06:13 | |
| *** javeriak has quit IRC | 06:13 | |
| *** strigazi_ has joined #openstack-infra | 06:14 | |
| *** arnewiebalck_ has joined #openstack-infra | 06:15 | |
| *** strigazi has quit IRC | 06:17 | |
| *** arnewiebalck has quit IRC | 06:17 | |
| *** arnewiebalck_ is now known as arnewiebalck | 06:17 | |
| *** bhavik has joined #openstack-infra | 06:19 | |
| *** yamamoto has quit IRC | 06:22 | |
| *** yamamoto has joined #openstack-infra | 06:23 | |
| *** jcoufal has quit IRC | 06:26 | |
| *** sdake has joined #openstack-infra | 06:27 | |
| *** Goneri has quit IRC | 06:28 | |
| *** mikelk has joined #openstack-infra | 06:29 | |
| *** mrmartin has joined #openstack-infra | 06:33 | |
| *** markvoelker has joined #openstack-infra | 06:34 | |
| *** jamesmcarthur has quit IRC | 06:34 | |
| *** javeriak has joined #openstack-infra | 06:36 | |
| *** mrmartin has quit IRC | 06:36 | |
| *** shashank_hegde has quit IRC | 06:37 | |
| *** PollyZ_ has quit IRC | 06:37 | |
| *** Goneri has joined #openstack-infra | 06:38 | |
| *** markvoelker has quit IRC | 06:39 | |
| *** pcaruana has joined #openstack-infra | 06:40 | |
| openstackgerrit | sandhya proposed openstack/diskimage-builder: Add support for building images capable of UEFI https://review.openstack.org/287784 | 06:42 |
| *** jamesmcarthur has joined #openstack-infra | 06:43 | |
| *** sputnik13 has joined #openstack-infra | 06:43 | |
| *** dingyichen has quit IRC | 06:44 | |
| *** yonglihe has joined #openstack-infra | 06:47 | |
| *** roxanaghe has joined #openstack-infra | 06:48 | |
| *** hashar is now known as hasharAway | 06:48 | |
| *** jamesmcarthur has quit IRC | 06:50 | |
| *** roxanaghe has quit IRC | 06:52 | |
| *** isaacb has joined #openstack-infra | 06:53 | |
| *** pabelanger has quit IRC | 06:57 | |
| *** mrmartin has joined #openstack-infra | 06:57 | |
| *** trown|outtypewww has quit IRC | 06:57 | |
| *** berendt has joined #openstack-infra | 06:58 | |
| *** e0ne has joined #openstack-infra | 06:58 | |
| *** asettle has joined #openstack-infra | 06:58 | |
| *** trown has joined #openstack-infra | 06:59 | |
| *** sorantis has joined #openstack-infra | 06:59 | |
| *** pabelanger has joined #openstack-infra | 06:59 | |
| *** berendt has quit IRC | 07:00 | |
| *** jamesmcarthur has joined #openstack-infra | 07:00 | |
| *** ociuhandu has joined #openstack-infra | 07:01 | |
| *** psachin has quit IRC | 07:02 | |
| *** asettle has quit IRC | 07:02 | |
| *** psachin has joined #openstack-infra | 07:02 | |
| *** zshuo has joined #openstack-infra | 07:04 | |
| *** Jeffrey4l has joined #openstack-infra | 07:04 | |
| *** rcernin has quit IRC | 07:05 | |
| *** jamesmcarthur has quit IRC | 07:06 | |
| *** nmagnezi has joined #openstack-infra | 07:06 | |
| *** jaosorior has joined #openstack-infra | 07:08 | |
| *** pahuang has quit IRC | 07:10 | |
| *** tesseract- has joined #openstack-infra | 07:10 | |
| *** ccamacho has joined #openstack-infra | 07:11 | |
| *** Julien-zte has quit IRC | 07:12 | |
| *** Julien-zte has joined #openstack-infra | 07:13 | |
| *** kushal has joined #openstack-infra | 07:14 | |
| *** jlanoux has joined #openstack-infra | 07:20 | |
| *** rcernin has joined #openstack-infra | 07:20 | |
| *** shardy has joined #openstack-infra | 07:21 | |
| *** e0ne has quit IRC | 07:21 | |
| *** sshnaidm|afk is now known as sshnaidm | 07:24 | |
| *** Ravikiran_K has quit IRC | 07:26 | |
| *** markvoelker has joined #openstack-infra | 07:28 | |
| *** sarob has joined #openstack-infra | 07:33 | |
| *** markvoelker has quit IRC | 07:35 | |
| *** sarob has quit IRC | 07:38 | |
| *** esker has joined #openstack-infra | 07:38 | |
| *** arxcruz has joined #openstack-infra | 07:39 | |
| *** tonytan_brb has quit IRC | 07:40 | |
| *** Jeffrey4l has quit IRC | 07:41 | |
| *** rossella_s has joined #openstack-infra | 07:41 | |
| *** esikachev has joined #openstack-infra | 07:41 | |
| *** yuanying has quit IRC | 07:42 | |
| *** esker has quit IRC | 07:43 | |
| *** yuanying has joined #openstack-infra | 07:43 | |
| *** akscram has joined #openstack-infra | 07:44 | |
| *** matthewbodkin has joined #openstack-infra | 07:44 | |
| *** pt_15 has quit IRC | 07:46 | |
| *** jlanoux has quit IRC | 07:47 | |
| *** jlanoux has joined #openstack-infra | 07:47 | |
| *** zeih has joined #openstack-infra | 07:49 | |
| *** itisha has quit IRC | 07:50 | |
| *** sdake_ has joined #openstack-infra | 07:50 | |
| *** berendt has joined #openstack-infra | 07:50 | |
| *** nadya has joined #openstack-infra | 07:51 | |
| *** nadya has quit IRC | 07:51 | |
| *** sdake has quit IRC | 07:51 | |
| *** sdake has joined #openstack-infra | 07:52 | |
| *** sflanigan has quit IRC | 07:53 | |
| *** bhavik has quit IRC | 07:54 | |
| *** sdake_ has quit IRC | 07:55 | |
| *** bhavik has joined #openstack-infra | 07:55 | |
| *** zeih has quit IRC | 07:55 | |
| *** berendt has quit IRC | 07:57 | |
| *** d0ugal has joined #openstack-infra | 07:59 | |
| *** zzzeek has quit IRC | 08:00 | |
| *** zzzeek has joined #openstack-infra | 08:00 | |
| *** rcernin has quit IRC | 08:00 | |
| *** zeih has joined #openstack-infra | 08:00 | |
| *** [1]Thelo has joined #openstack-infra | 08:02 | |
| *** jpich has joined #openstack-infra | 08:02 | |
| openstackgerrit | Sagi Shnaidman proposed openstack-infra/tripleo-ci: WIP: DONT MERGE TESTING https://review.openstack.org/316436 | 08:04 |
| *** Thelo has quit IRC | 08:04 | |
| *** [1]Thelo is now known as Thelo | 08:04 | |
| *** asettle has joined #openstack-infra | 08:06 | |
| *** jamesmcarthur has joined #openstack-infra | 08:06 | |
| *** hamzy has quit IRC | 08:11 | |
| *** jianghuaw has joined #openstack-infra | 08:11 | |
| *** jistr is now known as jistr|off | 08:11 | |
| *** jamesmcarthur has quit IRC | 08:11 | |
| *** derekh has joined #openstack-infra | 08:12 | |
| *** rcernin has joined #openstack-infra | 08:13 | |
| *** strigazi_ is now known as strigazi | 08:13 | |
| coolsvap | SergeyLukjanov, hi | 08:14 |
| openstackgerrit | Ricardo Carrillo Cruz proposed openstack-infra/shade: Change operating to interacting with in README https://review.openstack.org/341927 | 08:15 |
| coolsvap | SergeyLukjanov, I need your help to update topic in #openstack-requirements channel, can you help me with that? | 08:15 |
| openstackgerrit | Merged openstack-infra/puppet-openstack_infra_spec_helper: Dependon ~> 3.8 for puppet. https://review.openstack.org/341901 | 08:16 |
| *** eezhova has joined #openstack-infra | 08:17 | |
| *** ifarkas has joined #openstack-infra | 08:19 | |
| *** amotoki has quit IRC | 08:20 | |
| *** markvoelker has joined #openstack-infra | 08:23 | |
| *** markvoelker has quit IRC | 08:27 | |
| *** akscram has quit IRC | 08:27 | |
| *** ociuhandu has quit IRC | 08:28 | |
| *** shardy has quit IRC | 08:29 | |
| *** lucas|afk is now known as lucasagomes | 08:29 | |
| *** cody-somerville has quit IRC | 08:29 | |
| *** hamzy has joined #openstack-infra | 08:30 | |
| *** yuanying has quit IRC | 08:31 | |
| *** amotoki has joined #openstack-infra | 08:34 | |
| *** javeriak has quit IRC | 08:36 | |
| *** yuanying has joined #openstack-infra | 08:37 | |
| *** Srinu has joined #openstack-infra | 08:38 | |
| Srinu | Hi | 08:40 |
| Srinu | Nodepool image is not booting a nodepool vm in Openstack CI | 08:41 |
| *** adrian_otto has quit IRC | 08:41 | |
| Srinu | how to eliminate the Nodepool Bad Images | 08:41 |
| *** asettle has quit IRC | 08:41 | |
| *** ams__ has joined #openstack-infra | 08:42 | |
| *** amotoki has quit IRC | 08:42 | |
| Srinu | asselin__: hi | 08:44 |
| rcarrillocruz | Srinu: http://docs.openstack.org/infra/system-config/nodepool.html#bad-images | 08:44 |
| rcarrillocruz | Srinu: it's too early for asselin__ | 08:45 |
| rcarrillocruz | he's on US west coast | 08:45 |
| *** akscram has joined #openstack-infra | 08:45 | |
| *** pt_15 has joined #openstack-infra | 08:45 | |
| Srinu | rcarrillocruz: If the image is bad so i have to delete every time. i have to monitor ci daily | 08:45 |
| *** asettle has joined #openstack-infra | 08:47 | |
| Srinu | DuncanT: Nodepool image is not booting a nodepool vm in Openstack CI. how to eliminate the Nodepool Bad Images | 08:48 |
| DuncanT | Srinu: I know just about nothing about nodepool, sorry. | 08:48 |
| kun_huang | what is status of JJB and playbook? would we abandon JJB some days later? | 08:49 |
| Srinu | Did anyone face this issue nodepool bad images | 08:50 |
| Srinu | rcarrillocruz:how can i fix that issue. please can you suggest me | 08:51 |
| *** e0ne has joined #openstack-infra | 08:51 | |
| *** javeriak has joined #openstack-infra | 08:51 | |
| *** yuanying has quit IRC | 08:53 | |
| *** nadya has joined #openstack-infra | 08:55 | |
| *** e0ne has quit IRC | 08:59 | |
| *** jtomasek has joined #openstack-infra | 09:00 | |
| openstackgerrit | Markus Zoeller (markus_z) proposed openstack-infra/release-tools: release_notes.py: search your feelings https://review.openstack.org/342053 | 09:00 |
| *** amotoki has joined #openstack-infra | 09:01 | |
| *** e0ne has joined #openstack-infra | 09:01 | |
| *** sputnik13 has quit IRC | 09:03 | |
| *** dizquierdo has joined #openstack-infra | 09:06 | |
| openstackgerrit | Markus Zoeller (markus_z) proposed openstack-infra/release-tools: release_notes.py: search your feelings https://review.openstack.org/342053 | 09:07 |
| *** shardy has joined #openstack-infra | 09:08 | |
| Srinu | Hi. | 09:10 |
| *** javeriak has quit IRC | 09:10 | |
| Srinu | How to eliminate the Nodepool bad images. any work around for that | 09:11 |
| Srinu | hi.How to eliminate the Nodepool bad images( by using this image nodepool vm is not booting properly). any work around for that | 09:11 |
| *** sambetts|afk is now known as sambetts | 09:13 | |
| *** GheRivero has joined #openstack-infra | 09:13 | |
| *** zshuo has quit IRC | 09:15 | |
| *** markvoelker has joined #openstack-infra | 09:17 | |
| *** tosky has joined #openstack-infra | 09:18 | |
| *** dizquierdo has quit IRC | 09:20 | |
| *** markvoelker has quit IRC | 09:21 | |
| Srinu | Hi | 09:25 |
| *** clenimar_ has quit IRC | 09:30 | |
| *** Julien-zte has quit IRC | 09:32 | |
| openstackgerrit | Marton Kiss proposed openstack-infra/groups: Upgrade to Drupal Core 7.50 https://review.openstack.org/342067 | 09:33 |
| *** amotoki has quit IRC | 09:33 | |
| *** asettle has quit IRC | 09:34 | |
| *** sarob has joined #openstack-infra | 09:34 | |
| *** sambetts has quit IRC | 09:36 | |
| coolsvap | Sam-I-Am, hi | 09:36 |
| Srinu | coolsvap: hi | 09:37 |
| *** akscram has quit IRC | 09:38 | |
| *** sarob has quit IRC | 09:38 | |
| coolsvap | Srinu, hi whatsup | 09:39 |
| Srinu | coolsvap: How to eliminate the Nodepool bad images( by using this image nodepool vm is not booting properly). | 09:39 |
| openstackgerrit | Jin Jing Lin proposed openstack-infra/project-config: Fix error in Neutron VPNaaS API CI job. https://review.openstack.org/342068 | 09:40 |
| *** pfallenop has quit IRC | 09:40 | |
| *** _degorenko|afk is now known as degorenko | 09:40 | |
| *** mrmartin has quit IRC | 09:42 | |
| coolsvap | Srinu, I do not know that details about nodepool | 09:42 |
| Srinu | coolsvap: thank you | 09:42 |
| *** akscram has joined #openstack-infra | 09:43 | |
| *** asettle has joined #openstack-infra | 09:43 | |
| Srinu | asettle: hi | 09:45 |
| openstackgerrit | Flavio Percoco proposed openstack-infra/meetbot: Add support for round-table discussions https://review.openstack.org/342069 | 09:46 |
| *** amotoki has joined #openstack-infra | 09:46 | |
| asettle | Srinu: HI | 09:47 |
| Srinu | asettle:How to eliminate the Nodepool bad images( by using this image nodepool vm is not booting properly) | 09:49 |
| *** jaosorior has quit IRC | 09:50 | |
| *** jaosorior has joined #openstack-infra | 09:50 | |
| *** asettle has quit IRC | 09:50 | |
| GheRivero | Srinu: nodepool image-delete IMAGE_ID | 09:57 |
| *** pfallenop has joined #openstack-infra | 09:58 | |
| Srinu | GheRivero: for that i have to monitor the CI everyday | 09:58 |
| Srinu | GheRivero: right | 09:58 |
| Srinu | GheRivero: thats it is not the correct right | 09:58 |
| Srinu | GheRivero: Why it is not building image correctly? | 09:59 |
| GheRivero | that's another question.... the logs should have the answer | 09:59 |
| *** akscram has quit IRC | 10:00 | |
| GheRivero | Srinu: nodepool.openstack.org <- there are the logs of the images built | 10:04 |
| Srinu | GheRivero: ok. what can i get from their | 10:04 |
| *** ociuhandu has joined #openstack-infra | 10:05 | |
| *** zhurong_ has quit IRC | 10:05 | |
| GheRivero | just look for the image that is not booting properly and check that the built process and the params are correct | 10:06 |
| *** pfallenop has quit IRC | 10:06 | |
| *** jamesmcarthur has joined #openstack-infra | 10:08 | |
| *** jamesmcarthur has quit IRC | 10:10 | |
| *** markvoelker has joined #openstack-infra | 10:11 | |
| *** yamamoto has quit IRC | 10:11 | |
| *** akscram has joined #openstack-infra | 10:12 | |
| *** dtantsur|afk is now known as dtantsur | 10:12 | |
| *** markvoelker has quit IRC | 10:16 | |
| *** sdake has quit IRC | 10:16 | |
| openstackgerrit | Dougal Matthews proposed openstack-infra/project-config: Add documentation publishing for tripleo-common https://review.openstack.org/342082 | 10:17 |
| *** Na3iL has joined #openstack-infra | 10:19 | |
| *** roxanaghe has joined #openstack-infra | 10:24 | |
| *** akscram has quit IRC | 10:24 | |
| *** mordred has joined #openstack-infra | 10:27 | |
| *** akscram has joined #openstack-infra | 10:27 | |
| *** roxanaghe has quit IRC | 10:29 | |
| openstackgerrit | Adam Coldrick proposed openstack-infra/storyboard: Limit oslo.db to <4.8.0 https://review.openstack.org/342087 | 10:29 |
| *** jamesmcarthur has joined #openstack-infra | 10:30 | |
| *** eezhova has quit IRC | 10:30 | |
| *** coolsvap_ has joined #openstack-infra | 10:35 | |
| *** jamesmcarthur has quit IRC | 10:35 | |
| *** coolsvap has quit IRC | 10:38 | |
| *** clenimar_ has joined #openstack-infra | 10:41 | |
| *** coolsvap_ is now known as coolsvap | 10:41 | |
| *** zeih has quit IRC | 10:42 | |
| *** bhavik has quit IRC | 10:42 | |
| *** zeih has joined #openstack-infra | 10:45 | |
| *** asettle has joined #openstack-infra | 10:50 | |
| *** dizquierdo has joined #openstack-infra | 10:50 | |
| *** claudiub has joined #openstack-infra | 10:54 | |
| *** deadnull has joined #openstack-infra | 10:54 | |
| *** pt_15 has quit IRC | 10:56 | |
| openstackgerrit | Dougal Matthews proposed openstack-infra/project-config: Add documentation publishing for tripleo-common https://review.openstack.org/342082 | 10:56 |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Temporarily remove puppet-ceph from ci slaves https://review.openstack.org/340717 | 10:59 |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Temporarily disable introspection https://review.openstack.org/342106 | 10:59 |
| d0ugal | AJaeger: Thanks for the quick review. I updated ^ | 10:59 |
| *** yamamoto has joined #openstack-infra | 11:01 | |
| *** flepied has joined #openstack-infra | 11:02 | |
| *** Julien-zte has joined #openstack-infra | 11:03 | |
| openstackgerrit | Merged openstack-infra/tripleo-ci: Temporarily remove puppet-ceph from ci slaves https://review.openstack.org/340717 | 11:03 |
| openstackgerrit | Merged openstack-infra/shade: Change operating to interacting with in README https://review.openstack.org/341927 | 11:04 |
| openstackgerrit | Merged openstack-infra/tripleo-ci: Temporarily disable introspection https://review.openstack.org/342106 | 11:04 |
| *** markvoelker has joined #openstack-infra | 11:05 | |
| Srinu | asselin__: hi | 11:06 |
| *** markvoelker has quit IRC | 11:09 | |
| openstackgerrit | fumihiko kakuma proposed openstack-infra/project-config: Use ovs-interface-undefault instead of ovs-native job https://review.openstack.org/338944 | 11:10 |
| *** javeriak has joined #openstack-infra | 11:13 | |
| coolsvap | Srinu, I think you should send a mail to openstack-infra mailing list with your query | 11:16 |
| coolsvap | that would be a better alternative for better response | 11:16 |
| *** kzaitsev_mb has joined #openstack-infra | 11:18 | |
| openstackgerrit | Joshua Hesketh proposed openstack-infra/zuul: Merge branch 'master' into v3_merge https://review.openstack.org/341559 | 11:21 |
| *** lucasagomes is now known as lucas-hungry | 11:23 | |
| Srinu | Hi | 11:23 |
| Srinu | hi | 11:26 |
| *** Na3iL has quit IRC | 11:27 | |
| *** Na3iL has joined #openstack-infra | 11:27 | |
| coolsvap | jhesketh, hi I need your help to update topic in #openstack-requirements channel, can you help me with that? | 11:28 |
| Srinu | when the nodepool service is running i am trying to build an nodepool image and i saw the nodepool vm is going offline with this status as slave went offline | 11:28 |
| Srinu | coolsvap:when the nodepool service is running i am trying to build an nodepool image and i saw the nodepool vm is going offline with this status as slave went offline | 11:29 |
| Srinu | coolsvap: untill the completion of nodepool image | 11:29 |
| *** john-davidge has joined #openstack-infra | 11:30 | |
| *** john-dav_ has joined #openstack-infra | 11:31 | |
| Srinu | untill the completion of nodepool image | 11:31 |
| *** yamahata has quit IRC | 11:32 | |
| *** akscram has quit IRC | 11:32 | |
| *** mhickey has joined #openstack-infra | 11:33 | |
| *** john-dav_ is now known as john-davidge_ | 11:33 | |
| jhesketh | coolsvap: sure, what do you want it set to? | 11:33 |
| *** amotoki has quit IRC | 11:33 | |
| coolsvap | jhesketh, I want to add the information about wiki and meeting let me give it to you | 11:33 |
| *** bhavik has joined #openstack-infra | 11:34 | |
| Srinu | jhesketh: hi | 11:34 |
| *** john-davidge has quit IRC | 11:34 | |
| jhesketh | Srinu: Howdy | 11:35 |
| *** sarob has joined #openstack-infra | 11:35 | |
| Srinu | jhesketh: i have one doubt | 11:35 |
| Srinu | jhesketh:when the nodepool service is running i am trying to build an nodepool image and i saw the nodepool vm is going offline with this status as slave went offline | 11:35 |
| jhesketh | Srinu: which status? | 11:37 |
| *** javeriak has quit IRC | 11:37 | |
| Srinu | jhesketh: Slave went offline is coming for me when the nodepool vm running job | 11:38 |
| openstackgerrit | Vasyl Saienko proposed openstack-infra/project-config: Move ironic-multitenant-network-nv to check pipeline https://review.openstack.org/342122 | 11:39 |
| *** sarob has quit IRC | 11:39 | |
| *** javeriak has joined #openstack-infra | 11:40 | |
| *** thorst_ has joined #openstack-infra | 11:40 | |
| *** isaacb has quit IRC | 11:41 | |
| *** jcoufal has joined #openstack-infra | 11:44 | |
| jhesketh | Srinu: sorry I don't quire follow... are you building snapshot style images and while they are being made the template node is disappearing? | 11:45 |
| openstackgerrit | amrith proposed openstack-infra/irc-meetings: Proposing OpenStack SWG Meeting https://review.openstack.org/338134 | 11:46 |
| openstackgerrit | amrith proposed openstack-infra/irc-meetings: Proposing OpenStack SWG Meeting https://review.openstack.org/338134 | 11:46 |
| *** akscram has joined #openstack-infra | 11:46 | |
| *** esker has joined #openstack-infra | 11:47 | |
| Srinu | jhesketh: I am building Nodepool image and with using previous nodepool image jenkins running the Job and it is going offline | 11:47 |
| openstackgerrit | Joshua Hesketh proposed openstack-infra/zuul: Merge branch 'master' into v3_merge https://review.openstack.org/341559 | 11:47 |
| Srinu | jhesketh: Slave went offline and it is repeating untill the Nodepool image build success | 11:48 |
| *** amotoki has joined #openstack-infra | 11:49 | |
| Srinu | jhesketh: their no info in logs also | 11:49 |
| *** zhurong has joined #openstack-infra | 11:50 | |
| *** markusry has joined #openstack-infra | 11:50 | |
| jhesketh | Srinu: are you having trouble building the new image? ie, is that related or is it just the slave nodes going offline prematurely? | 11:51 |
| jhesketh | Srinu: are you building the new image with DIB? | 11:51 |
| Srinu | jhesketh: yes | 11:52 |
| Srinu | jhesketh: while the build in progress slave nodes going offline prematurely | 11:52 |
| Srinu | jhesketh: | 11:52 |
| jhesketh | Srinu: so only while building a new nodepool image this occurs? | 11:52 |
| Srinu | jhesketh: yes and creating a new nodepool vm and running the job and some time after then again the nodepool vm going slave went offline | 11:53 |
| jhesketh | Srinu: are you using DIB or snapshots? and what version of nodepool? | 11:54 |
| Srinu | jhesketh: i am using nodepool image-build dpc | 11:54 |
| Srinu | jhesketh: i am using dib only | 11:54 |
| *** esker has quit IRC | 11:57 | |
| *** eezhova has joined #openstack-infra | 11:57 | |
| jhesketh | Srinu: are you running from master? | 11:58 |
| Srinu | jhesketh: i am using jenkins master | 11:58 |
| jhesketh | Srinu: while the nodes are running a job does their state switch to "used"? | 11:58 |
| Srinu | jhesketh: yes | 11:58 |
| Srinu | jhesketh: and after some time it is going done by showing this message slave went offline | 11:59 |
| *** markvoelker has joined #openstack-infra | 11:59 | |
| *** rodrigods has quit IRC | 11:59 | |
| *** rodrigods has joined #openstack-infra | 11:59 | |
| jhesketh | Srinu: do you have the logs from the jenkins build? is jenksin shutting it down somehow perhaps? | 11:59 |
| *** ldnunes has joined #openstack-infra | 12:00 | |
| Srinu | jhesketh: in that No info | 12:00 |
| Srinu | jhesketh: in nodepool.log | 12:00 |
| jhesketh | Srinu: I was thinking the log of the jenkins job itself | 12:00 |
| jhesketh | Srinu: also see if you have a nodepool-debug.log file (or similar) | 12:00 |
| *** amotoki has quit IRC | 12:01 | |
| openstackgerrit | Jim Rollenhagen proposed openstack-infra/project-config: Make python35 jobs voting on ironic projects https://review.openstack.org/341830 | 12:01 |
| Srinu | jhesketh: if the nodepool image build success, the nodepool vm is unable boot properly and not attaching to jenkins and jobs are not running | 12:02 |
| *** rossella_s has quit IRC | 12:03 | |
| *** markvoelker has quit IRC | 12:03 | |
| *** YorikSar has quit IRC | 12:03 | |
| *** Ravikiran_K has joined #openstack-infra | 12:04 | |
