| lbragstad | jamielennox: yep, that sounds right. I believe it is within context switching | 00:01 |
|---|---|---|
| *** galstrom is now known as galstrom_zzz | 00:01 | |
| lbragstad | in the oslo-incubator code | 00:01 |
| jamielennox | lbragstad: the ideal way would be to get oslo to stop assuming that eventlet is going to be present | 00:01 |
| jamielennox | but that's going to be a slow process | 00:02 |
| lbragstad | jamielennox: right, kinda what I was thinking | 00:02 |
| *** zhuadl has joined #openstack-dev | 00:03 | |
| lbragstad | jamielennox: specifically, https://github.com/openstack/oslo-incubator/blob/master/openstack/common/local.py#L22 | 00:03 |
| jamielennox | lbragstad: what that requires i've really got no idea, i like my environment separation but from a quick look through the RPC code it would be a pretty big change | 00:03 |
| jamielennox | lbragstad: even looking at https://review.openstack.org/#/c/29803/5/keystone/openstack/common/rpc/amqp.py there is a lot of dependant stuff | 00:04 |
| jamielennox | lbragstad: the other hack i've seen is to check if socket or os is patched and in which case import eventlet stuff | 00:05 |
| lbragstad | jamielennox: right, I just saw that. I know we have a blueprint in keystone for notifications as well. Which there doesn't seem to be a good way of getting around | 00:05 |
| jamielennox | but again looking at rpc they are specifically doing green thread pools | 00:05 |
| lbragstad | jamielennox: have there been other projects that have done that? | 00:05 |
| jamielennox | not as far as i'm aware, but there are a number of projects that are trying to rid themselves of eventlet | 00:06 |
| jamielennox | swift i know is | 00:06 |
| jamielennox | mind you i just grepped there source and there is eventlet scattered everywhere | 00:07 |
| lbragstad | keystone source or other projects as well? | 00:08 |
| jamielennox | as i said i was under the impression swift was trying to accomodate running without it but a quick look through the code shows a lot of hard dependencies on eventlet | 00:09 |
| lbragstad | most of it in Keystone seems to be isolated in keystone/common/environment/*. I remember that going through | 00:11 |
| jamielennox | lbragstad: yea, that was mine and that seperation is part of why adam -2ed the logging stuff | 00:11 |
| *** markwash has quit IRC | 00:12 | |
| lbragstad | right, the logging stuff would have introduced it into another place | 00:12 |
| jamielennox | lbragstad: so i guess it would be interesting to know if you ignore the eventlet stuff of the oslo patch how much work is in writing our own RPC call | 00:13 |
| jamielennox | lbragstad: i dont know what it would take, but again looking at rpc most of the complication seems to be related to using eventlet anyway | 00:13 |
| jamielennox | if we are running synchronously then we don't need to worry about any thread pooling and return wait checking | 00:15 |
| jamielennox | maybe? interesting implications to that if you are doing a network call on every log | 00:15 |
| *** epim has quit IRC | 00:16 | |
| jamielennox | no idea why they made a logging interface on top of a synchronous network call, that just seems dumb | 00:16 |
| *** cliu has joined #openstack-dev | 00:16 | |
| lbragstad | so at this point its worth it to consider writing our own unified logging implementation | 00:17 |
| lbragstad | and holding off on pulling in anything from oslo | 00:18 |
| *** msmedved has left #openstack-dev | 00:18 | |
| *** msmedved has quit IRC | 00:18 | |
| lbragstad | in which case we would either have to hold off on the notifier service as well, or implement our own | 00:18 |
| jamielennox | lbragstad: i'm not sure, i don't claim to be an authority on this and i really don't know much about how the rpc works in oslo | 00:19 |
| jamielennox | as we are only a sender and never a receiver or rpcs it might be easy to wrap simply the logging interface until there is some sort of eventlet consensus | 00:20 |
| *** msmedved has joined #openstack-dev | 00:20 | |
| jamielennox | but i've got no idea what this would take, there is also some talk about how keystone should be queing notifications about this like when a user is deleted | 00:21 |
| jamielennox | so i guess if it can be interfaced fairly easily it's worth it, otherwise we have to suck it up and take eventlet back | 00:22 |
| jamielennox | lbragstad: but how i became an authority on this i've got no idea :) | 00:22 |
| lbragstad | jamielennox: haha you'll have to talk to Adam on that one, | 00:23 |
| lbragstad | jamielennox: when you say interfaced you're talking about notification, logging or both? | 00:23 |
| *** RajeshMohan has quit IRC | 00:24 | |
| jamielennox | lbragstad: so i'm not sure what is underpinning logging and notifications, i understand/think it is a synchronous rpc call, you would probably know better | 00:24 |
| *** cliu has quit IRC | 00:25 | |
| jamielennox | looking through oslo's rpc there is a lot of work with eventlet and format checking that is very generic | 00:26 |
| *** RajeshMohan has joined #openstack-dev | 00:26 | |
| jamielennox | if all logging boils down to is call ampq with this json data then we can probably do that ourselves | 00:26 |
| jamielennox | i think there is a fair bit of pressure to do notifications so that will probably have to happen soon anyway so we will need to either bring that in from oslo or look at it ourselves, but again i'm not really sure how that mechanism works | 00:28 |
| lbragstad | I think the problem with logging was it was using corolocal to switch between contexts | 00:28 |
| lbragstad | I don't seeing logging in oslo having deps on rpc or notifier | 00:28 |
| lbragstad | see* | 00:29 |
| *** portante has joined #openstack-dev | 00:30 | |
| *** msmedved has quit IRC | 00:30 | |
| jamielennox | ok so i'm likely wrong and maybe the easiest way of handling this is just a hack in oslo that if eventlet isn't isntalled then don't use the corolocal stuff | 00:30 |
| *** msmedved has joined #openstack-dev | 00:30 | |
| jamielennox | so i was looking at the failed review rather than the oslo code, why does brining in logging also bring in RPC? | 00:33 |
| lbragstad | well, there are a couple ways to do it. After playing around with oslo and asking some questions you can either apply a module independently or try applying multiple oslo modules to a project at once | 00:34 |
| lbragstad | sometimes it is common to applie rpc and notifier in the same patch since notifier is dependent