| *** amotoki has joined #openstack-infra | 12:04 | |
| jhesketh | Srinu: after the image is built have you uploaded it to your provider? | 12:04 |
| openstackgerrit | Adam Coldrick proposed openstack-infra/storyboard: Limit oslo.db to < 4.8.0 https://review.openstack.org/342087 | 12:05 |
| Srinu | jhesketh: yes and nodepool launched vm | 12:05 |
| *** sarob has joined #openstack-infra | 12:05 | |
| jhesketh | Srinu: but it fails to launch and terminates it? | 12:05 |
| *** YorikSar has joined #openstack-infra | 12:05 | |
| Srinu | jhesketh: it is showing Active and running in provider but it is not booting correctly | 12:06 |
| jhesketh | Srinu: do you have another provider you can test on? Perhaps just with a node or two? | 12:06 |
| jhesketh | Srinu: or otherwise directly try to boot the image on the provider yourself (rather than through nodepool) and see if you can get into it | 12:07 |
| Srinu | jhesketh: i tried with on same provider with manually creating a vm using the nodepool image but the same issue not booting correctly | 12:09 |
| jhesketh | so either the provider is having issues or the image isn't complete | 12:09 |
| *** sarob has quit IRC | 12:10 | |
| Srinu | jhesketh: in provider i tested with other images like cirros and ubuntu it is working fine | 12:10 |
| Srinu | jhesketh: but the nodepool image is not working correctly..how can i overcome this issue | 12:10 |
| jhesketh | Srinu: you could try downloading the image and booting on a local hypervisor | 12:11 |
| jhesketh | Srinu: or even running diskimage-builder locally too | 12:11 |
| *** roxanaghe has joined #openstack-infra | 12:12 | |
| Srinu | jhesketh: were i can run diskimage-builder either in jenkin master or provider | 12:13 |
| jhesketh | Srinu: on a fresh vm or a local computer | 12:13 |
| jhesketh | Srinu: preferably somewhere where you can run it on a hypervisor too (ie locally) | 12:13 |
| Srinu | jhesketh: can you tell me the command i have to be run | 12:14 |
| *** pradk has joined #openstack-infra | 12:14 | |
| Srinu | jhesketh: Nodepool by default use this diskimage-builder to create an nodepool image | 12:14 |
| jhesketh | Srinu: well it might be better to start with trying to run the image you've already built locally | 12:14 |
| *** flepied has quit IRC | 12:15 | |
| Srinu | jhesketh: Nodepool will update the image every 15 hours and upload to the provider and the nodepool vm is not booting correctly and Jnekins is not running jobs | 12:15 |
| jhesketh | Srinu: right, so the built image is buggy (well that's my guess)... which means you need to debug that somehow | 12:16 |
| jhesketh | the easiest way would be to run it locally | 12:16 |
| *** roxanaghe has quit IRC | 12:16 | |
| jhesketh | or in a cloud where you can attach to the console | 12:16 |
| Srinu | jhesketh: can you suggest me how to debug that issue | 12:17 |
| *** matthewbodkin has quit IRC | 12:17 | |
| *** amotoki has quit IRC | 12:17 | |
| *** kgiusti has joined #openstack-infra | 12:18 | |
| Srinu | jhesketh: is it mandatory to use every time the new nodepool image r it is not required | 12:18 |
| Srinu | jhesketh: i can use the old nodepool image everytime | 12:18 |
| *** Jeffrey4l has joined #openstack-infra | 12:18 | |
| Srinu | jeblair: hi | 12:19 |
| *** zeih has quit IRC | 12:19 | |
| *** zeih has joined #openstack-infra | 12:20 | |
| jhesketh | Srinu: sorry that's a bit difficult to describe any further.. you might need to do some reading on dib (http://docs.openstack.org/developer/diskimage-builder/) and play/test around configuratoins (it depends on what elements etc you have configured too) | 12:21 |
| jhesketh | Srinu: nope, not mandatory.. so yes, you can still use the old image | 12:21 |
| Srinu | jhesketh: if it is not required to use i can easly use old nodepool image | 12:22 |
| jhesketh | Srinu: that might be a good temporary solution... the only reason to build new images is to get updates to your configuration or system packages | 12:23 |
| Srinu | jhesketh: but it is not booting right | 12:23 |
| jhesketh | Srinu: the old images aren't booting right either? | 12:23 |
| *** lucas-hungry is now known as lucasagomes | 12:24 | |
| Srinu | jhesketh: Old image is booting correctly but the new image builded automatically by nodepool after 15 hours . this is not booting correctly | 12:25 |
| jhesketh | Srinu: right, so you can delay rebuilding images, but you should probably debug why it isn't working at some point | 12:25 |
| Srinu | jhesketh: i increased the rebuilding the image but i have to solve that issue right | 12:26 |
| *** cbader has joined #openstack-infra | 12:26 | |
| Srinu | jhesketh: i am not getting how to solve that issue | 12:26 |
| *** yamamoto has quit IRC | 12:26 | |
| *** rossella_s has joined #openstack-infra | 12:27 | |
| openstackgerrit | Adam Coldrick proposed openstack-infra/storyboard-webclient: Complex priorities UI in stories https://review.openstack.org/312666 | 12:28 |
| *** ddieterly has joined #openstack-infra | 12:28 | |
| *** clenimar_ has quit IRC | 12:28 | |
| *** EricGonczer_ has joined #openstack-infra | 12:28 | |
| *** dprince has joined #openstack-infra | 12:28 | |
| *** krtaylor has joined #openstack-infra | 12:29 | |
| *** zeih has quit IRC | 12:31 | |
| *** bswartz has joined #openstack-infra | 12:31 | |
| jhesketh | Srinu: sorry it's difficult to be more specific.. as suggested try reading up on DIB and boot your built image somewhere where you can attach/watch the console | 12:31 |
| *** ddieterly has quit IRC | 12:31 | |
| Srinu | jhesketh: ok | 12:31 |
| *** mdrabe has joined #openstack-infra | 12:31 | |
| openstackgerrit | Dmitry Tantsur proposed openstack-infra/tripleo-ci: Properly configure introspection HTTP port and reenable it in gate https://review.openstack.org/342146 | 12:31 |
| *** baoli has joined #openstack-infra | 12:32 | |
| *** arxcruz has quit IRC | 12:33 | |
| *** baoli_ has joined #openstack-infra | 12:33 | |
| *** Jeffrey4l has quit IRC | 12:34 | |
| *** raildo has joined #openstack-infra | 12:36 | |
| *** baoli has quit IRC | 12:36 | |
| openstackgerrit | Joshua Hesketh proposed openstack-infra/zuul: Merge branch 'master' into v3_merge https://review.openstack.org/341559 | 12:37 |
| *** daemontool_ has quit IRC | 12:38 | |
| Srinu | Hi | 12:39 |
| *** YorikSar_ has joined #openstack-infra | 12:39 | |
| *** YorikSar has quit IRC | 12:39 | |
| Srinu | did any one face this issue.How to eliminate the Nodepool bad images( by using this image nodepool vm is not booting properly). | 12:39 |
| *** links has quit IRC | 12:40 | |
| *** flepied has joined #openstack-infra | 12:41 | |
| openstackgerrit | Merged openstack-infra/storyboard: Limit oslo.db to < 4.8.0 https://review.openstack.org/342087 | 12:42 |
| Srinu | clarkb: hi | 12:47 |
| *** yamamoto has joined #openstack-infra | 12:49 | |
| *** tpsilva has joined #openstack-infra | 12:49 | |
| *** Jeffrey4l has joined #openstack-infra | 12:52 | |
| *** weshay has joined #openstack-infra | 12:53 | |
| *** d0ugal has quit IRC | 12:53 | |
| *** rlandy has joined #openstack-infra | 12:53 | |
| *** xyang1 has joined #openstack-infra | 12:54 | |
| Srinu | asselin__: hi | 12:56 |
| *** bhavik has quit IRC | 12:57 | |
| *** bhavik has joined #openstack-infra | 12:57 | |
| *** matthewbodkin has joined #openstack-infra | 12:57 | |
| *** javeriak has quit IRC | 12:57 | |
| *** tonytan4ever has joined #openstack-infra | 12:58 | |
| Srinu | fungi: hi | 12:59 |
| Srinu | pabelanger: hi | 12:59 |
| *** psilvad has joined #openstack-infra | 13:00 | |
| *** gordc has joined #openstack-infra | 13:00 | |
| *** ddieterly has joined #openstack-infra | 13:00 | |
| *** cardeois_ has joined #openstack-infra | 13:01 | |
| *** cardeois_ has quit IRC | 13:01 | |
| *** maishsk has joined #openstack-infra | 13:01 | |
| *** cardeois_ has joined #openstack-infra | 13:02 | |
| *** pgadiya has quit IRC | 13:03 | |
| *** Srinu has quit IRC | 13:03 | |
| *** edmondsw has joined #openstack-infra | 13:06 | |
| *** julim has joined #openstack-infra | 13:06 | |
| *** sorantis has quit IRC | 13:06 | |
| *** sdake has joined #openstack-infra | 13:08 | |
| *** yamamoto has quit IRC | 13:09 | |
| *** psilvad has quit IRC | 13:12 | |
| *** EricGonczer_ has quit IRC | 13:12 | |
| *** yamamoto has joined #openstack-infra | 13:12 | |
| *** psilvad has joined #openstack-infra | 13:13 | |
| *** cbader has quit IRC | 13:13 | |
| *** eharney has joined #openstack-infra | 13:14 | |
| *** sdake__ has joined #openstack-infra | 13:15 | |
| openstackgerrit | Adam Coldrick proposed openstack-infra/storyboard-webclient: Allow comments to be edited by their owner https://review.openstack.org/333418 | 13:16 |
| *** igorn has quit IRC | 13:16 | |
| *** zz_dimtruck is now known as dimtruck | 13:19 | |
| openstackgerrit | Adam Coldrick proposed openstack-infra/storyboard-webclient: Fix duplicate description preview in card detail modal https://review.openstack.org/342172 | 13:19 |
| *** woodster_ has joined #openstack-infra | 13:21 | |
| *** mrmartin has joined #openstack-infra | 13:24 | |
| *** eezhova has quit IRC | 13:24 | |
| *** yaume has joined #openstack-infra | 13:26 | |
| *** amotoki has joined #openstack-infra | 13:26 | |
| *** _ari_ has joined #openstack-infra | 13:26 | |
| *** markusry has quit IRC | 13:27 | |
| *** eezhova has joined #openstack-infra | 13:28 | |
| *** Srinu has joined #openstack-infra | 13:28 | |
| Srinu | hi | 13:28 |
| Srinu | Hi.i am building image using this command nodepool image-build dpc,after success and the image is uploaded to provider and launched a nodepool vm and it is not booting correclty.How to eliminate the Nodepool bad images( by using this image nodepool vm is not booting properly) | 13:29 |
| *** mrmartin has quit IRC | 13:31 | |
| *** mriedem has joined #openstack-infra | 13:31 | |
| *** amitgandhinz has quit IRC | 13:31 | |
| AJaeger | Srinu, I think you got more than enough answers from people here already, I doubt anybody else can help you more than that. People in this channel read backscroll, so if anybody has any further suggestions, they will speak up. But there's no need to repeat your question every hour here. | 13:32 |
| *** amitgandhinz has joined #openstack-infra | 13:32 | |
| *** esberglu has joined #openstack-infra | 13:32 | |
| Srinu | AJaeger: ok | 13:33 |
| Srinu | AJaeger: Thanks for your info | 13:33 |
| *** Jeffrey4l has quit IRC | 13:34 | |
| *** adrian_otto has joined #openstack-infra | 13:35 | |
| *** coolsvap_ has joined #openstack-infra | 13:36 | |
| *** rhallisey has joined #openstack-infra | 13:36 | |
| coolsvap_ | Srinu, as I suggested earlier it is better to send detailed mail to openstack-infra mailing list with details as what you are doing and where you are stuck possibly with logs | 13:37 |
| *** thiagop has joined #openstack-infra | 13:37 | |
| *** akscram has quit IRC | 13:38 | |
| *** coolsvap has quit IRC | 13:39 | |
| pabelanger | zigo: looks like your package was published properly in reprepro | 13:40 |
| *** akscram has joined #openstack-infra | 13:40 | |
| *** pt_15 has joined #openstack-infra | 13:41 | |
| *** itisha has joined #openstack-infra | 13:42 | |
| *** markusry has joined #openstack-infra | 13:43 | |
| Srinu | asselin__: Hi | 13:46 |
| *** Jeffrey4l has joined #openstack-infra | 13:47 | |
| AJaeger | wow, pabelanger ! | 13:47 |
| *** daemontool has joined #openstack-infra | 13:48 | |
| *** coolsvap_ has quit IRC | 13:49 | |
| *** tidwellr has joined #openstack-infra | 13:53 | |
| *** r-mibu has quit IRC | 13:54 | |
| *** r-mibu has joined #openstack-infra | 13:54 | |
| *** ddieterly is now known as ddieterly[away] | 13:55 | |
| *** rossella_s has quit IRC | 13:55 | |
| *** amitgandhinz has quit IRC | 13:56 | |
| *** amitgandhinz has joined #openstack-infra | 13:56 | |
| *** hasharAway is now known as hasharAudio | 13:57 | |
| *** savihou has joined #openstack-infra | 13:57 | |
| *** PollyZ has joined #openstack-infra | 13:58 | |
| *** roxanaghe has joined #openstack-infra | 13:58 | |
| Srinu | pabelanger: hi | 13:58 |
| *** jamesmcarthur has joined #openstack-infra | 13:59 | |
| *** links has joined #openstack-infra | 14:01 | |
| *** ddieterly[away] is now known as ddieterly | 14:01 | |
| openstackgerrit | Adam Coldrick proposed openstack-infra/storyboard-webclient: Make the notifications block fixed on large and medium screens https://review.openstack.org/342164 | 14:02 |
| *** roxanaghe has quit IRC | 14:03 | |
| *** jamesmcarthur has quit IRC | 14:04 | |
| *** amotoki has quit IRC | 14:04 | |
| *** amitgandhinz has quit IRC | 14:04 | |
| *** amitgandhinz has joined #openstack-infra | 14:05 | |
| *** dimtruck is now known as zz_dimtruck | 14:06 | |
| *** arxcruz has joined #openstack-infra | 14:06 | |
| *** amotoki has joined #openstack-infra | 14:07 | |
| *** akshai has joined #openstack-infra | 14:09 | |
| openstackgerrit | Jp Maxwell proposed openstack-infra/puppet-mediawiki: Modified settings file to remove questy captcha and add recaptcha. Re-enabled new account creation allowing new users to edit the wiki again. Setup two new variables to hold public and private recaptcha keys wg_recaptchapublickey and wg_recaptchaprivate https://review.openstack.org/341800 | 14:09 |
| *** rbrndt has joined #openstack-infra | 14:10 | |
| *** markusry has quit IRC | 14:10 | |
| *** edmondsw has quit IRC | 14:10 | |
| *** Jeffrey4l has quit IRC | 14:10 | |
| *** coolsvap_ has joined #openstack-infra | 14:10 | |
| openstackgerrit | Jp Maxwell proposed openstack-infra/puppet-mediawiki: Modified settings file to remove questy captcha and add recaptcha. Re-enabled new account creation allowing new users to edit the wiki again. Setup two new variables to hold public and private recaptcha keys wg_recaptchapublickey and wg_recaptchaprivateke https://review.openstack.org/341800 | 14:11 |
| *** markusry has joined #openstack-infra | 14:16 | |
| *** bhavik has quit IRC | 14:16 | |
| *** coolsvap_ has quit IRC | 14:17 | |
| *** garyk has joined #openstack-infra | 14:17 | |
| *** gouthamr has joined #openstack-infra | 14:18 | |
| *** markusry has quit IRC | 14:19 | |
| rcarrillocruz | jpmaxman: you may want to add a new line between the commit title and commit msg description ^ | 14:20 |
| *** nmagnezi has quit IRC | 14:21 | |
| *** markusry has joined #openstack-infra | 14:21 | |
| *** markusry has quit IRC | 14:21 | |
| *** amotoki has quit IRC | 14:21 | |
| jpmaxman | ah ok - still getting the form right ;) | 14:22 |
| openstackgerrit | Jp Maxwell proposed openstack-infra/puppet-mediawiki: Setup ReCaptcha in Wiki to Prevent Spam https://review.openstack.org/341800 | 14:23 |
| jpmaxman | better? | 14:23 |
| *** Jeffrey4l has joined #openstack-infra | 14:23 | |
| rcarrillocruz | yup | 14:23 |
| *** greghaynes has quit IRC | 14:23 | |
| *** adrian_otto has quit IRC | 14:25 | |
| *** greghaynes has joined #openstack-infra | 14:25 | |
| *** adrian_otto has joined #openstack-infra | 14:25 | |
| *** ajmiller has joined #openstack-infra | 14:28 | |
| *** zz_dimtruck is now known as dimtruck | 14:29 | |
| *** mrmartin has joined #openstack-infra | 14:29 | |
| *** eezhova has quit IRC | 14:30 | |
| *** burgerk has joined #openstack-infra | 14:31 | |
| *** jaosorior has quit IRC | 14:32 | |
| *** adrian_otto has quit IRC | 14:35 | |
| *** hasharAudio is now known as hashar | 14:35 | |
| *** markvoelker has joined #openstack-infra | 14:35 | |
| *** mikelk has quit IRC | 14:36 | |
| *** zul has quit IRC | 14:36 | |
| *** john-davidge_ has quit IRC | 14:38 | |
| *** john-davidge has joined #openstack-infra | 14:38 | |
| *** pcaruana has quit IRC | 14:39 | |
| *** zul has joined #openstack-infra | 14:40 | |
| *** Swami has joined #openstack-infra | 14:41 | |
| AJaeger | fungi, do you have a few minutes now to talk about constraints/post jobs? | 14:41 |
| *** markus__ has joined #openstack-infra | 14:42 | |
| *** savihou has quit IRC | 14:43 | |
| *** adrian_otto has joined #openstack-infra | 14:43 | |
| *** savihou has joined #openstack-infra | 14:44 | |
| *** david-lyle_ has quit IRC | 14:45 | |
| *** Nikolay_St has quit IRC | 14:45 | |
| fungi | AJaeger: sure | 14:45 |
| *** Srinivas has joined #openstack-infra | 14:45 | |
| Srinivas | Hi | 14:46 |
| Srinivas | Jeblair:hi | 14:46 |
| *** d0ugal has joined #openstack-infra | 14:46 | |
| AJaeger | fungi, do you have a good idea on how to test constraints in post jobs? | 14:48 |
| *** amotoki has joined #openstack-infra | 14:48 | |
| Srinivas | Fungi: hi | 14:48 |
| AJaeger | I could setup a specific job for openstack-manuals (which has branches) that does not publish anything - just checks out a bit. And then do some test merges... | 14:48 |
| AJaeger | Or do you have a better idea? | 14:49 |
| openstackgerrit | Merged openstack-infra/project-config: Add console-log to gate-generate-specs-site https://review.openstack.org/341953 | 14:49 |
| fungi | AJaeger: yeah, something with multiple branches that gets frequent commits merged on not-master, with a simple job that doesn't do anything other than just exercise zuul-cloner/constraints file retrieval and confirm that we get one from the correct branch | 14:50 |
| fungi | AJaeger: your idea is probably the simplest solution | 14:50 |
| anteaya | Srinivas: just saying hi to people is irritating and won't help you achieve your goal. I suggest you clearly state your question, if someone can help you they will. I also suggest you read http://www.catb.org/esr/faqs/smart-questions.html | 14:50 |
| Srinivas | Fungi: i am building nodepool image and after build i uploaded to provider and launched a nodepool vm and itis not booting correctly | 14:51 |
| AJaeger | fungi, ok, then I'll look at this tomorrow and propose a job - since I'm core in openstack-manuals, let me use that as test | 14:51 |
| * AJaeger will be offline the rest of the day | 14:51 | |
| fungi | Srinivas: sounds like you need to do some additional troubleshooting. "not booting correctly" is pretty vague. i suggest you figure out what's happening and that may lend some clues as to why it's happening | 14:52 |
| pabelanger | AJaeger: 339636, we should be using bindep for that patch | 14:52 |
| *** tidwellr has left #openstack-infra | 14:53 | |
| pabelanger | left a comment on the review too | 14:53 |
| anteaya | AJaeger: have a good day | 14:54 |
| AJaeger | thanks, anteaya | 14:54 |
| anteaya | :) | 14:54 |
| *** psilvad has quit IRC | 14:54 | |
| *** edtubill has joined #openstack-infra | 14:54 | |
| Srinivas | Fungi: i will remove the image update in nodepool yml | 14:54 |
| Srinivas | Fungi: cab i use single nodepool image entirely | 14:55 |
| AJaeger | pabelanger: was considering that as well but left it for now... I let you followup | 14:55 |
| openstackgerrit | Merged openstack-infra/release-tools: update the script to expire old bug reports https://review.openstack.org/337642 | 14:55 |
| openstackgerrit | Merged openstack-infra/project-config: Add openstack/monasca-grafana-datasource project https://review.openstack.org/341176 | 14:55 |
| fungi | Srinivas: i don't know what "remove the image update" means in the context of the nodepool.yaml config, nor what you mean by "use single nodepool image entirely" | 14:56 |
| *** mrmartin has quit IRC | 14:56 | |
| *** rossella_s has joined #openstack-infra | 14:57 | |
| *** psilvad has joined #openstack-infra | 14:57 | |
| *** Julien-zte has quit IRC | 14:58 | |
| *** maishsk has quit IRC | 14:58 | |
| Srinivas | Fungi: in nodepool yml config there is a cron feild in that check feild is we are mentioning to update the image after certain houra | 14:58 |
| *** mrmartin has joined #openstack-infra | 14:59 | |
| fungi | Srinivas: right, so you mean you will configure nodepool to stop updating its configured images? | 14:59 |
| Srinivas | Fungi: yes, it will effect the ci | 15:00 |
| *** andreas_s has quit IRC | 15:00 | |
| fungi | i don't know off the top of my head whether there is a default for that which kicks in if you comment it out/remove it completely, but you should be able to set it for a sufficiently low frequency (monthly? yearly?) to give you an opportunity to troubleshoot it while you just reuse your existing images | 15:00 |
| *** karthikp_ has joined #openstack-infra | 15:01 | |
| fungi | obviously i wouldn't recommend leaving it that way | 15:01 |
| *** jamesmcarthur has joined #openstack-infra | 15:02 | |
| *** shardy_ has joined #openstack-infra | 15:02 | |
| Srinivas | Fungi: ok but will it effect the ci performance? | 15:02 |
| Srinivas | Fungi: ok | 15:03 |
| fungi | Srinivas: over time it will, yes, as your images get more stale (your jobs have to fetch more refs from teh upstream git mirror, pull in newer package updates from the distro, et cetera) | 15:03 |
| fungi | i don't think you should consider it a solution, but it may be a useful temporary measure depending on what you're trying to troubleshoot | 15:04 |
| *** annegentle has joined #openstack-infra | 15:04 | |
| *** akscram has quit IRC | 15:04 | |
| openstackgerrit | Merged openstack-infra/project-config: Enable publishing for Charm documentation https://review.openstack.org/341378 | 15:05 |
| *** edmondsw has joined #openstack-infra | 15:05 | |
| *** zhurong has quit IRC | 15:05 | |
| openstackgerrit | Merged openstack-infra/project-config: Make python35 jobs voting on ironic projects https://review.openstack.org/341830 | 15:05 |
| clarkb | Srinivas: fungi: one thing that can help debugging nodepool images is to boot them directly | 15:06 |
| openstackgerrit | Merged openstack-infra/project-config: Add documentation publishing for tripleo-common https://review.openstack.org/342082 | 15:06 |
| *** akscram has joined #openstack-infra | 15:06 | |
| openstackgerrit | Merged openstack-infra/project-config: Change mistral-extra jobs https://review.openstack.org/340346 | 15:06 |
| clarkb | assuming it is getting uploaded to glance you can nova boot that and inspect it by hand | 15:06 |
| openstackgerrit | Merged openstack-infra/project-config: Disable all gate jobs for fuel-mirror master https://review.openstack.org/340814 | 15:06 |
| openstackgerrit | Merged openstack-infra/project-config: Add experimental Fedora 24 job https://review.openstack.org/340672 | 15:06 |
| fungi | yep, that's what i'm hoping Srinivas meant by "use single nodepool image entirely" | 15:06 |
| fungi | but i was unable to parse that into anything that made a lot of sense | 15:07 |
| jeblair | Shrews: comment on https://review.openstack.org/341223 | 15:07 |
| *** dmsimard is now known as dmsimard|afk | 15:08 | |
| *** coolsvap has joined #openstack-infra | 15:08 | |
| *** yamamoto has quit IRC | 15:08 | |
| *** eharney has quit IRC | 15:09 | |
| Srinivas | Clarkb: i have created a vm by using nova but the same not booting it is getting stucked | 15:10 |
| openstackgerrit | Oleksii Zamiatin proposed openstack-infra/project-config: Turn zmq gates to voting https://review.openstack.org/340439 | 15:10 |
| *** shardy has quit IRC | 15:10 | |
| *** deadnull has quit IRC | 15:10 | |
| clarkb | Srinivas: you should be able to inspect why though. Get the consoloe logs from nova is probably a good first step | 15:11 |
| *** Srinivas has quit IRC | 15:12 | |
| openstackgerrit | Merged openstack-infra/project-config: Add check-requirements job to all openstack-ansible repo's https://review.openstack.org/341508 | 15:12 |
| openstackgerrit | Merged openstack-infra/project-config: Add valgrind test to liberasurecode gerrit gate https://review.openstack.org/341665 | 15:12 |
| openstackgerrit | Merged openstack-infra/zuul: Merge branch 'master' into v3_merge https://review.openstack.org/341559 | 15:13 |
| *** karthikp_ has quit IRC | 15:13 | |
| *** karthikp_ has joined #openstack-infra | 15:14 | |
| openstackgerrit | Merged openstack-infra/project-config: Bindep: Use python3-all https://review.openstack.org/341093 | 15:15 |
| *** eharney has joined #openstack-infra | 15:16 | |
| *** d0ugal has quit IRC | 15:16 | |
| openstackgerrit | Merged openstack-infra/project-config: Add designate charms https://review.openstack.org/341372 | 15:16 |
| openstackgerrit | Merged openstack-infra/project-config: [zmq] Use proper driver name to configure devstack https://review.openstack.org/339050 | 15:16 |
| *** links has quit IRC | 15:17 | |
| openstackgerrit | Merged openstack-infra/project-config: Add charm-aodh https://review.openstack.org/339485 | 15:17 |
| openstackgerrit | Merged openstack-infra/project-config: Enable py35 voting on oslo.utils https://review.openstack.org/340662 | 15:17 |
| *** sdake has quit IRC | 15:17 | |
| *** d0ugal has joined #openstack-infra | 15:17 | |
| *** armax has joined #openstack-infra | 15:17 | |
| *** esberglu has quit IRC | 15:17 | |
| *** esberglu has joined #openstack-infra | 15:18 | |
| openstackgerrit | Merged openstack-infra/project-config: Enable py35 voting on oslo.policy https://review.openstack.org/341270 | 15:18 |
| *** Kaiyan has joined #openstack-infra | 15:19 | |
| *** mtanino has joined #openstack-infra | 15:22 | |
| Shrews | jeblair: aye. thx | 15:25 |
| *** ramishra_ has quit IRC | 15:25 | |
| *** esberglu has quit IRC | 15:26 | |
| openstackgerrit | Merged openstack-infra/project-config: publish osc-lib docs https://review.openstack.org/340981 | 15:26 |
| *** ramishra has joined #openstack-infra | 15:26 | |
| coolsvap | how can one I add patchbot to a channel? | 15:26 |
| *** esberglu has joined #openstack-infra | 15:27 | |
| *** psilvad has quit IRC | 15:27 | |
| fungi | coolsvap: try again with different phrasing. i couldn't make an answerable question out of that | 15:29 |
| *** karthikp_ has quit IRC | 15:29 | |
| rcarrillocruz | i guess you mean getting 'Foo proposed blah patch' | 15:30 |
| coolsvap | fungi, i need patchbot in #openstack-requirement channel | 15:30 |
| pabelanger | what is patchbot? | 15:30 |
| coolsvap | pabelanger, its what rcarrillocruz said | 15:31 |
| coolsvap | which updates about the patch submitted/merged | 15:31 |
| pabelanger | ah, gerritbot | 15:31 |
| mugsie | it replies to a channel when a review is mentioned with the title | 15:31 |
| coolsvap | pabelanger, sorry | 15:32 |
| coolsvap | not that | 15:32 |
| mugsie | coolsvap: oh, do you mean patchbot, or gerritbot? | 15:32 |
| *** puiterwijk is now known as puiterwijk|mnt | 15:32 | |
| coolsvap | whenever we have a patch url in message | 15:32 |
| coolsvap | it provides the patch url | 15:32 |
| mugsie | [16:31] [freenode] --- | [patchbot] (john@not.mn) : Supybot 0.84.0-dev | 15:32 |
| fungi | coolsvap: that's not something we run | 15:32 |
| pabelanger | ya, that isn't a bot we maintain | 15:32 |
| coolsvap | oh | 15:32 |
| *** esikachev has quit IRC | 15:33 | |
| pabelanger | also, not sure we should be adding random bots into our channels, do we? | 15:33 |
| mugsie | looks like notmyname runs it | 15:33 |
| *** clenimar_ has joined #openstack-infra | 15:33 | |
| fungi | we can't control people joining bots to channels, though if the residents of those channels find them a nuisance we can ban them | 15:33 |
| *** ihrachys has joined #openstack-infra | 15:33 | |
| *** isaacb has joined #openstack-infra | 15:33 | |
| fungi | on a case-by-case basis | 15:34 |
| *** admcleod has joined #openstack-infra | 15:34 | |
| pabelanger | true | 15:34 |
| coolsvap | sorry guys i didnt know its channel specific | 15:35 |
| *** jaosorior has joined #openstack-infra | 15:36 | |
| *** jaosorior has quit IRC | 15:36 | |
| *** ihrachys has quit IRC | 15:36 | |
| *** ajmiller has quit IRC | 15:37 | |
| *** ajmiller has joined #openstack-infra | 15:37 | |
| *** jaosorior has joined #openstack-infra | 15:38 | |
| *** EricGonczer_ has joined #openstack-infra | 15:38 | |
| *** mrmartin has quit IRC | 15:39 | |
| *** yamamoto has joined #openstack-infra | 15:39 | |
| *** mriedem has quit IRC | 15:39 | |
| *** mriedem1 has joined #openstack-infra | 15:39 | |
| Kaiyan | fungi: Can you take a look at https://review.openstack.org/#/c/341091/ when you get a chance please? Thanks! This review is to remove check requirements CI job from monasca-agent. Because there are libraries such as tornado and supervisor that we can not get into global requirements. | 15:40 |
| *** javeriak has joined #openstack-infra | 15:40 | |
| *** nmagnezi has joined #openstack-infra | 15:40 | |
| *** bethwhite has quit IRC | 15:41 | |
| anteaya | pabelanger: I had a mailing list thread up about that topic some time ago | 15:41 |
| anteaya | interest in it kind of fizzled | 15:42 |
| *** nadya has quit IRC | 15:42 | |
| anteaya | and given the larger conversation of go, I decided to give it a pass for a bit | 15:42 |
| anteaya | however if you feel it is time to revive the conversation, the thread is still out there | 15:42 |
| *** krtaylor has joined #openstack-infra | 15:43 | |
| *** mrmartin has joined #openstack-infra | 15:43 | |
| *** yamahata has joined #openstack-infra | 15:43 | |
| openstackgerrit | Philip Marc Schwartz proposed openstack-infra/infra-specs: Add extension to openstackci for next phase of work https://review.openstack.org/239810 | 15:44 |
| phschwartz | asselin__: nibalizer: ^ | 15:44 |
| *** yamamoto has quit IRC | 15:44 | |
| *** mriedem1 is now known as mriedem | 15:45 | |
| asselin__ | oh wow an update! looking | 15:45 |
| openstackgerrit | David Shrewsbury proposed openstack-infra/nodepool: Use random numbers for ZK test paths https://review.openstack.org/341223 | 15:45 |
| phschwartz | asselin__: nibalizer: I will be focusing time the next week to knocking this out, so if we can push to get it agreed upon and merged it would be great | 15:45 |
| phschwartz | asselin__: Yeah, I have the time and the need to focus on it | 15:45 |
| *** dizquierdo has quit IRC | 15:45 | |
| *** mrmartin has quit IRC | 15:45 | |
| *** javeriak has quit IRC | 15:45 | |
| openstackgerrit | David Shrewsbury proposed openstack-infra/nodepool: Use random strings for ZK test paths https://review.openstack.org/341223 | 15:45 |
| phschwartz | asselin__: I removed all of the talk of the server template, user template, and iptables. | 15:46 |
| *** roxanaghe has joined #openstack-infra | 15:47 | |
| zubchick | folks, please review my patch I want to create a new project https://review.openstack.org/#/c/341443/ | 15:48 |
| *** Jeffrey4l has quit IRC | 15:48 | |
| anteaya | asselin__: lennyb mmedvede if one of you is available to chair the Monday 15:00 third party meeting next week, I'd be grateful, I will be traveling to the nova mid-cycle | 15:49 |
| lennyb | anteaya, sorry, cant take it | 15:49 |
| asselin__ | anteaya, I have a conflict & double-booked during that time, so cannot chair. | 15:50 |
| *** sshnaidm is now known as sshnaidm|afk | 15:50 | |
| anteaya | lennyb: asselin__ thanks for letting me know, thank you | 15:50 |
| lennyb | anteaya, i can start, but I will have only about 20mins | 15:50 |
| anteaya | lennyb: understood | 15:50 |
| openstackgerrit | Peter Stachowski proposed openstack-infra/project-config: [mistral] Add check-requirements to mistral-extra https://review.openstack.org/342255 | 15:50 |
| anteaya | perhaps mmedvede has time | 15:50 |
| *** roxanaghe has quit IRC | 15:52 | |
| lennyb | anteaya, if you, asselin__, mmedvede and me wont be there, most likely noone else would:) | 15:52 |
| *** akscram has quit IRC | 15:52 | |
| phschwartz | Are there any guides for setting up jeepyb? I haven't noticed any, but haven't search well yet. | 15:53 |
| anteaya | lennyb: good point, I'll wait to hear from mmedvede | 15:53 |
| clarkb | phschwartz: I don't think so beyond whatever docs are in jeepyb itself | 15:53 |
| *** bethwhite- has joined #openstack-infra | 15:54 | |
| *** akscram has joined #openstack-infra | 15:54 | |
| mmedvede | anteaya: I think I can, but I need to check calendar. I'll get back to you in a bit. busy right now | 15:54 |
| clarkb | phschwartz: I want to saw the projects.yaml config is documented as a docstring | 15:54 |
| clarkb | s/saw/say/ | 15:54 |
| rcarrillocruz | phschwartz: there's also puppet-jeepyb in case is not in your radar | 15:54 |
| *** rossella_s has quit IRC | 15:54 | |
| phschwartz | clarkb: yay, will be a learning experience and maybe I will blog the work so it is stored for furture documenting | 15:54 |
| phschwartz | rcarrillocruz: I did not know there was. That is good to know | 15:54 |
| phschwartz | rcarrillocruz: in openstack-ci? | 15:55 |
| fungi | phschwartz: it's worth keeping in mind that jeepyb is really just a grab-bag of random gerrit-related (mostly) scripts and tooling | 15:55 |
| phschwartz | rcarrillocruz: your re-review of https://review.openstack.org/#/c/239810/ would be appreciated | 15:55 |
| rcarrillocruz | openstack-infra/puppet-jeepyb | 15:55 |
| phschwartz | fungi: understandable. A grab bag I want to use ;) | 15:55 |
| rcarrillocruz | you could look at system-config for how to declare it on nodes, etc | 15:55 |
| anteaya | mmedvede: yup, thanks | 15:55 |
| rcarrillocruz | saw the change, i'll review later | 15:55 |
| fungi | phschwartz: so depending on _what_ from it you want to set up, documentation is in different places and of varying quality/modernity | 15:55 |
| *** coolsvap_ has joined #openstack-infra | 15:55 | |
| fungi | phschwartz: it was just a bunch of arbitrary semi-related stuff we ripped out of our old config repo really | 15:56 |
| phschwartz | fungi: ah, so it will probably just require a bit of searching. And I want project creation/management and acl management from it mostly. | 15:56 |
| *** coolsvap has quit IRC | 15:57 | |
| *** karthikp_ has joined #openstack-infra | 15:57 | |
| *** ddieterly is now known as ddieterly[away] | 15:58 | |
| fungi | phschwartz: okay, so that's really just http://git.openstack.org/cgit/openstack-infra/jeepyb/tree/jeepyb/cmd/manage_projects.py which might be easier understood in the context of http://git.openstack.org/cgit/openstack-infra/puppet-jeepyb/tree/manifests/manage_projects.pp | 15:58 |
| *** ramishra has quit IRC | 15:58 | |
| *** hurgleburgler has joined #openstack-infra | 15:58 | |
| *** ramishra has joined #openstack-infra | 15:59 | |
| fungi | phschwartz: and from a broader perspective, also http://git.openstack.org/cgit/openstack-infra/puppet-gerrit/tree/manifests/init.pp | 15:59 |
| clarkb | I think i may have just run itno a unittest2 bug! | 16:00 |
| *** ddieterly[away] is now known as ddieterly | 16:01 | |
| clarkb | lifeless: your eyes on http://paste.openstack.org/show/532830/ would be appreciated if you have a moment for an odd unittest2 traceback | 16:01 |
| *** isaacb has quit IRC | 16:01 | |
| *** berendt has joined #openstack-infra | 16:02 | |
| openstackgerrit | Anita Kuno proposed openstack-infra/storyboard: Add example commands for the User Tokens api https://review.openstack.org/339297 | 16:02 |
| *** matt-borland has joined #openstack-infra | 16:02 | |
| openstackgerrit | Anita Kuno proposed openstack-infra/storyboard: Add example commands for the Users api https://review.openstack.org/338570 | 16:02 |
| openstackgerrit | Anita Kuno proposed openstack-infra/storyboard: Add example commands for the Worklists api https://review.openstack.org/340506 | 16:02 |
| phschwartz | fungi: ty for the pointers I will look at those closely | 16:03 |
| *** asettle has quit IRC | 16:03 | |
| *** asettle has joined #openstack-infra | 16:04 | |
| clarkb | hrm and that seemed to have caused the entire test suite to bail out | 16:04 |
| fungi | phschwartz: and apologies for the state of documentation for jeepyb. i think there are psychological barriers at play because it's a dumping ground for various ugly hacks we wish would eventually go away, so it's hard to get excited about documenting it | 16:04 |
| phschwartz | fungi: all good :) | 16:05 |
| *** flepied has quit IRC | 16:05 | |
| *** hashar is now known as hasharAway | 16:06 | |
| rcarrillocruz | anyone got issues on tests where if running func tests to create servers 'too soon' Nova would crash with a 500 ? | 16:07 |
| *** shashank_hegde has joined #openstack-infra | 16:07 | |
| *** xarses has quit IRC | 16:07 | |
| rcarrillocruz | i've had to put an artificial delay last week on a change, but today it seems any increase i put is not enough... | 16:07 |
| fungi | nibalizer: crinkle: any idea if there are scope issues at play in https://review.openstack.org/341897 ? (see the depending change i linked in a review comment for the resulting failing jobs) | 16:07 |
| *** jlanoux has quit IRC | 16:09 | |
| *** asettle has quit IRC | 16:09 | |
| mmedvede | anteaya: I can chair the next Monday's 15:00 UTC third party meeting | 16:10 |
| anteaya | mmedvede: awesome, thank you | 16:12 |
| *** e0ne has quit IRC | 16:13 | |
| openstackgerrit | Adam Coldrick proposed openstack-infra/storyboard: Allow timeline events to be related to worklists and boards https://review.openstack.org/342263 | 16:13 |
| notmyname | rcarrillocruz: you want patchbot? | 16:14 |
| *** coolsvap_ has quit IRC | 16:14 | |
| *** Swami has quit IRC | 16:14 | |
| *** bhavik has joined #openstack-infra | 16:16 | |
| zaro | morning | 16:16 |
| openstackgerrit | Derek Higgins proposed openstack-infra/tripleo-ci: Use delorean-current for puppet modules https://review.openstack.org/341429 | 16:18 |
| rcarrillocruz | nah, coolsvap was asking something about it | 16:18 |
| *** shashank_hegde has quit IRC | 16:18 | |
| nibalizer | fungi: at a conference, will check when i get a break | 16:19 |
| fungi | nibalizer: it's not urgent. enjoy your conferencing | 16:19 |
| *** mrmartin has joined #openstack-infra | 16:19 | |
| *** ajmiller has quit IRC | 16:19 | |
| rcarrillocruz | Shrews: heya, ever got something like http://logs.openstack.org/99/342099/7/check/gate-ansible-role-cloud-launcher-dsvm-ansible-func-ubuntu-trusty/04fcad3/console.html when running os_server tests | 16:19 |
| rcarrillocruz | ? | 16:19 |
| rcarrillocruz | i'm crashing nova it seems :/ | 16:19 |
| *** ajmiller has joined #openstack-infra | 16:20 | |
| *** david-lyle__ has joined #openstack-infra | 16:20 | |
| anteaya | morning zaro | 16:20 |
| *** sdake__ is now known as sdake | 16:20 | |
| Shrews | rcarrillocruz: eek. nope | 16:20 |
| clarkb | mtreinish: is it a known issue that when using os-testr I can't do tox -epy27 -- --until-failure some.test.name.here ? | 16:20 |
| clarkb | mtreinish: it gets cranky saying that some.test.name.here is not a valid option? | 16:20 |
| *** david-lyle__ is now known as david-lyle | 16:21 | |
| *** rcernin has quit IRC | 16:21 | |
| openstackgerrit | Merged openstack/python-jenkins: Setup required packages with bindep https://review.openstack.org/341143 | 16:22 |
| *** puiterwijk|mnt is now known as puiterwijk | 16:22 | |
| armax | hello folks, there’s a change https://review.openstack.org/#/c/321146/ in the oven that will allow us to avod installing/running neutron-fwaas in the gate by default, this was a long time coming effort, can anyone give it a nudge? Many thanks! | 16:24 |
| *** trown is now known as trown|lunch | 16:24 | |
| openstackgerrit | Merged openstack-infra/git-review: Install bc libs when running testsuite https://review.openstack.org/341833 | 16:25 |
| *** akshai has quit IRC | 16:27 | |
| *** piet has joined #openstack-infra | 16:27 | |
| openstackgerrit | Anita Kuno proposed openstack-infra/storyboard: Add example commands for the User Tokens api https://review.openstack.org/339297 | 16:28 |
| anteaya | armax: it is a devstack-gate patch so unfortunately I can't help | 16:29 |
| armax | anteaya: thanks nonetheless | 16:29 |
| openstackgerrit | Peter Stachowski proposed openstack-infra/project-config: [mistral] Add check-requirements to mistral-extra https://review.openstack.org/342255 | 16:29 |
| anteaya | armax: thanks for the kindly message of salutation and introduction to the patch | 16:29 |
| *** garyk has quit IRC | 16:30 | |
| *** akshai has joined #openstack-infra | 16:30 | |
| openstackgerrit | Merged openstack/python-jenkins: Fixed a minor typo in example of Python-jenkins https://review.openstack.org/337208 | 16:30 |
| armax | anteaya: I can do it more often :) | 16:30 |
| *** daemontool has quit IRC | 16:30 | |
| anteaya | armax: :) | 16:32 |
| *** dtantsur is now known as dtantsur|afk | 16:32 | |
| *** Swami has joined #openstack-infra | 16:33 | |
| *** cody-somerville has joined #openstack-infra | 16:33 | |
| *** cody-somerville has joined #openstack-infra | 16:33 | |
| *** garyk has joined #openstack-infra | 16:33 | |
| *** Swami_ has joined #openstack-infra | 16:33 | |
| *** Swami_ has quit IRC | 16:34 | |
| *** Swami has quit IRC | 16:34 | |
| *** Swami has joined #openstack-infra | 16:35 | |
| *** lucasagomes is now known as lucas|afk | 16:35 | |
| *** aviau has quit IRC | 16:36 | |
| *** aviau has joined #openstack-infra | 16:36 | |
| *** savihou has quit IRC | 16:36 | |
| armax | fungi: thanks! | 16:37 |
| *** gordc has quit IRC | 16:37 | |
| openstackgerrit | Milan Kováčik proposed openstack-infra/project-config: Run smoke tests after upgrade https://review.openstack.org/342275 | 16:40 |
| openstackgerrit | Merged openstack-infra/nodepool: Fix zk startup detection https://review.openstack.org/341771 | 16:41 |
| *** tphummel has joined #openstack-infra | 16:46 | |
| openstackgerrit | K Jonathan Harker proposed openstack-infra/puppet-hound: Add ServerAlias support to the vhost config https://review.openstack.org/339274 | 16:46 |
| jeblair | clarkb, notmorgan: does this look familiar: http://logs.openstack.org/84/341784/3/check/gate-nodepool-python27-db/7e02de1/testr_results.html.gz | 16:46 |
| *** rdx_matt has quit IRC | 16:48 | |
| *** baoli_ has quit IRC | 16:48 | |
| *** d0ugal has quit IRC | 16:49 | |
| *** matthewbodkin has quit IRC | 16:49 | |
| *** derekh has quit IRC | 16:51 | |
| *** karthikp_ has quit IRC | 16:52 | |
| *** julim has quit IRC | 16:53 | |
| *** xarses has joined #openstack-infra | 16:53 | |
| *** pvaneck has joined #openstack-infra | 16:54 | |
| *** Apoorva has joined #openstack-infra | 16:55 | |
| *** cody-somerville has quit IRC | 16:56 | |
| *** cody-somerville has joined #openstack-infra | 16:56 | |
| *** hasharAway has quit IRC | 16:57 | |
| *** baoli has joined #openstack-infra | 16:57 | |
| *** jpich has quit IRC | 16:57 | |
| *** ifarkas has quit IRC | 16:57 | |
| *** xarses has quit IRC | 16:58 | |
| *** weshay is now known as weshay_food | 16:58 | |
| *** IlyaG has joined #openstack-infra | 16:58 | |
| openstackgerrit | Anita Kuno proposed openstack-infra/storyboard: Add example commands for the User Tokens api https://review.openstack.org/339297 | 16:59 |
| *** mrmartin has quit IRC | 16:59 | |
| *** tkelsey has joined #openstack-infra | 17:00 | |
| *** sdake has quit IRC | 17:00 | |
| *** flepied has joined #openstack-infra | 17:00 | |
| *** karthikp_ has joined #openstack-infra | 17:01 | |
| *** sdake has joined #openstack-infra | 17:01 | |
| *** arxcruz has quit IRC | 17:02 | |
| *** weshay_food is now known as weshay | 17:04 | |
| *** jtomasek has quit IRC | 17:04 | |
| *** asettle has joined #openstack-infra | 17:04 | |
| *** ddieterly is now known as ddieterly[away] | 17:05 | |
| *** esberglu has quit IRC | 17:06 | |
| *** karthikp_ has quit IRC | 17:07 | |
| *** akshai has quit IRC | 17:08 | |
| *** flepied has quit IRC | 17:08 | |
| thiagop | Hello folks. This commit has introduced a change in devstack's output that is undesirable -> https://github.com/openstack-dev/devstack/commit/657cdcdbd18cbdbc185fbf6ef853382b357b3370 | 17:08 |
| thiagop | it doesn't restore the xtrace state | 17:08 |
| thiagop | Is it fine for me to send a patch to just restore the state or is it better to rollback the change and implement it again? | 17:09 |
| openstackgerrit | Merged openstack-infra/nodepool: Use random strings for ZK test paths https://review.openstack.org/341223 | 17:09 |
| *** garyk has quit IRC | 17:09 | |
| *** asettle has quit IRC | 17:09 | |
| anteaya | thiagop: devstack is a repo most often discussed in #openstack-qa | 17:10 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack-infra/project-config: Add check-requirements job to OSA lxc-container-create repo https://review.openstack.org/342289 | 17:10 |
| anteaya | thiagop: you probably will have a more beneificial discussion about your issue in that channel | 17:10 |
| thiagop | anteaya: right, so I'll take the question there | 17:10 |
| anteaya | sounds good | 17:10 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack-infra/project-config: Add check-requirements job to OSA lxc-container-create repo https://review.openstack.org/342289 | 17:11 |
| thiagop | anteaya: tks | 17:11 |
| anteaya | you're welcome | 17:11 |
| *** nadya has joined #openstack-infra | 17:15 | |
| *** PollyZ has quit IRC | 17:15 | |
| *** ddieterly[away] is now known as ddieterly | 17:16 | |
| *** rkukura has quit IRC | 17:16 | |
| *** psilvad has joined #openstack-infra | 17:17 | |
| openstackgerrit | Zara proposed openstack-infra/storyboard-webclient: Comment some settings in worklist resource https://review.openstack.org/342290 | 17:17 |
| *** PollyZ has joined #openstack-infra | 17:17 | |
| *** garyk has joined #openstack-infra | 17:17 | |
| notmorgan | jeblair: transient error iirc | 17:17 |
| *** rderose has joined #openstack-infra | 17:18 | |
| notmorgan | jeblair: it looks like setup failed in allowing access via localhost | 17:18 |
| openstackgerrit | Merged openstack-infra/storyboard: Update .gitignore to ignore all .egg things https://review.openstack.org/341670 | 17:18 |
| notmorgan | jeblair: it's happened infrewquently in my experience and has been low on my list to debug because it was not super common [in my tests/py3 conversion work] | 17:18 |
| rderose | hello, I'm having an issue where Jenkins giving me a -1 when all voting gates are successful | 17:19 |
| openstackgerrit | Merged openstack/python-jenkins: Do not force DeprecationWarning to be displayed https://review.openstack.org/341409 | 17:19 |
| rderose | https://review.openstack.org/#/c/340074/ | 17:19 |
| *** e0ne has joined #openstack-infra | 17:19 | |
| jeblair | notmorgan: ok, i don't recall having seen it so wasn't sure if it was new | 17:19 |
| rderose | any ideas how I can resolve this? | 17:19 |
| *** ajmiller has quit IRC | 17:19 | |
| notmorgan | jeblair: yeah very infrequent | 17:19 |
| *** flepied has joined #openstack-infra | 17:20 | |
| *** dmsimard|afk is now known as dmsimard | 17:20 | |
| *** ajmiller has joined #openstack-infra | 17:20 | |
| *** akshai has joined #openstack-infra | 17:20 | |
| anteaya | rderose: voting jobs | 17:20 |
| *** ddieterly is now known as ddieterly[away] | 17:21 | |
| anteaya | rderose: your patch went through the check pipeline | 17:21 |
| rderose | anteaya: mean, voting gates. | 17:21 |
| rderose | anteaya: yeah, everything passes except: gate-keystone-python35-db-nv, but that is non-voting | 17:22 |
| anteaya | rderose: gates don't vote, jobs do | 17:22 |
| rderose | anteaya: yet Jenkins is giving me a -1 | 17:22 |
| rderose | anteaya: okay, jobs then | 17:22 |
| *** jaosorior has quit IRC | 17:22 | |
| anteaya | gate is a pipline, as is check | 17:22 |
| anteaya | your patch went through the check pipeline | 17:22 |
| rderose | anteaya: yes, this is the patch: https://review.openstack.org/#/c/340074/ | 17:23 |
| anteaya | if you click the button in the gui entitled "Toggle CI" you see the comment Jenkins left | 17:23 |
| anteaya | you will see the following in the comment | 17:23 |
| anteaya | gate-keystone-requirements http://logs.openstack.org/74/340074/21/check/gate-keystone-requirements/11940f4/ : Incompatible requirement found; see http://docs.openstack.org/developer/requirements/ in 1m 41s | 17:23 |
| anteaya | it appears that the requirements job is unhappy | 17:23 |
| rderose | anteaya: hmm... I am adding freezegun to the test-requirments.txt | 17:24 |
| *** jaosorior has joined #openstack-infra | 17:24 | |
| *** OSNEXUS has joined #openstack-infra | 17:25 | |
| clarkb | jeblair: not to me no. Did notmorgan's patch to update db creation for xenial merge? possibly related to that maybe? | 17:25 |
| rderose | anteaya: but again, isn't this a non-voting? | 17:25 |
| *** PollyZ has quit IRC | 17:25 | |
| *** shardy_ has quit IRC | 17:25 | |
| OSNEXUS | Anyone around who can help in troubleshooting my third party CI setup? Having issues with nodepool | 17:25 |
| anteaya | rderose: requirements jobs should only be voting, have you read http://docs.openstack.org/developer/requirements/ as the comment suggests? | 17:26 |
| *** psachin has quit IRC | 17:27 | |
| *** harlowja has joined #openstack-infra | 17:28 | |
| rderose | anteaya: reading it now | 17:28 |
| anteaya | rderose: my read of the keystone jobs is the check-requirements jobs are voting: http://git.openstack.org/cgit/openstack-infra/project-config/tree/zuul/layout.yaml#n7546 | 17:28 |
| *** e0ne has quit IRC | 17:29 | |
| anteaya | OSNEXUS: many people are around, the best strategy is to ask your question, clearly stating each time that you are asking about your third party ci setup | 17:29 |
| anteaya | OSNEXUS: if/when someone is available to help, they will do so | 17:29 |
| OSNEXUS | thanks anteaya | 17:29 |
| anteaya | OSNEXUS: you're welcome | 17:29 |
| clarkb | OSNEXUS: anteaya though it usually helps to start with a specific issue | 17:30 |
| *** PollyZ has joined #openstack-infra | 17:30 | |
| anteaya | clarkb: OSNEXUS agreed | 17:31 |
| annegentle | anteaya or pleia2 or clarkb or fungi : could I ask one of you a couple of questions about SSH keys and OpenStack instances? My conceptual base is weak and I think I need a q&a session. | 17:31 |
| clarkb | annegentle: go for it | 17:31 |
| fungi | absolutely | 17:32 |
| *** rcernin has joined #openstack-infra | 17:32 | |
| anteaya | annegentle: I'm happy to witness you ask questions until you realize you knew the answer all along | 17:32 |
| clarkb | OSNEXUS: I am also happy to help though nodepool issues is pretty nebulous. What is the specific issue or issues | 17:32 |
| *** IlyaG has quit IRC | 17:32 | |
| *** IlyaG has joined #openstack-infra | 17:33 | |
| OSNEXUS | clarkb: I'm setting up a 3rd party CI for a cinder driver, and when trying to read from the sandbox, nodepool seems to error in creating a server, deletes it, then starts up again, it's just looping trying and failing to create nodes | 17:33 |
| OSNEXUS | http://paste.openstack.org/show/532894/ | 17:33 |
| annegentle | anteaya: that's exactly what I need! | 17:34 |
| annegentle | (rubber duck) | 17:34 |
| rderose | anteaya: thank you for your help, I think I see the problem now | 17:34 |
| anteaya | annegentle: :) | 17:34 |
| anteaya | rderose: wonderful, glad you are on your way | 17:34 |
| clarkb | OSNEXUS: looks like you may ahve run out of quota according to that log? I would start there and make sure you ahve enough quota | 17:34 |
| annegentle | Background: I have access to 2 openstack clouds. One cloud creates a user called "user". One cloud creates a user called "cloud-user". | 17:34 |
| anteaya | rderose: and you are welcome | 17:34 |
| annegentle | I'm practicing my empathy for OpenStack users by creating a Heat template that runs an Ansible playbook. | 17:35 |
| notmorgan | clarkb: that error was on my patch. it has been infrequnt and occational but not isolated there | 17:35 |
| *** PollyZ has quit IRC | 17:35 | |
| clarkb | notmorgan: huh | 17:35 |
| notmorgan | clarkb: i've seen it one or two other times. | 17:35 |
| *** SumitNaiksatam has quit IRC | 17:35 | |
| *** roxanaghe has joined #openstack-infra | 17:35 | |
| clarkb | I wonder if its side effect of our db setup in jjb | 17:36 |
| clarkb | jeblair: ^ | 17:36 |
| OSNEXUS | clarkb: okay, where, not seeing anything about quota in nodepool.yaml | 17:36 |
| annegentle | Right now, that playbook has remote_user: cloud-user become: yes become_method: sudo | 17:36 |
| clarkb | OSNEXUS: quota is a cloud construct, you will need to ask your cloud | 17:36 |
| clarkb | OSNEXUS: specifically in that log it says you ahve run out of cores, which is managed by nova so you will need to ask nova | 17:36 |
| annegentle | So, my thinking currently is that well, one, I'll need to make a variable passed in to the playbook for the "remote_user" | 17:36 |
| jeblair | OSNEXUS: OpenStackCloudException: Error in creating instance (Inner Exception: Quota exceeded for cores: Requested 4, but already used 20 of 20 cores (HTTP 403) (Request-ID: req-0dbc5222-c31f-4c6f-9489-4726638ed317)) | 17:36 |
| jeblair | OSNEXUS: ^ is the log line about quota | 17:36 |
| openstackgerrit | Peter Stachowski proposed openstack-infra/project-config: [mistral] Add check-requirements to mistral-extra https://review.openstack.org/342255 | 17:36 |
| jeblair | OSNEXUS: there is another error in there too | 17:36 |
| jeblair | OSNEXUS: OpenStackCloudException: Error in creating the server. | 17:36 |
| jeblair | OSNEXUS: that means that the instance went into the "ERROR" state, which is another thing that your cloud operator can look into | 17:37 |
| annegentle | but two, the part I don't get is, if I need cloud-user to do something as me, in this case, clone a private git repo, I don't want to copy my private key over to that server, that's, ugh. | 17:37 |
| annegentle | so, I think what I want is a "deploy key" according to Github docs. https://developer.github.com/guides/managing-deploy-keys/ | 17:38 |
| jeblair | clarkb: i don't quite follow the 'db setup in jjb' comment | 17:38 |
| clarkb | jeblair: one sec | 17:38 |
| annegentle | Still has some cons, but you know, seems safe enough. | 17:38 |
| OSNEXUS | jeblair , clarkb : thanks, looking through nova conf | 17:38 |
| *** isaacb has joined #openstack-infra | 17:38 | |
| annegentle | My question is this: how do I tell Ansible that those keys are for a github user? | 17:38 |
| clarkb | jeblair: https://git.openstack.org/cgit/openstack-infra/project-config/tree/jenkins/jobs/macros.yaml#n681 the macros for db setup taht start there | 17:39 |
| annegentle | or am I telling the wrong entity the wrong thing? How do you all do this in your multicloud environment? (not the private repos of course, but the "hey, I'm this other user with these keys for this purpose" | 17:39 |
| clarkb | jeblair: since in theory it is responsible for setting that initial connection stuff? | 17:39 |
| notmorgan | clarkb this looks to be not initial connection as i look more closely | 17:39 |
| notmorgan | since only 1 test failed. | 17:40 |
| *** cody-somerville has quit IRC | 17:40 | |
| clarkb | annegentle: you can tell ssh to use a different identity using the -i flag | 17:40 |
| *** roxanaghe has quit IRC | 17:40 | |
| clarkb | annegentle: so in the simplest case you would do ssh -i /path/to/other/key otheruser@wherever | 17:40 |
| fungi | i do exactly that for my openstack infra accounts for example | 17:40 |
| jeblair | clarkb, notmorgan: and that initial setup seems to have worked without error: http://logs.openstack.org/84/341784/3/check/gate-nodepool-python27-db/7e02de1/console.html#_2016-07-14_16_04_13_820613 | 17:40 |
| notmorgan | yah | 17:40 |
| anteaya | annegentle: I also use the -i flag, I create a key for use on one server and ssh into that server specifying the path to the key in the ssh command | 17:41 |
| *** e0ne has joined #openstack-infra | 17:41 | |
| anteaya | annegentle: as clarkb pointed out | 17:41 |
| annegentle | yep, I'm doing that to ssh into it, -i private/key/generated/from/heat/template | 17:41 |
| clarkb | jeblair: notmorgan ya and all the other tests work an in theory they are using new DBs and connections for every test | 17:42 |
| annegentle | but clarkb is onto something I hadn't thought of... -i private/key/as/other/user.... but where is that private key generated? I guess I have to do ssh-keygen once to generate that deploy key? | 17:42 |
| *** sdake_ has joined #openstack-infra | 17:42 | |
| clarkb | annegentle: yes you'd generate it somewhere, put the public key in github or whever, then have ansible/heat install the private key where you need it | 17:43 |
| fungi | annegentle: an alternative is that you can use ssh-agent forwarding, though keep in mind that will result in your forwarded private key being usable while your connection to that system is active (especially a risk if you want unattended use and so don't have the agent require unlocking on every use) | 17:43 |
| clarkb | annegentle: typically we just have service specifc users and they get a single key so it works by default | 17:43 |
| annegentle | clarkb: ok, that helps a ton, that there's this other private key, not necessarily the one for cloud-user or user from the heat template... | 17:43 |
| notmorgan | i did add a cleanup of users at the end of the tests? | 17:43 |
| notmorgan | i wonder if this is a leak of user re-use | 17:43 |
| *** jaosorior_ has joined #openstack-infra | 17:43 | |
| annegentle | fungi: yeah I read a ton about that, and unfortunately I have Windows users in the mix where key forwarding is a non-starter | 17:44 |
| *** tesseract- has quit IRC | 17:44 | |
| *** PollyZ has joined #openstack-infra | 17:44 | |
| fungi | annegentle: and yes, tightly-scoped keys are generally the simplest compromise for the use case you're describing | 17:44 |
| *** sdake has quit IRC | 17:44 | |
| annegentle | clarkb: fungi: anteaya: thanks for this! fungi Yeah I guess that deploy key is more or less a passphrase you protect as such. | 17:45 |
| annegentle | whew!!! unstuck feels good :) | 17:45 |
| anteaya | annegentle: yay, well done | 17:45 |
| clarkb | ok, I think the earlier hostname issue may have been related to the fails in the neutron stadium projects too | 17:45 |
| *** rkukura has joined #openstack-infra | 17:46 | |
| clarkb | workign on confirming that. Also zaqar works on an instance I held from nodepool so no idea what is going on there | 17:46 |
| clarkb | but this is good news means that everything either works or has fixes proposed at this point or is not an active project | 17:46 |
| clarkb | dhellmann: ^ | 17:46 |
| anteaya | yay good news | 17:46 |
| *** jaosorior has quit IRC | 17:46 | |
| *** degorenko is now known as _degorenko|afk | 17:46 | |
| *** tosky has quit IRC | 17:47 | |
| *** trown|lunch is now known as trown | 17:47 | |
| *** jaosorior_ is now known as jaosorior | 17:47 | |
| OSNEXUS | clarkb: jeblair: I am running out of cores, but it seems to me like that's simply a tangential error to the fact that there are failed builds going on that are eating up those cores, and then leaving me with nothing else to work with. Also how would my client check why it's entering an error state? | 17:49 |
| nibalizer | ianw: fungi see my comments on https://review.openstack.org/#/c/341888/2 | 17:49 |
| *** PollyZ has quit IRC | 17:49 | |
| OSNEXUS | Not seeing any nova logging | 17:49 |
| pabelanger | anteaya: thanks, that was the thread I was thinking of | 17:50 |
| *** rdx_matt has joined #openstack-infra | 17:50 | |
| *** jamesmcarthur has quit IRC | 17:50 | |
| *** shashank_hegde has joined #openstack-infra | 17:50 | |
| clarkb | OSNEXUS: the simplest way is to stop nodepool and boot off of that image directly | 17:51 |
| sdake_ | AJaeger pabelanger do either of you know if the mirror issues with available.d are resolved now? | 17:51 |
| *** ociuhandu has quit IRC | 17:51 | |
| clarkb | OSNEXUS: then you can `nova show instanceuuid` or `openstack server show instanceuuid` which shoudl give you a reason for the error state if it enters an error state too | 17:51 |
| pabelanger | sdake_: all images should be uploaded now | 17:51 |
| sdake_ | cool thanks | 17:51 |
| *** sarob has joined #openstack-infra | 17:52 | |
| fungi | nibalizer: thanks a bunch! | 17:54 |
| dhellmann | clarkb : nice! | 17:54 |
| clarkb | sdake_: pabelanger rax-iad centos-7 and rax-ord debian-jessie and rh1 centos-7 may not have the update yet looking at our images | 17:54 |
| *** sarob has quit IRC | 17:54 | |
| fungi | nibalizer: i completely overlooked that. i must be going blind | 17:55 |
| sdake_ | oops ok | 17:55 |
| dulek | Hi, I'm looking for feedback on https://review.openstack.org/#/c/327619/ | 17:55 |
| sdake_ | well let me workflow-1 my patch then | 17:55 |
| *** sarob has joined #openstack-infra | 17:55 | |
| clarkb | sdake_: pabelanger: you'd have to get more precise on when stuff merged and when we succeeded in those locations but they all have older images | 17:56 |
| openstackgerrit | Merged openstack-infra/nodepool: Make ZK cleanup more resilient https://review.openstack.org/341784 | 17:56 |
| sdake_ | the only data i have clarkb is as of 2 days ago rax.ord was lacking the files | 17:56 |
| pabelanger | clarkb: sdake_: Ah, I was only looking at ubuntu-trusty | 17:56 |
| clarkb | sdake_: on what image? | 17:56 |
| *** karthikp_ has joined #openstack-infra | 17:56 | |
| sdake_ | trusty | 17:56 |
| clarkb | the trusty image in rax-ord is a day old | 17:57 |
| clarkb | so should be corrected now | 17:57 |
| sdake_ | cool | 17:57 |
| *** PollyZ has joined #openstack-infra | 17:57 | |
| anteaya | pabelanger: wonderful and you're welcome | 17:58 |
| clarkb | http://git.openstack.org/cgit/openstack/charm-aodh prevented things from updating today | 17:58 |
| pabelanger | Hmm | 17:59 |
| *** mhickey has quit IRC | 17:59 | |
| *** julim has joined #openstack-infra | 17:59 | |
| clarkb | new project creation failing for some reason? | 18:00 |
| * clarkb looks in git logs if thats new | 18:00 | |
| fungi | clarkb: not sure, but there were a ton of new project changes approved earlier today | 18:01 |
| clarkb | yup its new | 18:01 |
| fungi | probably were merging happening at the same time image updates were starting | 18:01 |
| clarkb | well the repo is still not workign | 18:01 |
| fungi | likely something is mistyped on it | 18:02 |
| *** jaosorior has quit IRC | 18:02 | |
| clarkb | I should look into making the dib caching just ignore failed things | 18:02 |
| *** sdake has joined #openstack-infra | 18:02 | |
| clarkb | since we use it as a cahce its not actually fatal unlike what most people will use it for | 18:02 |
| *** [1]Thelo has joined #openstack-infra | 18:02 | |
| *** SumitNaiksatam has joined #openstack-infra | 18:02 | |
| jeblair | clarkb, fungi: unlikely to have an effect, but also recall pabelanger's 'free' ansible change went in recently | 18:03 |
| *** sdake_ has quit IRC | 18:03 | |
| openstackgerrit | Merged openstack-infra/nodepool: Handle NO_AUTO_CREATE_USER default mode https://review.openstack.org/338619 | 18:03 |
| rkukura | meetbot seems dead on openstack-meeting-3 | 18:05 |
| *** Thelo has quit IRC | 18:05 | |
| *** [1]Thelo is now known as Thelo | 18:05 | |
| SumitNaiksatam | yeah, trying to start a meeting | 18:05 |
| SumitNaiksatam | topic shows previous meeting | 18:05 |
| pabelanger | let me check puppetmaster | 18:07 |
| pabelanger | we were having OOMs again | 18:07 |
| pabelanger | so it is possible ansible-playbook is hung again | 18:07 |
| fungi | the meetbot issue was a syntax error. apparently it doesn't like leading space before #commands | 18:07 |
| openstackgerrit | Chris Spencer proposed openstack-infra/project-config: add craton project https://review.openstack.org/332470 | 18:08 |
| *** jamesmcarthur has joined #openstack-infra | 18:08 | |
| OSNEXUS | clarkb: Trying to boot up images is still giving me core issues, how can I clear out my cores now that nodepool is stopped? | 18:08 |
| *** nadya has quit IRC | 18:08 | |
| clarkb | OSNEXUS: you can either increase your quota or delete instances using up quota | 18:08 |
| pabelanger | clarkb: ++ to ignore failures | 18:08 |
| *** pvaneck has quit IRC | 18:09 | |
| *** tkelsey has quit IRC | 18:09 | |
| pabelanger | but ya, strategy free was merged last night for git servers | 18:09 |
| clarkb | pabelanger: the problem is that element is used genericly and I think the default fail is good | 18:09 |
| clarkb | so have to sort out how I can make it not happen for us | 18:09 |
| pabelanger | right | 18:09 |
| *** rajinir has joined #openstack-infra | 18:10 | |
| odyssey4me | pabelanger a quick review of https://review.openstack.org/342289 would be appreciated - it's just adding a missing job | 18:11 |
| pabelanger | I have also seen project creation fail because ansible will fail to run properly on git.o.o servers, and never get to the review.o.o task for remote_puppet_git.yaml | 18:12 |
| odyssey4me | thanks | 18:13 |
| pabelanger | project creation failed, because puppet didn't run, in that case | 18:13 |
| odyssey4me | any other project config cores available for a final vote on https://review.openstack.org/342289 ? | 18:13 |
| *** bhunter71 has joined #openstack-infra | 18:13 | |
| *** kzaitsev_mb has quit IRC | 18:14 | |
| *** psilvad has quit IRC | 18:14 | |
| pabelanger | odyssey4me: have you started using the UCA mirrors yet? | 18:14 |
| *** rbrndt has quit IRC | 18:14 | |
| odyssey4me | pabelanger For out integrated builds, yes. Not yet for all the roles as I have a var precedence issue to sort out to make that work. | 18:15 |
| odyssey4me | pabelanger We were talking the other day about the possibility of having a disabled apt list file placed on the nodepool image to make it easier to use in jobs. | 18:16 |
| *** sdake_ has joined #openstack-infra | 18:16 | |
| pabelanger | odyssey4me: okay cool, was just wanting to confirm they are working as expected | 18:16 |
| odyssey4me | pabelanger yeah, it seems so - thanks! | 18:16 |
| odyssey4me | thanks anteaya | 18:16 |
| *** PollyZ_ has joined #openstack-infra | 18:16 | |
| anteaya | odyssey4me: you're welcome | 18:16 |
| pabelanger | odyssey4me: ya, clarkb came up with the idea of adding /etc/apt/sources.list.available.d so projects could symlink to that at will: http://git.openstack.org/cgit/openstack-infra/project-config/tree/nodepool/scripts/configure_mirror.sh#n78 | 18:17 |
| pabelanger | odyssey4me: so, as we add mirrors, we'll add them into their | 18:18 |
| pabelanger | there* | 18:18 |
| odyssey4me | oh neat, thanks for the pointer - I'll adjust our current method to make use of that :) | 18:18 |
| *** sdake has quit IRC | 18:18 | |
| pabelanger | for example: https://review.openstack.org/#/c/341856/ adds the source.list bits for debian for the debian-openstack-jessie mirror | 18:19 |
| *** ociuhandu has joined #openstack-infra | 18:19 | |
| *** Na3iL has quit IRC | 18:19 | |
| *** PollyZ has quit IRC | 18:19 | |
| OSNEXUS | clarkb: {"message": "No valid host was found. There are not enough hosts available.", "code": 500, "details": " File \"/opt/stack/nova/nova/conductor/manager.py\", line 459, in build_instances | 18:20 |
| clarkb | OSNEXUS: that means you don't have any compute hosts to your your instances on | 18:21 |
| clarkb | OSNEXUS: you'll have to debug your cloud to determine why | 18:21 |
| OSNEXUS | I thought as long as I have open cores nova goes about that? Hmmm, okay | 18:21 |
| *** john-davidge has quit IRC | 18:22 | |
| anteaya | OSNEXUS: well you have a cloud that nodepool is connected to to create nodes on, your cloud account has a quota of some sort | 18:22 |
| anteaya | OSNEXUS: if you surpase the quota, your cloud won't let you create any more nodes, this is what the nodepool logs are trying to tell you | 18:22 |
| *** mikelk has joined #openstack-infra | 18:23 | |
| anteaya | s/surpase/surpass | 18:23 |
| *** rdx_matt has quit IRC | 18:23 | |
| asselin__ | rcarrillocruz, I might've found an issue still with shade & the floating ip address. So I can launch a node, it can ssh to it via public ip, but openstack server list doesn't show the public ip. | 18:24 |
| OSNEXUS | anteaya: Right, but even starting with all my cores free it errors out, and my quota is 20 cores, are there other factors to the quota that might limit it? | 18:24 |
| openstackgerrit | Merged openstack-infra/project-config: Add check-requirements job to OSA lxc-container-create repo https://review.openstack.org/342289 | 18:25 |
| anteaya | OSNEXUS: have you manually created vms on that cloud? | 18:25 |
| anteaya | to ensure your account credentials are valid and the cloud can create nodes? | 18:26 |
| *** PollyZ_ has quit IRC | 18:26 | |
| fungi | could be you have a flavor defined that none of your compute nodes provide? | 18:27 |
| *** maishsk has joined #openstack-infra | 18:27 | |
| OSNEXUS | anteaya: Didn't I just attempt that with nova boot? | 18:27 |
| asselin__ | rcarrillocruz, log output..check test-puppet3 http://paste.openstack.org/show/532903/ | 18:27 |
| *** jamesmcarthur has quit IRC | 18:28 | |
| *** nadya has joined #openstack-infra | 18:29 | |
| *** jamesmcarthur has joined #openstack-infra | 18:31 | |
| asselin__ | rcarrillocruz, so nova caught up and now has the floating ip. I guess it's not really a shade issue, but the symptom shade is fixing (takes too long for nova to get the ip address) | 18:33 |
| clarkb | OSNEXUS: yes, you will need to debug your nova install | 18:33 |
| clarkb | OSNEXUS: nova is saying I have nowhere to run this VM. Make sure you have nova compute(s) running and if you do check their logs to see their available resource calculations | 18:34 |
| *** rbrndt has joined #openstack-infra | 18:36 | |
| *** maishsk has quit IRC | 18:37 | |
| *** xarses has joined #openstack-infra | 18:40 | |
| openstackgerrit | Chris Spencer proposed openstack-infra/project-config: add craton project https://review.openstack.org/332470 | 18:40 |
| OSNEXUS | \HELP | 18:41 |
| *** Sukhdev has joined #openstack-infra | 18:41 | |
| *** eric_lopez has quit IRC | 18:41 | |
| *** pvaneck has joined #openstack-infra | 18:41 | |
| *** pvaneck has quit IRC | 18:41 | |
| *** elo has joined #openstack-infra | 18:42 | |
| isaacb | clarkb: can you please help with processing: https://review.openstack.org/#/c/321480/ | 18:42 |
| clarkb | isaacb: my preference would be that one of the people that already +2'd it approved... but I can take a look I guess | 18:43 |
| *** ddieterly[away] is now known as ddieterly | 18:43 | |
| isaacb | clarkb: thanks | 18:43 |
| *** Sukhdev has quit IRC | 18:44 | |
| *** weshay is now known as weshay_afk | 18:45 | |
| *** Sukhdev has joined #openstack-infra | 18:45 | |
| *** maishsk has joined #openstack-infra | 18:46 | |
| *** dizquierdo has joined #openstack-infra | 18:46 | |
| openstackgerrit | David Shrewsbury proposed openstack-infra/nodepool: WIP: ZooKeeper API https://review.openstack.org/340658 | 18:48 |
| openstackgerrit | David Shrewsbury proposed openstack-infra/nodepool: WIP: ZooKeeper API https://review.openstack.org/340658 | 18:50 |
| *** dizquierdo has quit IRC | 18:52 | |
| *** jamesmcarthur has quit IRC | 18:53 | |
| *** PollyZ has joined #openstack-infra | 18:53 | |
| *** dprince has quit IRC | 18:54 | |
| *** garyk has quit IRC | 18:54 | |
| *** e0ne has quit IRC | 18:57 | |
| *** e0ne has joined #openstack-infra | 18:58 | |
| rcarrillocruz | asselin__: that's what presumably the last change merged fixes | 18:58 |
| rcarrillocruz | use shade from master | 18:58 |
| openstackgerrit | Merged openstack-infra/nodepool: Add log config option to nodepool cmd https://review.openstack.org/321480 | 18:58 |
| rcarrillocruz | should have that fixed/alleviated | 18:58 |
| rcarrillocruz | as you said, it's the way neutron/nova work, it's silly | 18:59 |
| rcarrillocruz | i have to debug an issue where my os_server tests crash nova | 18:59 |
| rcarrillocruz | http://logs.openstack.org/99/342099/7/check/gate-ansible-role-cloud-launcher-dsvm-ansible-func-ubuntu-trusty/04fcad3/console.html | 18:59 |
| rcarrillocruz | :D | 18:59 |
| rcarrillocruz | yay! | 18:59 |
| *** Kaiyan has quit IRC | 19:01 | |
| clarkb | oh its spec day /me quickly goes to do things | 19:01 |
| openstackgerrit | Chris Spencer proposed openstack-infra/project-config: add craton project https://review.openstack.org/332470 | 19:01 |
| *** shamail has joined #openstack-infra | 19:02 | |
| *** ddieterly is now known as ddieterly[away] | 19:05 | |
| fungi | yes, please to do the things. i was going to approve some specs as of 5 minutes ago but am slack about it as usual so it's not done yet | 19:05 |
| fungi | i'll continue to be slack a little longer if you're reviewing still | 19:05 |
| clarkb | I won't vote on my spec but did on the other :) | 19:05 |
| clarkb | so all good | 19:05 |
| clarkb | I was just reflecting on how "uses 0mq" is not a good thing due to our use of it | 19:06 |
| *** ddieterly[away] is now known as ddieterly | 19:06 | |
| *** asettle has joined #openstack-infra | 19:06 | |
| clarkb | and feeling both glad that lessons were learned btu also sad that I am to blame for using it in the first place | 19:06 |
| fungi | you introduced us to its badnez | 19:07 |
| clarkb | mtreinish: fungi another excellent thing to throw on that mqtt bus would be nodepool image infos | 19:07 |
| fungi | absolutely. there are pretty limitless options for things we _could_ add to it (as long as it's things we're comfortable with being 100% public) | 19:08 |
| *** asettle has quit IRC | 19:11 | |
| *** Nikolay_St has joined #openstack-infra | 19:12 | |
| *** nadya has quit IRC | 19:14 | |
| *** sarob has quit IRC | 19:15 | |
| *** IlyaG has quit IRC | 19:17 | |
| *** sdake__ has joined #openstack-infra | 19:19 | |
| asselin__ | rcarrillocruz, right...I realized that after the fact. Ansible/shade can keep going and manage the node, which is awesome, but everyone else needs to wait until the sync happens to get the floating ip. | 19:20 |
| *** rdx_matt has joined #openstack-infra | 19:20 | |
| *** sdake_ has quit IRC | 19:21 | |
| *** tonytan4ever has quit IRC | 19:21 | |
| *** tonytan4ever has joined #openstack-infra | 19:22 | |
| *** karthikp_ has quit IRC | 19:22 | |
| *** rderose has quit IRC | 19:23 | |
| *** karthikp_ has joined #openstack-infra | 19:23 | |
| *** ddieterly is now known as ddieterly[away] | 19:24 | |
| *** e0ne has quit IRC | 19:24 | |
| *** sdake_ has joined #openstack-infra | 19:25 | |
| *** cody-somerville has joined #openstack-infra | 19:26 | |
| *** cody-somerville has joined #openstack-infra | 19:26 | |
| *** PollyZ has quit IRC | 19:26 | |
| *** cody-somerville has quit IRC | 19:28 | |
| *** sdake__ has quit IRC | 19:28 | |
| *** cody-somerville has joined #openstack-infra | 19:28 | |
| *** cody-somerville has quit IRC | 19:28 | |
| *** cody-somerville has joined #openstack-infra | 19:29 | |
| clarkb | pabelanger: get anywhere debugging new project creation? | 19:29 |
| * clarkb digs into that since charms-aodh still is afk | 19:29 | |
| openstackgerrit | John Dickinson proposed openstack-infra/project-config: updated branches gerribot alerts for in #openstack-swift https://review.openstack.org/342370 | 19:29 |
| notmyname | clarkb: ^ follow up to that branch | 19:29 |
| openstackgerrit | John Dickinson proposed openstack-infra/project-config: updated branches gerritbot alerts for in #openstack-swift https://review.openstack.org/342370 | 19:30 |
| fungi | zaro: thanks for all those reviews on open git-review changes! | 19:30 |
| clarkb | https://review.openstack.org/#/admin/projects/openstack/charm-aodh,access looks good | 19:31 |
| *** PollyZ has joined #openstack-infra | 19:31 | |
| pabelanger | clarkb: I didn't see anything specific but the project is updated to | 19:32 |
| pabelanger | I can dig more into it | 19:32 |
| *** rderose has joined #openstack-infra | 19:32 | |
| clarkb | pabelanger: its ok I am looking | 19:34 |
| *** asettle has joined #openstack-infra | 19:35 | |
| pabelanger | clarkb: Yup, puppet just started to run on review.o.o again | 19:35 |
| pabelanger | last run was Jul 14 15:43:37 review puppet-user[24749]: Finished catalog run in 1526.32 seconds | 19:36 |
| clarkb | pabelanger: review.o.o is fine though | 19:36 |
| clarkb | see my link above | 19:36 |
| openstackgerrit | Jeremy Stanley proposed openstack-infra/system-config: Setup ReCaptcha in Wiki to Prevent Spam https://review.openstack.org/342372 | 19:36 |
| pabelanger | most recent was | 19:36 |
| pabelanger | Jul 14 18:26:31 review puppet-user[39818]: Finished catalog run in 1386.21 seconds | 19:36 |
| clarkb | ah and now git.o.o is fine too | 19:36 |
| pabelanger | yup | 19:36 |
| clarkb | wow that was a really long puppet run | 19:36 |
| pabelanger | checking puppetmaster.o.o so see if ansible was failing | 19:36 |
| pabelanger | Ya | 19:36 |
| clarkb | ansible was stuck? | 19:37 |
| openstackgerrit | Jeremy Stanley proposed openstack-infra/puppet-mediawiki: Setup ReCaptcha in Wiki to Prevent Spam https://review.openstack.org/341800 | 19:37 |
| pabelanger | looking now | 19:37 |
| *** ajmiller has quit IRC | 19:38 | |
| *** ajmiller has joined #openstack-infra | 19:38 | |
| pabelanger | clarkb: yes | 19:38 |
| pabelanger | http://paste.openstack.org/show/532911/ | 19:39 |
| pabelanger | timestamps between runs | 19:39 |
| *** e0ne has joined #openstack-infra | 19:39 | |
| pabelanger | Yup, it is blocked now too | 19:39 |
| *** e0ne has quit IRC | 19:39 | |
| clarkb | pabelanger: does that imply it was stuck on phabricator? | 19:39 |
| pabelanger | we are getting SError: [Errno 12] Cannot allocate memory again | 19:39 |
| pabelanger | clarkb: no | 19:39 |
| pabelanger | this is the failure to post our logs to puppetdb | 19:40 |
| *** e0ne has joined #openstack-infra | 19:40 | |
| pabelanger | and running out of memory | 19:40 |
| openstackgerrit | Jeremy Stanley proposed openstack-infra/puppet-mediawiki: Clean up unused questy captcha variable https://review.openstack.org/342374 | 19:40 |
| *** asettle has quit IRC | 19:40 | |
| *** gordc has joined #openstack-infra | 19:41 | |
| fungi | pabelanger: could it be fact posting that's actually bombing memory? i know in the past there have been issues with "facts" containing absurd amounts of data | 19:41 |
| pabelanger | possible | 19:41 |
| *** shamail has quit IRC | 19:42 | |
| *** bhunter71 has left #openstack-infra | 19:42 | |
| pabelanger | the amount of troubles we are having with posting facts, I almost want to refactor that code into a separate ansible-playbook so we can debug it easier. Right now, we have it glob'd into ansible-role-puppet task | 19:42 |
| *** bhunter71 has joined #openstack-infra | 19:43 | |
| *** Goneri has quit IRC | 19:43 | |
| *** nmagnezi has quit IRC | 19:44 | |
| clarkb | the thing that gets me is puppet itself didn't have this problem and it was processing the same data | 19:44 |
| odyssey4me | pabelanger something that may relate, or be useful - you may wish to reduce the fact subset that's gathered... https://github.com/openstack/openstack-ansible/commit/a2326764d3034080ada67f777b546eb4f5c700e8 | 19:44 |
| clarkb | though I guess nodes could talk directly to puppetdb then | 19:44 |
| clarkb | because they had certs and now don't so we proxy through puppetmaster | 19:45 |
| jeblair | clarkb, fungi, mtreinish: i'm pretty excited about adding nodepool/zuul stuff both as producer and potential consumer to mqtt -- but i'd like to defer that until later in v3. the spec reads to me as "shake some things out with gerrit first, then add stuff", so i think we're probably on the same page there, but just want to make sure. | 19:45 |
| fungi | jeblair: absolutely | 19:45 |
| clarkb | yup | 19:45 |
| dhellmann | I'm seeing an error from a validation script in the releases repo that makes me think the local git repo is out of date w.r.t. the existing tags http://logs.openstack.org/07/342307/2/check/gate-releases-tox-validate/ad0562b/console.html#_2016-07-14_19_23_00_906312 | 19:45 |
| fungi | jeblair: if we have gerrit reporting its event stream there by the summit i'll consider that a success | 19:45 |
| clarkb | I was just thinking about sdake_'s question earlier of "are the images up to date" and reading the spec thinking thats a perfect place to solve that problem | 19:46 |
| pabelanger | odyssey4me: thanks, will look at it | 19:46 |
| dhellmann | it's trying to figure out if tag 11.2.17 exists, and what sha it has, and it's not finding the tag even though it's clearly there | 19:46 |
| jeblair | clarkb: we do have another potential solution to that problem that is immediately available: https://review.openstack.org/293706 | 19:46 |
| pabelanger | clarkb: ya, puppetmaster.o.o is MITM now | 19:46 |
| *** weshay_afk is now known as weshay | 19:46 | |
| dhellmann | these are old tags, or I would wonder about a cache issue | 19:46 |
| fungi | jeblair: neither mtreinish nor i have a ton of bandwidth for the firehose project, so i expect it'll be pretty slow-going/low-priority unless a ton of other excited people show up | 19:47 |
| jeblair | clarkb: if you are in principle okay with that, i can refresh it now. | 19:47 |
| dhellmann | the tag isn't on master though; does zuul-cloner limit the branches it clones? | 19:47 |
| dhellmann | or does git log limit the branches it scans somehow? | 19:47 |
| *** julim has quit IRC | 19:47 | |
| *** ddieterly[away] is now known as ddieterly | 19:49 | |
| *** julim has joined #openstack-infra | 19:49 | |
| clarkb | jeblair: ya as a way to query the data that looks fine. My only concern (and this is true of anything we do here) is making sure people understand the time deltas involved when wanting new things. New thing needs to be in before we run dib and the image-list counter only starts after we get image into glance | 19:50 |
| clarkb | jeblair: but that can be sorted out as we go | 19:50 |
| clarkb | dhellmann: yes I think it may not do a full remote update. if ZUUL_BRANCH is set to the branch the tag is on it should work. | 19:51 |
| dhellmann | clarkb : ok, this used to work, so is that a new behavior? | 19:51 |
| dhellmann | clarkb : I don't know in advance what branch the tag is on in a lot of cases | 19:51 |
| dhellmann | and fwiw, this works locally for me | 19:51 |
| clarkb | dhellmann: did it used to work with zuul cloner? | 19:51 |
| dhellmann | yes | 19:51 |
| clarkb | if so then I am likely wrong | 19:52 |
| *** Na3iL has joined #openstack-infra | 19:52 | |
| dhellmann | ok | 19:52 |
| jeblair | dhellmann: we merged a zuul-cloner change recently. there is a possibility there is an unintentional behavior change. | 19:52 |
| dhellmann | jeblair : was that released? or do our nodes install zc from source? | 19:52 |
| jeblair | oddly enough, the change i'm thinking of is so that z-c can be used in the post pipeline | 19:52 |
| odyssey4me | clarkb jeblair dhellmann perhaps this has something to do with it? https://review.openstack.org/293194 | 19:52 |
| jeblair | dhellmann: they install from source | 19:52 |
| dhellmann | that explains why it's different in ci and locally then | 19:52 |
| jeblair | odyssey4me: that's the one i'm thinking of | 19:52 |
| *** rdx_matt has quit IRC | 19:53 | |
| jeblair | dhellmann: can you help me with a test case -- i should see this if i z-c openstack-ansible and look for that tag? | 19:54 |
| dhellmann | I think I'm running "zuul-cloner --workspace $workdir --cache-dir $cachedir git://git.openstack.org openstack/openstack-ansible" | 19:54 |
| dhellmann | jeblair : yes | 19:55 |
| jeblair | git show 11.2.17 | 19:56 |
| jeblair | fatal: ambiguous argument '11.2.17': unknown revision or path not in the working tree. | 19:56 |
| jeblair | dhellmann: then that, right ^ | 19:56 |
| dhellmann | yes, that's the error | 19:56 |
| jeblair | cool | 19:56 |
| asselin__ | rcarrillocruz, wanted to say thanks for your help. I got a node launched using playbooks/launch_node.yml and puppet apply working on it using /opt/system-config/production/playbooks/remote_puppet_else.yaml | 19:56 |
| openstackgerrit | Khai Do proposed openstack-infra/puppet-gerrit: allow override to run git gc instead of repack on repos https://review.openstack.org/334714 | 19:57 |
| jeblair | dhellmann, odyssey4me, clarkb: i still get that error if i revert that commit, and even if i go back to zuul from june 22 | 19:58 |
| *** kzaitsev_mb has joined #openstack-infra | 19:58 | |
| *** asettle has joined #openstack-infra | 19:59 | |
| *** asettle has quit IRC | 19:59 | |
| clarkb | jeblair: there was a change on june 4th to the file to use six | 19:59 |
| clarkb | jeblair: possibly that changed behavior subtly? | 19:59 |
| *** nadya has joined #openstack-infra | 19:59 | |
| jeblair | dhellmann: is it possible we were inadvertently relying on the local cache being updated? ie -- this job ran after the daily image refresh after the tag created? | 19:59 |
| *** jamesmcarthur has joined #openstack-infra | 19:59 | |
| jeblair | clarkb: i'll try that | 19:59 |
| dhellmann | jeblair : this similar release worked 30 hours ago https://review.openstack.org/#/c/341263/2 | 20:00 |
| clarkb | jeblair: but if not that then my enxt guess would be gitpython | 20:00 |
| dhellmann | jeblair : maybe. most of these are really old tags though | 20:00 |
| clarkb | new gitpython on june 20th | 20:00 |
| *** jamesmcarthur has quit IRC | 20:00 | |
| jeblair | clarkb: i'm using gitpython 1.0.1 | 20:01 |
| jeblair | i don't update my venv often :) | 20:01 |
| clarkb | wow ya 2.0.6 is the new version :) | 20:01 |
| jeblair | yeah, still no joy | 20:02 |
| dhellmann | is it possible that the local cache doesn't have the other branches either? | 20:03 |
| *** PollyZ has quit IRC | 20:03 | |
| dhellmann | how is that created? | 20:03 |
| openstackgerrit | Khai Do proposed openstack-infra/system-config: Set cron job to gc repositories instead of repack https://review.openstack.org/342388 | 20:04 |
| clarkb | it is created using disk image builders' source repository element caching thing | 20:04 |
| dhellmann | I can try adding a "git remote update" to my clone instructions but is that going to pull in the extra branches? | 20:04 |
| clarkb | it won't have eg stable/foo but should have remotes/origin/stable/foo | 20:04 |
| dhellmann | yeah, that's fine | 20:04 |
| clarkb | which should be neough to point at that sha1 | 20:04 |
| clarkb | ya | 20:04 |
| dhellmann | as long as it gets all the tags too | 20:04 |
| *** tkelsey has joined #openstack-infra | 20:04 | |
| dhellmann | I'll give that a try | 20:04 |
| dhellmann | I still can't reproduce this locally, fwiw | 20:04 |
| fungi | jeblair: clarkb: will the zuul mergers necessarily have tags? | 20:05 |
| openstackgerrit | Khai Do proposed openstack-infra/puppet-gerrit: allow override to run git gc instead of repack on repos https://review.openstack.org/334714 | 20:05 |
| openstackgerrit | Andreas Florath proposed openstack/diskimage-builder: Refactor: block-device handling (local loop) https://review.openstack.org/319591 | 20:05 |
| clarkb | fungi: the mergers have gerrit job to update their repos for this situation | 20:05 |
| jeblair | fungi: no, the mergers are not guaranteed to be generally up to date | 20:05 |
| clarkb | fungi: but it might not fire for the check queue | 20:05 |
| clarkb | I think it fires for post? | 20:05 |
| jeblair | clarkb, fungi: yeah, for a post job, the repo for the project in question on the designated merger will be up to date | 20:06 |
| jeblair | but that's the only time they get updated. so you can count on it for a post job, but not otherwise | 20:06 |
| *** Na3iL has quit IRC | 20:06 | |
| *** akshai has quit IRC | 20:06 | |
| dhellmann | what is a "zuul merger" and how does that apply? | 20:06 |
| *** Apoorva has quit IRC | 20:06 | |
| fungi | it's the ZUUL_URL that zuul-cloner will pick up as its git remote, where it can expect the precalculated merge commits for jobs to be made available | 20:07 |
| dhellmann | so we might be cloning from a source that doesn't have the tags? | 20:07 |
| *** jamesmcarthur has joined #openstack-infra | 20:08 | |
| *** shashank_hegde has quit IRC | 20:08 | |
| *** tkelsey has quit IRC | 20:09 | |
| *** Apoorva has joined #openstack-infra | 20:09 | |
| fungi | dhellmann: well, we're definitely cloning from a source that might not have tags, since zuul-cloner clones from the local cached git repo on the job node | 20:09 |
| dhellmann | does zuul-cloner rewrite the origin or otherwise add a remote that would have the info? so if I do "git remote update" I'll get them? | 20:10 |
| *** rderose has quit IRC | 20:10 | |
| *** shashank_hegde has joined #openstack-infra | 20:11 | |
| *** pradk has quit IRC | 20:11 | |
| *** rderose has joined #openstack-infra | 20:11 | |
| fungi | i'm pulling up the source now to confirm which remotes it expects and retrieves updates from | 20:11 |
| *** akshai has joined #openstack-infra | 20:11 | |
| fungi | in which sequence | 20:11 |
| clarkb | I am pretty sure its only going to use the cache dir and the one provided | 20:12 |
| jeblair | i think zuul-cloner may never perform a remote update | 20:12 |
| dhellmann | ok, and I'm going ahead and testing with the update command in https://review.openstack.org/342389 | 20:12 |
| fungi | jeblair: yeah, that's what i was looking for | 20:12 |
| clarkb | jeblair: that too | 20:12 |
| dhellmann | oh, wow, ok, I assumed zuul-cloner did a remote update | 20:12 |
| fungi | if it doesn't, it won't automatically know about any new tags | 20:12 |
| dhellmann | does it reset the origin url? | 20:12 |
| jeblair | which is fine for check/gate, but probably is a latent bug in the new 'post' code | 20:12 |
| dhellmann | if not I can do something like "git review -s" to get the gerrit remote setup | 20:13 |
| fungi | dhellmann: it does reset the origin, at http://git.openstack.org/cgit/openstack-infra/zuul/tree/zuul/lib/cloner.py#n90 | 20:13 |
| clarkb | dhellmann: it does set origin url to the one on the command line | 20:13 |
| jeblair | dhellmann: yes: 2016-07-14 19:22:43.107581 | INFO:zuul.Cloner:Updating origin remote in repo openstack/openstack-ansible-os_horizon to git://git.openstack.org/openstack/openstack-ansible-os_horizon | 20:13 |
| *** gyee has joined #openstack-infra | 20:13 | |
| dhellmann | ah, cool | 20:13 |
| *** pradk has joined #openstack-infra | 20:13 | |
| rcarrillocruz | asselin__: np dude, a pleasure ;-) | 20:14 |
| fungi | so anyway, if we git remote update after zuul-cloner does its thing, you should have recent tags | 20:14 |
| clarkb | I wonder if we shouldn't add a remote update to zuul after it calls cloneUpstream? | 20:14 |
| clarkb | would potentially slow things down but be more accurate | 20:15 |
| fungi | i wouldn't want z-c to do that itself though, for performance reasons (remote update can be slow, and we don't usually need it) | 20:15 |
| clarkb | fungi: ya thats the trade off | 20:15 |
| fungi | if a job needs to pick up new tags or branches which aren't related to the commit set up for the change then it can explicitly do that | 20:15 |
| jeblair | clarkb, fungi: possibly in the case where it's not fetching a zuul ref? | 20:15 |
| jeblair | or we could add a --update option :) | 20:16 |
| fungi | yeah, that's a possibility | 20:16 |
| asselin__ | rcarrillocruz, is the intent to have ansible manage this file and have it removed from puppet? /etc/ansible/roles/puppet/templates/puppet.conf.j2 | 20:16 |
| rcarrillocruz | that's a good question | 20:16 |
| rcarrillocruz | nibalizer: ^ | 20:17 |
| dhellmann | adding a git remote update does not seem to have fixed it | 20:17 |
| dhellmann | http://logs.openstack.org/07/342307/3/check/gate-releases-tox-validate/f5fac71/console.html#_2016-07-14_20_15_16_421368 | 20:17 |
| *** yamahata has quit IRC | 20:17 | |
| rcarrillocruz | there's an overlap on template.pp managing puppet.conf and the ansible-role-puppet | 20:17 |
| rcarrillocruz | i assume it's a leftover from puppetmaster setup | 20:17 |
| rcarrillocruz | ? | 20:17 |
| fungi | dhellmann: i don't see git remote update run in that log | 20:18 |
| *** asettle has joined #openstack-infra | 20:18 | |
| dhellmann | I need to add some verbose output, I'll run it again | 20:18 |
| *** garyk has joined #openstack-infra | 20:18 | |
| fungi | just making sure it's run before it's needed | 20:18 |
| fungi | having teh actual output of the remote update call would be helpful so we can see what it's pulling in | 20:19 |
| *** rderose has quit IRC | 20:19 | |
| fungi | maybe also run git tag so we can see the list of known tags | 20:19 |
| jeblair | dhellmann, fungi: running that on my local outdated repo doesn't help | 20:19 |
| fungi | remote update isn't picking up a tag? | 20:19 |
| jeblair | dhellmann, fungi: i note that in gerrit-git-prep.sh, if we are in the post queue for a tag event, we run "git fetch --tags ..." | 20:20 |
| dhellmann | I'll try that, too | 20:20 |
| *** asettle has quit IRC | 20:20 | |
| *** rderose has joined #openstack-infra | 20:20 | |
| openstackgerrit | Alex Kavanagh proposed openstack-infra/project-config: Add charm-barbican, and related https://review.openstack.org/340945 | 20:20 |
| clarkb | jeblair: aha! | 20:20 |
| fungi | jeblair: i wonder why we do that. remote update will usually take care of that | 20:20 |
| clarkb | jeblair: though in this case its not the post queue right? its the check queue? | 20:20 |
| *** hurgleburgler has quit IRC | 20:20 | |
| *** asettle has joined #openstack-infra | 20:20 | |
| fungi | oh! | 20:21 |
| fungi | i bet that tag isn't on master | 20:21 |
| jeblair | clarkb: yeah -- we haven't (until now) had anything in check that cared about tags. the only place we have had something care about tags is 'release' which is a post-style pipeline matching on tag refs | 20:21 |
| asselin__ | rcarrillocruz, yeah, I think it's leftover. Same with hiera.yaml. But the environment.conf is different b/w ansible-puppet and system-config | 20:21 |
| dhellmann | fungi : no, the tag is not on master | 20:21 |
| *** sputnik13 has joined #openstack-infra | 20:21 | |
| fungi | there we go | 20:21 |
| rcarrillocruz | yeah, that's a bit screwy | 20:22 |
| rcarrillocruz | hrmpf | 20:22 |
| dhellmann | fungi : I've not yet made whatever connection you have | 20:22 |
| dhellmann | for this validation script to work, it needs a full clone of the repository with all branches and tags. if zuul-cloner isn't the best way to get that, what should I do instead? | 20:23 |
| fungi | dhellmann: http://paste.openstack.org/show/532913 | 20:23 |
| *** markus__ has quit IRC | 20:23 | |
| *** yaume has quit IRC | 20:23 | |
| dhellmann | does "fetch --tags" also fetch the object to which the tag points? | 20:24 |
| dhellmann | or does the remote update take care of that, but not include the tags? | 20:24 |
| fungi | dhellmann: a "full" clone (depending on how you define that) would be done with `git clone --mirror` which clones a lot more than just things referenced from HEAD | 20:24 |
| dhellmann | well, I need all of the merged commits on all the stable branches and master, and all the tags | 20:25 |
| dhellmann | I don't know if that's "full" or not :-) | 20:25 |
| dhellmann | sounds like maybe not | 20:25 |
| fungi | dhellmann: after fetch --tags, show 11.2.17 does show me the entire commit referenced from that tag too | 20:25 |
| *** asettle has quit IRC | 20:25 | |
| dhellmann | ok, cool | 20:25 |
| dhellmann | I've updated my patch with fetch --tags so I'll see how that goes | 20:25 |
| fungi | but yeah, remote update afaik only fetches tags for "tracked" branches | 20:26 |
| dhellmann | ah | 20:26 |
| *** hurgleburgler has joined #openstack-infra | 20:26 | |
| dhellmann | do I need to make similar changes in the post queue? | 20:28 |
| dhellmann | adding the fetch --tags seems to have helped | 20:28 |
| dhellmann | http://logs.openstack.org/07/342307/4/check/gate-releases-tox-validate/63384d0/console.html | 20:28 |
| jeblair | dhellmann: what does the job in the post pipeline do? | 20:29 |
| fungi | yeah, i suspect so | 20:29 |
| dhellmann | jeblair : it clones the specified branch of a repo, applies a signed tag, then pushes that to gerrit | 20:30 |
| fungi | (or will once it's able to run) | 20:30 |
| jeblair | dhellmann: the job runs after merges to openstack/releases and clones openstack/foo? | 20:30 |
| dhellmann | the job is attached to openstack/releases and uses zuul-cloner to download the repo to receive the tag | 20:30 |
| dhellmann | yes | 20:30 |
| jeblair | dhellmann, fungi: then i think i agree with fungi: yes | 20:31 |
| openstackgerrit | David Shrewsbury proposed openstack-infra/nodepool: WIP: ZooKeeper API https://review.openstack.org/340658 | 20:31 |
| dhellmann | fungi : waiting for your +2a to test :-) https://review.openstack.org/341058 | 20:31 |
| fungi | oh, right! ;) | 20:31 |
| dhellmann | ok, I will make that change | 20:31 |
| *** tonytan4ever has quit IRC | 20:31 | |
| fungi | yeah, i haven't finished the puppet changes 341058 will need to be able to push to gerrit. working on that right now actually | 20:32 |
| jeblair | fungi, clarkb, dhellmann: with that use case also in mind, i'm not sure we can ever be certain automatically when 'fetch --tags' would be needed. so if we want to update zuul-cloner, i think we'd either want to do it in all cases, or add a '--fetch-tags' argument. | 20:33 |
| fungi | now that this new complication is figured out, i'm getting back to it | 20:33 |
| fungi | jeblair: for precedence, as you point out, we already did fetch --tags in ggp (and as i'm looking at it, remote update too for that matter) | 20:33 |
| clarkb | maybe make it default to match ggp the opt out if you know its safe? | 20:35 |
| jeblair | fungi: yes, though the fetch tags is operating in neither of the places needed here :| | 20:35 |
| jeblair | fungi, clarkb: we also wanted to be very deliberate with z-c and remove the unecessary voodoo in ggp, while recognizing that we did learn some things from that :) | 20:35 |
| jeblair | i think that explains the lack of 'git remote update' (it was not always necessary) -- but certainly its use of fetch tags is a good indication of what we thought we needed to do when we really cared about tags | 20:36 |
| pabelanger | what does ggp stand for again? | 20:36 |
| fungi | pabelanger: http://git.openstack.org/cgit/openstack-infra/project-config/tree/jenkins/scripts/gerrit-git-prep.sh | 20:37 |
| jeblair | pabelanger: gerrit-git-prep.sh in project-config | 20:37 |
| openstackgerrit | Doug Hellmann proposed openstack-infra/project-config: force a fuller update of the repo being cloned for tagging https://review.openstack.org/342396 | 20:37 |
| pabelanger | ah | 20:37 |
| fungi | pabelanger: also a jjb builder macro by a similar name | 20:37 |
| fungi | (which simply calls the sforementioned script) | 20:37 |
| pabelanger | danke | 20:37 |
| *** hemanthm is now known as hemanthm|afk | 20:39 | |
| *** IlyaG has joined #openstack-infra | 20:39 | |
| jeblair | yeah, so both of these use cases look like the "checkout branch tip" case for zuul-cloner, which is a pretty common check/gate case where we would not care about tags and probably don't want the extra load on git.o.o | 20:40 |
| *** e0ne has quit IRC | 20:40 | |
| jeblair | so i think the potential zuul-cloner change here would be to add an argument | 20:40 |
| *** mikelk has quit IRC | 20:41 | |
| jeblair | dhellmann: ^ that may be worth doing so that we can have zuul-cloner retry on failure, etc. git.o.o is good, but not always perfect. | 20:41 |
| dhellmann | jeblair : ok, I'll put that on my todo list | 20:41 |
| dhellmann | jeblair : fwiw, when we run these tools by hand, we don't install zuul-cloner from source, we use the released version. I don't know how often you do releases. | 20:42 |
| jeblair | dhellmann: very rarely, though i reckon we'll have one soon. | 20:44 |
| *** rderose has quit IRC | 20:44 | |
| *** rderose has joined #openstack-infra | 20:45 | |
| dhellmann | jeblair : what name do you like for the flag? I can make a note in my todo list | 20:45 |
| dhellmann | --fetch-tags? | 20:45 |
| jeblair | i like that | 20:45 |
| dhellmann | and maybe --update-remotes? | 20:45 |
| dhellmann | for the 2 separate steps | 20:45 |
| dhellmann | or --fetch-remotes? | 20:45 |
| jeblair | i don't think you need to update remotes for this; i *think* fetch --tags is sufficient; i'll retest | 20:46 |
| dhellmann | oh, ok | 20:46 |
| jeblair | (assuming that's the case, i'd probably omit adding an update-remotes option to zc until we know we need it) | 20:46 |
| fungi | it should be, yes | 20:46 |
| fungi | remote update or fetch remotes gets other things like branch names i believe | 20:47 |
| fungi | which in this case we shouldn't need if the job really just cares about tags | 20:47 |
| jeblair | dhellmann: yes, 'git fetch --tags' after z-c runs is all that's needed to make 'git show 11.2.17' work | 20:47 |
| dhellmann | well, the other sort of validation I'm doing is ensuring that a newly tagged version X is an ancestor of existing version Y (or vice versa, whatever) | 20:47 |
| dhellmann | and to do that, I think I need that whole branch? | 20:47 |
| fungi | no, the history for each tag will be available, without branches present | 20:48 |
| dhellmann | ok | 20:48 |
| nibalizer | rcarrillocruz: asselin__ yep, we have to move management of puppet into ansible | 20:48 |
| fungi | once git knows about the tags, it knows their histories | 20:48 |
| *** shashank_hegde has quit IRC | 20:48 | |
| fungi | you would need branches if you want to know what commits are on the branch tips, and between those and the most recent tags on them | 20:49 |
| dhellmann | ok, cool, I wasn't sure if it was just pulling in part of the dag or the whole thing | 20:49 |
| dhellmann | ah, yeah, it does that, too | 20:49 |
| *** akshai has quit IRC | 20:49 | |
| dhellmann | basically I've tried to automate all of the visual validation we were doing to make sure people were putting tags in the right places | 20:49 |
| fungi | ahh, so if you care that there are unreleased commits on stable/mitaka then you'll either need to remote update/fetch remotes or do checkouts of them or something | 20:50 |
| *** rdx_matt has joined #openstack-infra | 20:50 | |
| *** gouthamr has quit IRC | 20:50 | |
| mwhahaha | fungi, greghaynes - what ever was the resolution with the locale stuff? | 20:50 |
| dhellmann | it also tries to show which branch a new tag would end up on by showing which branches contain the named commit, | 20:51 |
| dhellmann | so yeah, I think I do need the remote update step | 20:51 |
| dhellmann | I'm not clear on the difference between "remote update" and "fetch remote" | 20:51 |
| dhellmann | I can look into that when it comes time to implement it, though | 20:51 |
| jeblair | dhellmann, fungi: hrm -- somehow my repo shows the correct origin/stable/mitaka commit | 20:51 |
| jeblair | let me repeat and see if it's z-c or fetch tags that did that | 20:52 |
| fungi | mwhahaha: resolution yes (that dib should create the locale default config file since the systemd package doesn't actually create that file), but i don't know if it's been implemented yet | 20:52 |
| mwhahaha | ok | 20:52 |
| *** eharney has quit IRC | 20:52 | |
| jeblair | dhellmann, fungi: z-c alone causes 'git show remotes/origin/stable/mitaka' to be correct | 20:52 |
| *** sputnik13 has quit IRC | 20:53 | |
| fungi | jeblair: ahh, then that should be fine | 20:53 |
| fungi | though looking in the zuul.lib.cloner source i'm not sure why | 20:54 |
| dhellmann | jeblair : so all I need is the fetch --tags? | 20:54 |
| jeblair | dhellmann, fungi: oh, there is an 'update' in there | 20:55 |
| *** kgiusti has left #openstack-infra | 20:55 | |
| *** nadya has quit IRC | 20:55 | |
| *** akshai has joined #openstack-infra | 20:55 | |
| fungi | in z-c? i don't see it | 20:56 |
| jeblair | we call repo.prune() and then repo.reset() | 20:56 |
| jeblair | repo.reset() calls update() | 20:56 |
| jeblair | which does what you would expect | 20:56 |
| fungi | aha! | 20:56 |
| fungi | i didn't realize repo.reset did that | 20:56 |
| *** akshai has quit IRC | 20:56 | |
| jeblair | its log message is suppressed in z-c apparently -- though we see it on the mergers | 20:57 |
| jeblair | so, er, forget i said anything about update being unecessary. | 20:57 |
| openstackgerrit | Merged openstack-infra/project-config: updated branches gerritbot alerts for in #openstack-swift https://review.openstack.org/342370 | 20:57 |
| *** shashank_hegde has joined #openstack-infra | 20:57 | |
| *** jamesmcarthur has quit IRC | 20:57 | |
| *** Apoorva_ has joined #openstack-infra | 20:57 | |
| fungi | jeblair: that explains it. i missed that cloner is calling into merger for that | 20:58 |
| jeblair | dhellmann: so good news -- we can say with some confidence that z-c will always leave you with a repo that has had 'git remote update' run on it | 20:58 |
| jeblair | the 'fetch --tags' is the only thing missing | 20:58 |
| dhellmann | ah, good | 20:58 |
| dhellmann | ok, so the other step is extra but shouldn't be expensive so I'll just leave it for now | 20:58 |
| jeblair | dhellmann: 'other step'? | 20:59 |
| dhellmann | the remote update | 20:59 |
| *** claudiub has quit IRC | 20:59 | |
| dhellmann | I've already merged that patch in openstack/releases so we aren't blocked on tagging openstack-ansible | 20:59 |
| jeblair | dhellmann: oh, i'd recommend removing it. it's a potential network failure point, and it's actually surprisingly expensive on the git server side (at least until we get the gc stuff fixed) | 21:00 |
| dhellmann | oh, ok | 21:00 |
| *** adrian_otto has quit IRC | 21:00 | |
| dhellmann | hmm, that failed to merge anyway | 21:01 |
| fungi | at this point, if we're already incurring the overhead of remote update, fetch --tags is minor by comparison. i'd just hardcode it in and not bother with a separate option | 21:01 |
| *** Apoorva has quit IRC | 21:01 | |
| jeblair | fungi: oh, hrm. i thought fetch tags might end up shipping a lot of packs over? | 21:01 |
| fungi | more than remote update would? | 21:01 |
| ianw | nibalizer: oops :) | 21:02 |
| *** akshai has joined #openstack-infra | 21:03 | |
| jeblair | fungi: i just ran 'time git remote update' on an oldish nova -- about 5 seconds | 21:03 |
| *** flepied has quit IRC | 21:03 | |
| jeblair | fungi: 'time git fetch --tags' *after* the remote update was 11 seconds and transferred 589KiB | 21:04 |
| fungi | oh wow | 21:04 |
| fungi | yep, i am very wrong about that then | 21:04 |
| *** rfolco has quit IRC | 21:04 | |
| fungi | i bet the eol tags are to blame | 21:04 |
| fungi | i wasn't considering them | 21:04 |
| jeblair | fungi: yeah, there were 3 of them | 21:04 |
| jeblair | fungi: so.... perhaps this is not a representative case for us though? | 21:04 |
| openstackgerrit | Ian Wienand proposed openstack-infra/system-config: Add ianw key and to infra-root https://review.openstack.org/341888 | 21:04 |
| jeblair | fungi: perhaps our usual case is that eol tags are not in play, because our images are rarely *that* old (except for the day after), and maybe there would not be much of a difference? | 21:05 |
| fungi | jeblair: ideally we would be doing that in the git cloning for our dib elements | 21:05 |
| jeblair | fungi: right | 21:05 |
| jeblair | anyone have a week old nova repo? ;) | 21:05 |
| fungi | i _did_ | 21:05 |
| fungi | now it's current because i was testing remote update on it | 21:05 |
| jeblair | fungi: have you tested 'fetch tags' on it? | 21:06 |
| fungi | but also a normal clone wouldn't be representative | 21:06 |
| jeblair | we could move this testing to a node | 21:06 |
| fungi | i hadn't no, and got real 0m4.055s, user 0m0.140s, sys 0m0.136s (no new tags were fetched) | 21:07 |
| *** raildo has quit IRC | 21:07 | |
| fungi | rerunning fetch --tags now goes about twice as fast | 21:07 |
| jeblair | fungi: did it transfer any data? | 21:08 |
| jeblair | fungi: Receiving objects: 100% (2034/2034), 589.19 KiB | 0 bytes/s, done. | 21:08 |
| jeblair | that line | 21:08 |
| fungi | it didn't say that it transferred anything, no | 21:08 |
| fungi | there was no output | 21:08 |
| fungi | i've almost certainly fetched tags since the latest round of eoling | 21:08 |
| *** dizquierdo has joined #openstack-infra | 21:08 | |
| jeblair | okay, i'm persuaded that doing a fetch --tags will usually be fairly harmless for us then | 21:08 |
| fungi | well, the real test would be trying in a clone on one of our job nodes | 21:09 |
| *** piet has quit IRC | 21:09 | |
| fungi | though actually, since we're not doing a clone --mirror from them, we likely aren't pulling unreferenced tags from them | 21:10 |
| fungi | i'll hold one and poke around | 21:10 |
| dhellmann | fungi : http://logs.openstack.org/07/342307/5/check/gate-releases-tox-validate/f2b6cbb/console.html | 21:10 |
| *** _ari_ has quit IRC | 21:10 | |
| ianw | jeblair: any quick tips on debugging http://logs.openstack.org/16/341416/1/experimental/gate-tempest-dsvm-platform-fedora24/35ec3e3/console.html before i dive down the rabbit hole? just bails? | 21:11 |
| *** trown is now known as trown|outtypewww | 21:11 | |
| fungi | dhellmann: yep, that's what i was looking for (before we realized we needed fetch --tags anyway) | 21:11 |
| dhellmann | for some reason I'm not seeing the output of the print statements I added; I wonder what's eating those | 21:11 |
| *** roxanagh_ has joined #openstack-infra | 21:12 | |
| jeblair | ianw: you can find the actual error in http://logs.openstack.org/16/341416/1/experimental/gate-tempest-dsvm-platform-fedora24/35ec3e3/_zuul_ansible/ansible_log.txt | 21:12 |
| *** thorst_ has quit IRC | 21:13 | |
| *** ldnunes has quit IRC | 21:13 | |
| jeblair | ianw: we should consult with Shrews, pabelanger, and mordred since this is very similar to some errors that we think we have corrected or worked around. | 21:13 |
| rcarrillocruz | wow | 21:13 |
| rcarrillocruz | i held a node to debug my role test issue where it crashes nova | 21:14 |
| rcarrillocruz | http://paste.openstack.org/show/532918/ | 21:14 |
| rcarrillocruz | anyone saw this before? | 21:14 |
| *** dizquierdo has quit IRC | 21:14 | |
| rcarrillocruz | clarkb: have you seen ^ before? | 21:14 |
| *** ajmiller has quit IRC | 21:15 | |
| *** thorst_ has joined #openstack-infra | 21:15 | |
| *** ajmiller has joined #openstack-infra | 21:15 | |
| clarkb | rcarrillocruz: I haven't looks like a nova install issue | 21:16 |
| rcarrillocruz | hmm | 21:16 |
| rcarrillocruz | dhellmann: i see you pushed an oslo.config yesterday | 21:16 |
| *** Apoorva_ has quit IRC | 21:16 | |
| *** roxanagh_ has quit IRC | 21:16 | |
| rcarrillocruz | you reckon ^ as a thing that got fixed | 21:16 |
| rcarrillocruz | the logs shows 3.12 | 21:17 |
| rcarrillocruz | latest oslo.config is 3.13 | 21:17 |
| *** burgerk has quit IRC | 21:17 | |
| openstackgerrit | Ramy Asselin proposed openstack-infra/system-config: Remove ansible managed puppet configuration files https://review.openstack.org/342431 | 21:17 |
| dhellmann | rcarrillocruz : did the constraint update land? | 21:17 |
| asselin__ | rcarrillocruz, nibalizer ^^ | 21:17 |
| rcarrillocruz | dhellmann: i'm not sure, where can i look? | 21:18 |
| *** ddieterly has quit IRC | 21:18 | |
| dhellmann | rcarrillocruz : https://review.openstack.org/#/c/341542/ has not merged | 21:18 |
| *** Apoorva has joined #openstack-infra | 21:18 | |
| *** Apoorva has quit IRC | 21:19 | |
| *** maishsk has quit IRC | 21:19 | |
| rcarrillocruz | ahk | 21:19 |
| *** thorst_ has quit IRC | 21:19 | |
| *** Apoorva has joined #openstack-infra | 21:19 | |
| rcarrillocruz | anyway, let me check on git log from 3.12 to 3.13 if there's a patch about my issue | 21:19 |
| *** maishsk has joined #openstack-infra | 21:19 | |
| fungi | jeblair: http://paste.openstack.org/show/532919 trying fetch --tags after cloning from a cached repo on a trusty worker | 21:19 |
| fungi | looks reasonably fast with very little data transferred | 21:20 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul: Use fetch instead of update in merger https://review.openstack.org/342433 | 21:20 |
| *** edtubill has quit IRC | 21:20 | |
| * dhellmann signs off for the evening | 21:20 | |
| jeblair | fungi, clarkb, dhellmann, Nakato: ^ lots of people should double check the wild assertions i just made there | 21:20 |
| fungi | jeblair: also the remote update i did after the cloner call was mostly to confirm it's a no-op (seems to be) | 21:21 |
| jeblair | fungi, clarkb, dhellmann, Nakato: but if that turns out to be the case, maybe we can have nice things (at little cost)? | 21:21 |
| *** flepied has joined #openstack-infra | 21:21 | |
| dhellmann | yay! | 21:21 |
| dhellmann | jeblair : I'll look at that next week, tomorrow is a travel day | 21:22 |
| jeblair | dhellmann: cool, thanks! [what, no gertty? ;)] | 21:22 |
| fungi | dhellmann: planning to be around for the release team meeting? | 21:22 |
| clarkb | jeblair: does update not get tags? | 21:22 |
| *** ccamacho is now known as ccamacho|out | 21:22 | |
| jeblair | clarkb: experimentally, apparently not -- we know that z-c is running 'git remote update origin' for every clone, but we did not see the tags we wanted earlier | 21:23 |
| clarkb | I see. Wasn't the origin the zuul merger which may not have the tags though? | 21:23 |
| *** rdx_matt has quit IRC | 21:23 | |
| *** rcernin has quit IRC | 21:24 | |
| *** rhallisey has quit IRC | 21:24 | |
| *** karthikp_ has quit IRC | 21:24 | |
| jeblair | clarkb: it changes the origin before performing the update | 21:24 |
| fungi | clarkb: no, the origin is the git farm, it separately explicitly fetches from the merger | 21:24 |
| clarkb | ah | 21:24 |
| jeblair | er, yeah, fungi's answer is more correct for that question, sorry | 21:25 |
| fungi | at least in our typical invocation | 21:25 |
| jeblair | (though what i said is also true -- it is the answer to "is the origin the local cache?") | 21:25 |
| clarkb | its just that my reading of the git remote man page is that the default is to fetch tags on fetched branches | 21:25 |
| fungi | it switches the origin from the local cache (that it cloned) to the specified remote | 21:25 |
| clarkb | implying that fetch and update should behave the same there | 21:25 |
| greghaynes | mwhahaha: yep, as fungi said, I havent gotten around to implementing it yet, it should be pretty trivial to do though - just echo out a locale.conf in the yum-minimal element | 21:25 |
| clarkb | but you can't do an explicit grab tags with update so switching to fetch is slgihtly better regardless | 21:26 |
| fungi | clarkb: consider my paste above | 21:26 |
| openstackgerrit | Ramy Asselin proposed openstack-infra/system-config: Explicity use scm: git for roles.yaml https://review.openstack.org/342435 | 21:26 |
| jeblair | clarkb: the only ref i see to that is the '--tags' argument to 'add' which perhaps means that if you add the remote in a certain way, tags will be fetched? | 21:26 |
| jeblair | so maybe we are not adding the remote in that way | 21:26 |
| clarkb | jeblair: ya thats a possibility too | 21:27 |
| clarkb | fungi: ah its tags not on a remote | 21:27 |
| *** sputnik13 has joined #openstack-infra | 21:27 | |
| clarkb | ok I am sold fetch seems to be more fetchy | 21:27 |
| fungi | yeah, so it's retrieving tags which were only on the history of a branch that was subsequently deleted | 21:28 |
| fungi | in this particular case | 21:28 |
| jeblair | clarkb: if you 'git remote add --tags' you get 'tagopt = --tags' in .git/config. if you do it with --no-tags, you get 'tagopt = --no-tags'. and if you omit it, you get no tagopt. so, er, i dunno how that relates to the default. :) | 21:29 |
| clarkb | jeblair: By default, only tags on fetched branches are imported (see git-fetch(1)) | 21:29 |
| jeblair | ah | 21:29 |
| *** hparekh has quit IRC | 21:30 | |
| clarkb | also that was a quote I didn't quote but ya. But that fits into fungi's thing | 21:30 |
| *** amitgandhinz has quit IRC | 21:30 | |
| *** akshai has quit IRC | 21:31 | |
| pabelanger | jeblair: ianw: we should hold a node to see what is going on. Looks like async_wrapper is not writing valid json | 21:31 |
| *** tonytan4ever has joined #openstack-infra | 21:32 | |
| clarkb | pabelanger: thats the issue that modrred was working on iirc | 21:32 |
| fungi | jeblair: clarkb: if i just do a clone from the cached nova repo (not using zuul-cloner), switch the origin to the git farm and fetch --tags (so no remote update involved), i get http://paste.openstack.org/show/532922/ | 21:32 |
| clarkb | so rather than fail right away mordreds patch was keep trying until the timeout in hopes that the json becomes valid | 21:32 |
| pabelanger | clarkb: Oh, could me. Let me see if there is an open PR | 21:32 |
| jeblair | fungi: do you find the new branches surprising? | 21:33 |
| jeblair | oh, maybe not | 21:33 |
| *** cardeois_ has quit IRC | 21:33 | |
| clarkb | fungi: ya that looks correct to me | 21:34 |
| *** cardeois_ has joined #openstack-infra | 21:34 | |
| jeblair | that's just our clone not having the local branches yet, but not necessarily that the cached repo didn't | 21:34 |
| fungi | not surprising, just confirming | 21:34 |
| clarkb | jeblair: the catch doesn't have "local" branches for stable/ | 21:34 |
| clarkb | jeblair: it has the remote refs so has the commits but not the tracking branches | 21:34 |
| fungi | except we wanted it to | 21:34 |
| *** weshay is now known as weshay_afk | 21:34 | |
| clarkb | jeblair: I ran into this doing my big test thing on xenial. I had to do what fungi did in his paste essentially to get things correct | 21:34 |
| fungi | i think something's still not right with my dib patch to be able to specify a * ref to grab all branches and tags | 21:35 |
| jeblair | what if we git fetch from the cache before we change the remote? | 21:35 |
| clarkb | fungi: well and even when you clone from /opt/git it won't set up tracking branches by default. | 21:35 |
| fungi | but yeah, or cached repos do seem to just have master | 21:35 |
| jeblair | (that doesn't sound like it should be necessary...) | 21:35 |
| clarkb | fungi: so functionally I Think this is rouhgly the same | 21:35 |
| fungi | clarkb: right, i mean we may want to clone --mirror from the cache with z-c to reduce remote update/fetch doing more things over the network than it needs | 21:36 |
| jeblair | i guess what i'm worried about is -- does that git fetch have to pull stuff from git.o.o that should be local? | 21:36 |
| clarkb | fungi: ah ya | 21:36 |
| openstackgerrit | Anita Kuno proposed openstack-infra/storyboard: Add example commands for the User Tokens api https://review.openstack.org/339297 | 21:36 |
| fungi | basically what jeblair also said | 21:37 |
| clarkb | jeblair: no its already local because we have remotes/origin/master | 21:37 |
| pabelanger | clarkb: I think this is a little different error, this was mordred PR: https://github.com/ansible/ansible/pull/16458 | 21:37 |
| clarkb | er remotes/origin/stable/foo | 21:37 |
| *** tonytan4ever has quit IRC | 21:37 | |
| pabelanger | clarkb: looks like the failure is just before we start polling | 21:37 |
| *** thorst_ has joined #openstack-infra | 21:38 | |
| pabelanger | which means maybe we didn't copy things properly to the remote host | 21:38 |
| clarkb | and that ref can only point at commits or other refs in repo which means we have them in the cache we just don't have stable/foo tracking remotes/origin/stable/foo as I understand it | 21:38 |
| fungi | clarkb: our cache doesn't currently have those though, which (unrelated) needs fixing in dib still | 21:38 |
| pabelanger | or the first run of async_wrapper | 21:38 |
| clarkb | fungi: it did on my local image | 21:38 |
| fungi | weird | 21:38 |
| clarkb | fungi: whihc I built using the same elements | 21:38 |
| *** cardeois_ has quit IRC | 21:38 | |
| clarkb | oh wait maybe I am mistaken | 21:38 |
| * clarkb checks the repo he had problems with using reno | 21:39 | |
| fungi | clarkb: http://paste.openstack.org/show/532923/ | 21:39 |
| clarkb | ya | 21:39 |
| fungi | that's on a held node | 21:39 |
| pabelanger | ianw: we added the ability into nodepool to auto-hold specific jobs on failures, lets start there and get access to the async_wrapper json file | 21:39 |
| clarkb | thats annoying so yes we don't have the things and a git fetch against cache wouldn't change that | 21:39 |
| fungi | however, we should fix that separately (we intended to have them), so plan as if it's there for now i guess | 21:40 |
| clarkb | fungi: I wonder if the copy into the image is where it breaks | 21:41 |
| clarkb | fungi: let me check on my builder node if the cache for dib has them | 21:41 |
| fungi | here's the bit we wanted it to be doing http://git.openstack.org/cgit/openstack/diskimage-builder/tree/elements/source-repositories/extra-data.d/98-source-repositories#n134 | 21:41 |
| *** thorst_ has quit IRC | 21:42 | |
| clarkb | fungi: http://paste.openstack.org/show/532924/ | 21:42 |
| ianw | pabelanger: sounds good | 21:43 |
| clarkb | so I think the setup on the dib cache side is fine. Its the moving into the image that must be beraking that | 21:43 |
| *** yamahata has joined #openstack-infra | 21:43 | |
| *** ihrachys has joined #openstack-infra | 21:43 | |
| *** hparekh has joined #openstack-infra | 21:43 | |
| fungi | clarkb: yeah, this is probably another one of those "dib wants to clean up everything for you, even things you want it to leave untouched" issues | 21:43 |
| pabelanger | ianw: jeblair: Random guess, but is locale setup properly on the node? | 21:43 |
| clarkb | fungi: http://git.openstack.org/cgit/openstack/diskimage-builder/tree/elements/source-repositories/extra-data.d/98-source-repositories#n173 | 21:43 |
| fungi | clarkb: aww, nuts | 21:44 |
| clarkb | the handling for * there should do what we want I think | 21:44 |
| clarkb | maybe we aren't using *? | 21:44 |
| fungi | clarkb: yeah, i added that too iirc | 21:44 |
| fungi | and i _thought_ i turned it on in our elements | 21:45 |
| fungi | confirming now | 21:45 |
| clarkb | looking at the manifest in my cache it specifies specific sha1s | 21:45 |
| clarkb | so I am guessing no? | 21:45 |
| fungi | clarkb: http://git.openstack.org/cgit/openstack-infra/project-config/tree/nodepool/elements/openstack-repos/extra-data.d/50-create-repo-list#n53 defaults it to * | 21:46 |
| ianw | pabelanger: hmm, same issue with no locale.conf i guess, but that doesn't stop centos. en* locales are there -- it's moved to packages ATM | 21:46 |
| jeblair | pabelanger: i don't think the locale should not affect ansible -- it specifies its own locale settings | 21:46 |
| jeblair | er | 21:46 |
| jeblair | pabelanger: i don't think the locale should affect ansible -- it specifies its own locale settings | 21:47 |
| *** karthikp_ has joined #openstack-infra | 21:47 | |
| *** ihrachys has quit IRC | 21:48 | |
| fungi | pabelanger: ianw: jeblair: is this the same discussion we had earlier in the week about dib's minimal elements needing to create locale.conf because the systemd package only claims it as a "ghost" file and assumes anaconda or similar installer has already created it? | 21:48 |
| clarkb | fungi: I don't see anything setting CONFIG_REF either. But I definitely ahve a manifest without * in it and instead sha1s | 21:48 |
| pabelanger | jeblair: ianw: I only say, because I did find this on the goggles: http://stackoverflow.com/questions/37636257/the-async-task-did-not-return-valid-json | 21:49 |
| pabelanger | could be something to check | 21:49 |
| fungi | clarkb: yep, i'm deep in the weeds there--i probably overlooked something subtle when implementing it. maybe greghaynes might have bandwidth to try and spot what | 21:49 |
| *** aeng has joined #openstack-infra | 21:49 | |
| pabelanger | fungi: I don't think so, see the URL i linked for the reason I was asking | 21:49 |
| ianw | fungi: i'm not sure but yeah still root causing that -> the spec should at least touch it http://pkgs.fedoraproject.org/cgit/rpms/systemd.git/tree/systemd.spec#n340 | 21:49 |
| openstackgerrit | Merged openstack-infra/devstack-gate: Stop using DevStack fwaas code https://review.openstack.org/321146 | 21:50 |
| pabelanger | Looks like that logic in ansible has changed recently too: https://github.com/ansible/ansible/pull/16534 | 21:51 |
| clarkb | fungi: I think I see what it is | 21:51 |
| anteaya | ops midcycle tickets finally are available | 21:51 |
| pabelanger | which resolved: https://github.com/ansible/ansible/issues/16156 | 21:51 |
| clarkb | fungi: we need to set the ref at http://git.openstack.org/cgit/openstack-infra/project-config/tree/nodepool/elements/openstack-repos/extra-data.d/50-create-repo-list#n43 | 21:52 |
| clarkb | fungi: we only do it for the tmp config thing | 21:52 |
| clarkb | let me try making that change and running a local build | 21:52 |
| *** sputnik13 has quit IRC | 21:52 | |
| *** edmondsw has quit IRC | 21:52 | |
| *** amotoki has quit IRC | 21:53 | |
| fungi | clarkb: OH! i only did it for system-config | 21:54 |
| * fungi hangs head in shame | 21:54 | |
| *** sputnik13 has joined #openstack-infra | 21:54 | |
| fungi | totally missed that it needed to be in the projects loop instead | 21:55 |
| clarkb | we should know in about an hour if that fixes it | 21:55 |
| clarkb | I probably could've reduced the elements list to something quicker ohwell | 21:55 |
| fungi | clarkb: also, what i have there in dib _should_ fetch tags i think | 21:56 |
| clarkb | fungi: ya I think the dib stuff proper is probably ok | 21:56 |
| zaro | fungi: made git gc optional, https://review.openstack.org/#/c/342388/ hope this satisfies your concern. | 21:56 |
| clarkb | fungi: we'll know by checking the nova tags list | 21:57 |
| fungi | clarkb: on the other hand, this probably means a huge bump in image sizes when we land the fix :/ | 21:57 |
| clarkb | possibly, I can compare to my older image. Older image was 5.2GB qcow2 | 21:58 |
| fungi | now i'm glad you didn't strip down the projects list | 21:58 |
| fungi | yes please | 21:58 |
| fungi | zaro: thanks, looking now | 21:58 |
| clarkb | -rw-r--r-- 1 root root 5586769920 Jul 11 20:49 devstack-gate.qcow2 is the old one | 21:59 |
| clarkb | owned by root because you must be this root to drive the dib train | 21:59 |
| fungi | heh | 21:59 |
| *** bhunter71 has quit IRC | 21:59 | |
| *** flepied has quit IRC | 21:59 | |
| fungi | rootin' tootin' dib! | 21:59 |
| clarkb | unrelated but I'd like to restart nodepool today as soon as the image uploads complete. This will pick up the new demand calculation code | 22:00 |
| clarkb | jeblair: ^ | 22:00 |
| *** gouthamr has joined #openstack-infra | 22:02 | |
| *** openstackgerrit has quit IRC | 22:03 | |
| *** gouthamr_ has joined #openstack-infra | 22:03 | |
| *** openstackgerrit has joined #openstack-infra | 22:03 | |
| jeblair | clarkb: ++ | 22:03 |
| jeblair | i will see how fast i can fix my webapp change :) | 22:04 |
| clarkb | kk | 22:05 |
| *** sputnik13 has quit IRC | 22:05 | |
| clarkb | you probably have time. uploads are slow | 22:05 |
| *** mriedem has quit IRC | 22:06 | |
| *** armax has quit IRC | 22:06 | |
| *** armax has joined #openstack-infra | 22:07 | |
| *** gouthamr has quit IRC | 22:07 | |
| ianw | yeah so http://pkgs.fedoraproject.org/cgit/rpms/systemd.git/tree/systemd.spec#n340 is a package build-time thing, not install. lucky i wasted all that time going through strace's of package installs to figure that out :( | 22:08 |
| *** piet has joined #openstack-infra | 22:08 | |
| openstackgerrit | James E. Blair proposed openstack-infra/nodepool: Add a simple status webapp https://review.openstack.org/293706 | 22:08 |
| jeblair | clarkb, fungi: ^ rebased (the hold message was the conflict) | 22:08 |
| *** elo has quit IRC | 22:09 | |
| openstackgerrit | Kien Ha proposed openstack-infra/jenkins-job-builder: Update live-screenshot plugin to use convert xml https://review.openstack.org/342446 | 22:09 |
| *** elo has joined #openstack-infra | 22:09 | |
| jeblair | oh wait let me fix the 'zuul' thing i keep missing :) | 22:10 |
| openstackgerrit | James E. Blair proposed openstack-infra/nodepool: Add a simple status webapp https://review.openstack.org/293706 | 22:10 |
| jeblair | there we go | 22:10 |
| *** cody-somerville has quit IRC | 22:12 | |
| clarkb | nibalizer: hurgleburgler random feature request for your vinz stuff since its something that bugs me occasionally. Make file mode changes super apparent on the front page change summary | 22:12 |
| clarkb | nibalizer: hurgleburgler: file modes matter for things like dib elements and its easy to get them wrong and not notice since its only in the diff view at the top iirc | 22:13 |
| *** shashank_hegde has quit IRC | 22:13 | |
| *** rkukura has quit IRC | 22:14 | |
| *** karthikp_ has quit IRC | 22:17 | |
| *** sputnik13 has joined #openstack-infra | 22:18 | |
| *** tpsilva has quit IRC | 22:18 | |
| *** cody-somerville has joined #openstack-infra | 22:18 | |
| clarkb | jeblair: lgtm I did point out a piece of new dead code but you can do a followup change if thats easier | 22:19 |
| openstackgerrit | James E. Blair proposed openstack-infra/nodepool: Remove dead code from nodepoolcmd https://review.openstack.org/342450 | 22:20 |
| *** devananda has quit IRC | 22:20 | |
| jeblair | clarkb: thanks! ^ | 22:20 |
| *** rdx_matt has joined #openstack-infra | 22:21 | |
| *** devananda has joined #openstack-infra | 22:21 | |
| *** sputnik13 has quit IRC | 22:22 | |
| *** bhavik has quit IRC | 22:23 | |
| *** cody-somerville has quit IRC | 22:23 | |
| *** cody-somerville has joined #openstack-infra | 22:23 | |
| *** cody-somerville has quit IRC | 22:23 | |
| *** cody-somerville has joined #openstack-infra | 22:23 | |
| *** ihrachys has joined #openstack-infra | 22:24 | |
| *** rlandy has quit IRC | 22:24 | |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul: Add dynamic reconfiguration https://review.openstack.org/340597 | 22:24 |
| openstackgerrit | Kien Ha proposed openstack-infra/jenkins-job-builder: Update Logfilesizechecker to use convert xml https://review.openstack.org/342452 | 22:24 |
| openstackgerrit | Kien Ha proposed openstack-infra/jenkins-job-builder: Update Logfilesizechecker to use convert xml https://review.openstack.org/342452 | 22:25 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul: Remove v2.5 ansiblelaunchserver https://review.openstack.org/342454 | 22:25 |
| *** shashank_hegde has joined #openstack-infra | 22:25 | |
| *** garyk has quit IRC | 22:28 | |
| *** ihrachys has quit IRC | 22:30 | |
| *** shashank_hegde has quit IRC | 22:30 | |
| *** Apoorva_ has joined #openstack-infra | 22:31 | |
| clarkb | timothyb89: hey just noticing that npm complains about a few packages having potential dos issues when installing stackviz | 22:31 |
| clarkb | timothyb89: its also complaining about a couple pacakges htat will fail on newer node which we probably don't want. Not suire if these thigns are already on your radar | 22:32 |
| *** rkukura has joined #openstack-infra | 22:32 | |
| *** karthikp_ has joined #openstack-infra | 22:32 | |
| nibalizer | clarkb: cool | 22:33 |
| *** shashank_hegde has joined #openstack-infra | 22:33 | |
| nibalizer | i daresay you could make a storyboard story | 22:33 |
| *** tonytan4ever has joined #openstack-infra | 22:33 | |
| clarkb | nibalizer: I can do that if its how you would like things tracked | 22:33 |
| nibalizer | since probably I will forget | 22:33 |
| nibalizer | im donw to use storyboard | 22:33 |
| *** Apoorva has quit IRC | 22:33 | |
| nibalizer | hurgleburgler: thoughts? | 22:34 |
| clarkb | I am going to reenable account 20395 in gerrit as they have sent mail to the infra list sayign they won't run third party ci on their personal account | 22:34 |
| clarkb | any objections to that? | 22:35 |
| fungi | nope, thanks. i flagged that message in case nobody else got around to it | 22:36 |
| *** sputnik13 has joined #openstack-infra | 22:36 | |
| clarkb | done | 22:37 |
| clarkb | I will respond on the list | 22:37 |
| clarkb | fungi: my dib build failed (I think I ran out of disk space) but I am able to poke at hte half built state in /tmp and I have all the branches there for nova | 22:37 |
| clarkb | fungi: I will go ahead and push my fix to the element in a few | 22:37 |
| *** tonytan4ever has quit IRC | 22:38 | |
| anteaya | clarkb: I missed that post | 22:38 |
| *** mdrabe has quit IRC | 22:38 | |
| anteaya | ah now I found it | 22:39 |
| anteaya | okay, no objections, the person clearly states that they understand what a personal gerrit account is for | 22:39 |
| anteaya | thank you | 22:39 |
| *** kien-ha has joined #openstack-infra | 22:40 | |
| jhesketh | Morning | 22:40 |
| anteaya | morning jhesketh | 22:40 |
| fungi | clarkb: i guess no dice on getting an image size increase estimate then | 22:41 |
| clarkb | fungi: not yet at least I am in the process of freeing disk space and will rerun | 22:41 |
| *** karthikp_ has quit IRC | 22:41 | |
| clarkb | I just have to find all the places that dib stashes data because I am not deleting all of it based on df output | 22:42 |
| fungi | yeah, having a rough idea in a review comment of what size jump we're looking at would help | 22:43 |
| fungi | thanks | 22:43 |
| ianw | while we're talking about disk size, i'll still like to track it in nodepool with https://review.openstack.org/#/c/289678/ | 22:44 |
| jeblair | ianw: lgtm | 22:45 |
| *** pahuang has joined #openstack-infra | 22:49 | |
| *** sputnik13 has quit IRC | 22:49 | |
| *** esberglu has joined #openstack-infra | 22:50 | |
| *** xyang1 has quit IRC | 22:50 | |
| *** shashank_hegde has quit IRC | 22:52 | |
| openstackgerrit | Kien Ha proposed openstack-infra/jenkins-job-builder: Update MongoDB plugin https://review.openstack.org/342460 | 22:52 |
| *** sputnik13 has joined #openstack-infra | 22:54 | |
| *** rdx_matt has quit IRC | 22:54 | |
| openstackgerrit | Clark Boylan proposed openstack-infra/project-config: Cache all branches in dib repo cache https://review.openstack.org/342461 | 22:55 |
| clarkb | fungi: ^ will let you know the build results as soon as I manage to get a build | 22:56 |
| *** rbrndt has quit IRC | 22:57 | |
| *** xarses has quit IRC | 22:59 | |
| *** ddieterly has joined #openstack-infra | 22:59 | |
| *** flepied has joined #openstack-infra | 22:59 | |
| *** shashank_hegde has joined #openstack-infra | 23:00 | |
| craige | o/ | 23:02 |
| *** ams__ has quit IRC | 23:04 | |
| *** cardeois_ has joined #openstack-infra | 23:04 | |
| *** tonytan4ever has joined #openstack-infra | 23:05 | |
| *** dimtruck is now known as zz_dimtruck | 23:05 | |
| *** yamamoto has joined #openstack-infra | 23:06 | |
| *** jcoufal has quit IRC | 23:06 | |
| *** ddieterly has quit IRC | 23:06 | |
| *** jcoufal has joined #openstack-infra | 23:06 | |
| openstackgerrit | Morgan Fainberg proposed openstack-infra/zuul: Do not contest locks in PythonGit https://review.openstack.org/342462 | 23:07 |
| openstackgerrit | Morgan Fainberg proposed openstack-infra/zuul: Python 3 fixes: wrap iter() returns in list() https://review.openstack.org/342463 | 23:07 |
| openstackgerrit | Morgan Fainberg proposed openstack-infra/zuul: WIP: Create a helper to ease encode/decode on gear jobs https://review.openstack.org/342464 | 23:07 |
| notmorgan | jeblair: ^ those three patches have me down to ~3 python3 fails. | 23:07 |
| *** shashank_hegde has quit IRC | 23:08 | |
| notmorgan | oops | 23:08 |
| *** jcoufal has quit IRC | 23:08 | |
| openstackgerrit | Morgan Fainberg proposed openstack-infra/zuul: WIP: Create a helper to ease encode/decode on gear jobs https://review.openstack.org/342464 | 23:09 |
| *** cardeois_ has quit IRC | 23:09 | |
| *** berendt has quit IRC | 23:10 | |
| *** shashank_hegde has joined #openstack-infra | 23:10 | |
| *** bhunter71 has joined #openstack-infra | 23:12 | |
| *** thiagop has quit IRC | 23:13 | |
| *** adrian_otto has joined #openstack-infra | 23:13 | |
| jeblair | notmorgan: not interested in changing gear? | 23:14 |
| notmorgan | jeblair: a real headache | 23:14 |
| *** thorst_ has joined #openstack-infra | 23:14 | |
| notmorgan | jeblair: spent most of the week trying to get it to work, and short of restructuring internally a huge volume of things, it wasn't easy | 23:15 |
| notmorgan | since the same job objects are used in many different ways on both client/server side | 23:15 |
| notmorgan | i also spoke with SpamapS and greghaynes about it and decided it would be better/cleaner to isolate this on the consuming side | 23:15 |
| clarkb | 6 image uploads left, all in ovh bhs1 | 23:15 |
| jeblair | notmorgan: okay; i'm starting to think this may be becoming a distraction from v3. perhaps we should shelve it for now? | 23:16 |
| *** rkukura has quit IRC | 23:16 | |
| notmorgan | jeblair: eh, sure? *shrug* it's mostly done now. once i dumped the gear fixes. | 23:16 |
| pabelanger | clarkb: Ya, it seems slow theses days | 23:16 |
| openstackgerrit | Merged openstack-infra/storyboard: Add example commands for the User Tokens api https://review.openstack.org/339297 | 23:17 |
| SpamapS | IMO python3 should generally be be something where you make progress, ensure it has unit test coverage, but don't compromise urgent tasks for it. | 23:17 |
| notmorgan | jeblair: at least the first fix should be included regardless, the way we were writing out our temp configs was not guaranteed to be locked | 23:17 |
| *** karthikp_ has joined #openstack-infra | 23:17 | |
| SpamapS | notmorgan: so does gear basically accept bytes most of the time in its API? | 23:18 |
| notmorgan | SpamapS: part of the fixes for py3 were to get clean web-based console logs again | 23:18 |
| notmorgan | SpamapS: yes bytes is accepted everywhere important | 23:18 |
| jeblair | notmorgan: yeah, i still think that if gear is requiring that in zuul, then something is wrong in gear. but we've gone back to zero on this twice and i don't want to burn any more time on it. | 23:18 |
| SpamapS | notmorgan: that's unfortunate.. :-P | 23:18 |
| * SpamapS prefers his libraries to take strings :) | 23:19 | |
| mordred | I tihnk we can mange the console log thing without going al th eway down the py3 hole for zuul if the hole is too deep | 23:19 |
| greghaynes | SpamapS: You mean on the server side? | 23:19 |
| *** EricGonczer_ has quit IRC | 23:19 | |
| greghaynes | SpamapS: oh, as a client | 23:19 |
| *** thorst_ has quit IRC | 23:19 | |
| SpamapS | just in general, if I have a library giving me an object, I would prefer that it take strings most of the time. | 23:19 |
| jeblair | SpamapS, notmorgan, greghaynes: no, i believe gear accepts and/or itends to accept some string | 23:20 |
| *** thorst_ has joined #openstack-infra | 23:20 | |
| jeblair | i think there has been some misunderstanding about that | 23:20 |
| jeblair | i thought we had cleared it up on the mailing list | 23:20 |
| jeblair | but apparently not | 23:20 |
| SpamapS | There are exceptions, like I/O libraries. And gear is kind of dancing on that line.. since it might be that you really do need to give it raw bytes. | 23:20 |
| notmorgan | it accepts strings for name/unique but not for arguments/data | 23:20 |
| notmorgan | and that is perfectly fine | 23:20 |
| greghaynes | Yea, when you say gear I am unclear if that means gear the protocol, the server, or the client library | 23:20 |
| notmorgan | but it also returns raw bytes | 23:20 |
| greghaynes | and thats what keeps confusing me | 23:21 |
| jeblair | greghaynes: is the protocol named gear? | 23:21 |
| notmorgan | greghaynes: gear in this case is our lib. | 23:21 |
| jeblair | i thought it was called the "gearman protocol" | 23:21 |
| greghaynes | well, there is a protocol that we use.. not sure the name | 23:21 |
| greghaynes | ok, sure | 23:21 |
| jeblair | http://gearman.org/protocol/ | 23:21 |
| greghaynes | so then server or client? | 23:21 |
| jeblair | that's the title of that | 23:21 |
| *** kien-ha has quit IRC | 23:22 | |
| greghaynes | what I keep wondering is - if were talking about the server, where does the server ever do any string manipulation | 23:22 |
| *** PollyZ has joined #openstack-infra | 23:22 | |
| SpamapS | greghaynes: we're not talking about the server. Or at least, I'm not | 23:23 |
| jeblair | greghaynes: it performs comparisons, and aggregrates things like function names into status command output | 23:23 |
| greghaynes | ok, so comparison and function names seem like things that shouldnt require any decoding | 23:23 |
| *** sputnik13 has quit IRC | 23:24 | |
| *** thorst_ has quit IRC | 23:24 | |
| notmorgan | and most of that is in-fact compared in bytes-form, but it still consumes/uses the same data model. | 23:24 |
| SpamapS | actually as I think about it.. you really do need to accept bytes even in the client lib | 23:24 |
| SpamapS | because your gear client may be talking to java worker | 23:25 |
| SpamapS | and you can't really assume encodings then | 23:25 |
| clarkb | SpamapS: yes that was my comment from ysetday | 23:25 |
| SpamapS | you want the bytes to match up | 23:25 |
| jeblair | notmorgan: that's chiefly for convenience -- the server is explicitly less supported than the client. it's marked as being experimental. the api should not be considered frozen at all. | 23:25 |
| *** sputnik13 has joined #openstack-infra | 23:25 | |
| notmorgan | jeblair: right, it's a lot of work to split that out compared to a small helper object | 23:25 |
| *** PollyZ has quit IRC | 23:25 | |
| notmorgan | jeblair: i got about 30% into splitting it and felt that the shorter path was better for *now* and could be circled back. | 23:26 |
| *** asettle has joined #openstack-infra | 23:26 | |
| jeblair | notmorgan: for me, this is an argument that we should do it later at a time when we are prepared to | 23:26 |
| notmorgan | (30% was about an hour but the next 70% is going to be days and days) | 23:26 |
| openstackgerrit | Anita Kuno proposed openstack-infra/storyboard: Add example commands for the Users api https://review.openstack.org/338570 | 23:27 |
| jeblair | SpamapS, clarkb: i think i am comfortable saying that function names in gear will be utf8 on the wire by default, and if you need something else, you can use a less convenient api. | 23:27 |
| jeblair | SpamapS, clarkb: that is the historic behavior in gear and it is reasonable for almost every imaginable circumstance | 23:27 |
| *** rook- has quit IRC | 23:28 | |
| jeblair | (this is what i said in my email, i think) | 23:28 |
| jeblair | (by 'less convenient api' i mean an alternate attribute/argument in gear -- not a different library) | 23:28 |
| jeblair | ('name_bytes' instead of 'name', etc) | 23:28 |
| *** armax_ has joined #openstack-infra | 23:28 | |
| clarkb | ya I think accepting bytes too is fine probably | 23:29 |
| clarkb | my comment yesterday was specifically if you waent ot use a non utf8 encoding because talking to somethign that doesn't utf8 then bytse would be useful | 23:30 |
| SpamapS | jeblair: the protocol is silent on this. the strings are opaque strings of bytes. | 23:30 |
| jeblair | clarkb: to be clear, i'm saying that Job(name=u'foo') gets automatically encoded to utf8, and Job(name_bytes=b'bar') gets sent as-is. | 23:30 |
| *** asettle has quit IRC | 23:30 | |
| SpamapS | so it feels a bit weird to impose utf-8 | 23:30 |
| jeblair | SpamapS: understood. | 23:30 |
| *** armax has quit IRC | 23:31 | |
| *** armax_ is now known as armax | 23:31 | |
| clarkb | jeblair: I think as long as the api is there and tested then thats doable | 23:31 |
| *** amitgandhinz has joined #openstack-infra | 23:31 | |
| jeblair | SpamapS: i think that to expose function names as bytes only makes the library unusable in python3. i think it's worth the compromise. | 23:31 |
| openstackgerrit | James Slagle proposed openstack-infra/tripleo-ci: Add mulitnode CI job support to tripleo-ci https://review.openstack.org/324777 | 23:31 |
| *** SumitNaiksatam has quit IRC | 23:31 | |
| SpamapS | It would certainly feel clumsy to use if everything had to be bytes. | 23:32 |
| jeblair | notmorgan: do you have your original zuul patch that did that handy ^ ? | 23:32 |
| notmorgan | jeblair: hmm? | 23:32 |
| *** pradk_ has quit IRC | 23:32 | |
| notmorgan | that did what? | 23:32 |
| jeblair | SpamapS: yes, i would not choose to use gear if it required that | 23:32 |
| jeblair | notmorgan: your patch to zuul that did '.encode' everywhere | 23:32 |
| notmorgan | oh sure | 23:32 |
| *** rkukura has joined #openstack-infra | 23:32 | |
| SpamapS | and actually, there's really no reason for a 'name_bytes' | 23:32 |
| notmorgan | it's still in zuul master branch, sec | 23:32 |
| notmorgan | https://review.openstack.org/#/c/325661/ | 23:33 |
| *** dingyichen has joined #openstack-infra | 23:33 | |
| SpamapS | you should be in control of whatever is setting 'name', and be able to decode using whatever you know the encoding to be | 23:33 |
| notmorgan | ignore a few things are bundled there | 23:33 |
| notmorgan | that is because otherwise the transient between patches would take ~2hrs to fail the py3 job | 23:33 |
| SpamapS | feeling very corner-casey here | 23:33 |
| *** zz_dimtruck is now known as dimtruck | 23:33 | |
| mordred | SpamapS: check the patch above | 23:33 |
| mordred | this is what started us down this path | 23:34 |
| jeblair | yeah, that's the patch that made me say "we need to fix gear/py3" because that is not a usable api | 23:34 |
| notmorgan | SpamapS: i would rather have a library say "we impose utf-8" than "encode with whatever you want" | 23:34 |
| mordred | I think we'd all actually prefer that python3 bytes/strings worked differently than it did :) | 23:34 |
| notmorgan | SpamapS: if you don't like that about the library, either use a different one, or implement something yourself. | 23:34 |
| clarkb | could you have a GearUTF8 class for client that did these things and use that as the place to do your encopding? | 23:35 |
| notmorgan | esp. when it's fairly niche like gear is (we use it, and i guarnatee aomost no one uses it in py3 until recently) | 23:35 |
| clarkb | then its consumable by users like zuul, but leaves the thigns are bytes for people that need not utf8? also would serve as a template for extending it to add not utf8 things | 23:35 |
| notmorgan | since there were py3 breaking bugs that if you actually used it, it would just fail | 23:35 |
| *** amitgandhinz has quit IRC | 23:36 | |
| jeblair | notmorgan: i think we can do the name_bytes thing for those corner cases when someone wants to have a function name encoded in gb18030 | 23:36 |
| *** xarses has joined #openstack-infra | 23:36 | |
| notmorgan | jeblair: sure. | 23:36 |
| SpamapS | hah | 23:36 |
| SpamapS | durn gb18030 | 23:36 |
| * mordred proposes that we migrate all of openstack to gb18030 encoding | 23:37 | |
| jeblair | clarkb: that may be a possibility as well -- the slightly harder thing is to make sure that when you have a gear.Worker, it gives you a UTF8Job or whatever. | 23:37 |
| clarkb | jeblair: ya I haven't looked at the itnernals too carefully thinking more from a UX perspective | 23:37 |
| SpamapS | another idea.. | 23:37 |
| notmorgan | and keep in mind, we still need to explicitly encode/decode in zuul for data/arguments | 23:37 |
| SpamapS | is to just have an optional encoding parameter | 23:38 |
| jeblair | maybe that entails passing in a job factory argument or something | 23:38 |
| notmorgan | those are data blobs that could be any "struct" form | 23:38 |
| SpamapS | so when you instantiate your objects, you pass encoding='foo' and it decodes with that. | 23:38 |
| notmorgan | so the fact we use json is convienent | 23:38 |
| jeblair | notmorgan: agreed | 23:38 |
| notmorgan | but we can't impose that at the gear level | 23:38 |
| clarkb | SpamapS: ya that could even go under my idea if we want a strong default in place | 23:38 |
| openstackgerrit | Anita Kuno proposed openstack-infra/storyboard: Add example commands for the Users api https://review.openstack.org/338570 | 23:38 |
| clarkb | SpamapS: like utf8 calss just calls your thing with utf8 encoding | 23:38 |
| *** zhurong has joined #openstack-infra | 23:38 | |
| clarkb | *calls | 23:39 |
| mordred | clarkb: s/calls/class/ I believe :) | 23:39 |
| clarkb | mordred: bah | 23:39 |
| SpamapS | because really, the corner case we're trying to handle is "Oh man my gear blew up because of invalid utf-8 coming from {{ other side of gearman pipe }} what do I do?" Just letting people make a client object that encodes/decodes with whatever you need is entirely sufficient. | 23:39 |
| SpamapS | the payload is being decoded too yes? | 23:40 |
| notmorgan | SpamapS: the payload has to be decoded by the client, regardless of name/unique | 23:40 |
| notmorgan | SpamapS: we can't force utf-8 here | 23:40 |
| clarkb | fungi: looks like I am getting a working build this time. Will know image size soon I hope | 23:40 |
| notmorgan | SpamapS: it could be a packed data struct, it could be protobuf | 23:41 |
| notmorgan | SpamapS: that is "data" and "arguments". | 23:41 |
| SpamapS | makes sense and agree | 23:41 |
| jeblair | and i think it's entirely reasonable to insist that be passed in as bytes and blow up if not | 23:41 |
| notmorgan | we just happne to use json, which ahs the same issue as str() | 23:41 |
| notmorgan | it needs explicit encoding. | 23:41 |
| notmorgan | (since it's str() at it's heart) | 23:42 |
| *** thorst_ has joined #openstack-infra | 23:42 | |
| mordred | we could switch to protobuf | 23:42 |
| mordred | and punt on the whole thing | 23:42 |
| notmorgan | mordred: different time / place / argument, and i'd 100% support that though. | 23:42 |
| notmorgan | but thats just me. | 23:42 |
| jeblair | mordred: that doesn't solve the problem at hand | 23:43 |
| notmorgan | but we'd need to deal with today and getting there anyway :P | 23:43 |
| anteaya | mordred: do you think the P poll will be able to be fixed and continue or do you think you will have to restart it? | 23:43 |
| mordred | anteaya: it will be fixed and continue | 23:43 |
| anteaya | thank you | 23:43 |
| mordred | anteaya: it is actually not broken, just most people's links are | 23:43 |
| *** bhunter71 has quit IRC | 23:43 | |
| mordred | there are 675 votes cast in it, amainzgly enouhg | 23:44 |
| * notmorgan has a broken link. | 23:44 | |
| mordred | what I have learned is that NOBODY reads the mailing list | 23:44 |
| * notmorgan doesn't know how to fix the link or would. | 23:44 | |
| jeblair | and yeah, let's please have the protobuf argument some other time because i'm not convinced :) | 23:44 |
| notmorgan | mordred: i also generally ignore the ML atm. | 23:44 |
| mordred | notmorgan: I just have to resend you a link ... it's not hard, I just haven't had time yet | 23:44 |
| notmorgan | jeblair: i have strong opinions on that, but it's such a change it's not worth a headache. | 23:44 |
| notmorgan | jeblair: for not. | 23:45 |
| notmorgan | now* | 23:45 |
| jeblair | notmorgan: convince me over our next beer :) | 23:45 |
| notmorgan | also depends on if we ever want a language other than python | 23:45 |
| mordred | jeblair: I was mostly kidding. I think that if we were already ina world where the legwork and infrastructure were already in place for protobuf it would be neat, but it's not worth the cost to get from here to there | 23:45 |
| anteaya | mordred: I read the mailing list | 23:45 |
| notmorgan | if we don't / don't need to transit craziness like conductor(s), the benefits are meh | 23:45 |
| mordred | anteaya: yes, you do. the hundreds of people who have sent me messages telling me that something is wrong do not | 23:46 |
| notmorgan | mordred: oh let me send you a few hundred more emails :P | 23:46 |
| *** Swami has quit IRC | 23:46 | |
| anteaya | mordred: well I read the bit about the poll, I did miss the infra mailing list post about the request to reenable a gerrit account | 23:46 |
| mordred | on the positive side, I've only gotten one angry email from someone who is upset that they are receiving unsolicited email - so that's good | 23:46 |
| notmorgan | jeblair: so.. i think i'm saying... i cant/wont convince you :) | 23:46 |
| anteaya | so I don't _always_ read the mailing list | 23:46 |
| anteaya | but yay, fun recieving lots of personal email | 23:47 |
| notmorgan | jeblair: unless we determine the use-cases trend that way... cause... i'd rather drink beer and not talk protobuf w/ ya | 23:47 |
| mordred | anteaya: I never read the mailing list, statistically speaking | 23:47 |
| mordred | notmorgan, jeblair: we could drink beer and talk about smoking meat | 23:47 |
| anteaya | mordred: I believe that | 23:47 |
| *** pradk has quit IRC | 23:47 | |
| notmorgan | mordred: and hack on code around a pool? in dallas? | 23:47 |
| jeblair | so anyway, on this -- i'd still like to suggest that we shelve or back-burner this for now. but when we want to proceed, if we want to continue with the utf8/alternate attribute approach, that's fine. if we want to change to the "encoding=utf8" approach, that someone write that up and send it to the infra ml. | 23:47 |
| notmorgan | mordred: cause... that seems like a win | 23:47 |
| clarkb | fungi: -rw-r--r-- 1 root root 5627776000 Jul 14 23:46 devstack-gate.qcow2 so that file size doesn't really change much. I am slightly suspicious and will mount it and check the repos | 23:47 |
| mordred | jeblair: wfm. I can happily propose other ways of dealing with console streaming between now and the time that we're ready to deal with them | 23:48 |
| *** piet has quit IRC | 23:48 | |
| notmorgan | mordred: yeah we can probably brew something easy up w/o websockets | 23:48 |
| *** tonytan4ever has quit IRC | 23:49 | |
| notmorgan | mordred: wont be as nice to work with though. but meh. | 23:49 |
| jeblair | mordred: i'm thinking we just refocus on the mainline v3 stuff, then console streaming after | 23:49 |
| mordred | jeblair: right. sorry, that's what I meant | 23:49 |
| mordred | I just meant that when we get there, there are options we can discuss if py3 is too hard/distracting | 23:50 |
| jeblair | mordred: that way we can unblock all the work about testing, conversion, etc. | 23:50 |
| *** matt-borland has quit IRC | 23:50 | |
| *** thorst_ has quit IRC | 23:51 | |
| jeblair | mordred: ok | 23:51 |
| *** rdx_matt has joined #openstack-infra | 23:51 | |
| clarkb | fungi: all the branches show up tehre so I think it is safe to go ahead and merge my change and images won't be terribly affected | 23:52 |
| clarkb | also its not looking like i am going to get to restart nodepool today | 23:52 |
| clarkb | ianw: ^ you might not be interested in doing that on a firday | 23:53 |
| *** shashank_hegde has quit IRC | 23:53 | |
| *** cody-somerville has quit IRC | 23:54 | |
| jeblair | notmorgan: if you're game, i'd like to get your help on the v3 configuration changes. it's still a bit of a mess, and there are places where i don't think we've found a good api. | 23:54 |
| *** david-lyle has quit IRC | 23:54 | |
| jeblair | notmorgan: https://review.openstack.org/340597 doesn't make things much better :) | 23:54 |
| *** shashank_hegde has joined #openstack-infra | 23:55 | |
| *** cody-somerville has joined #openstack-infra | 23:55 | |
| jeblair | notmorgan: but if you can take a look at that (it will be hard and it will prompt lots of questions) and then maybe work with me on some of the next steps, maybe we can suss out what that stuff should actually look like | 23:55 |
| *** mtanino has quit IRC | 23:55 | |
| *** gordc has quit IRC | 23:56 | |
| clarkb | I guess we could just decide to drop the ovh-bhs1 uplaods and resetart | 23:56 |
| jeblair | notmorgan: (i'm basically trying to make all the job configuration examples in http://specs.openstack.org/openstack-infra/infra-specs/specs/zuulv3.html work) | 23:56 |
| *** mika has quit IRC | 23:56 | |
| clarkb | I dont' think the status thing has merged though | 23:57 |
| clarkb | so maybe we wait | 23:57 |
| jeblair | i will approve my change and ianw's so hopefully they will be there for the next restart | 23:57 |
| *** mriedem has joined #openstack-infra | 23:57 | |
| notmorgan | looking at the patch and the spec. | 23:58 |
| jeblair | i also can't restart them now, but could do so tomorrow if image builds aren't precious | 23:58 |
| *** puiterwijk is now known as puiterwijk|mnt | 23:59 | |
| jeblair | notmorgan: cool. i am happy to talk at length about anything in there. | 23:59 |
| clarkb | jeblair: sounds good | 23:59 |
| *** hurgleburgler has quit IRC | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!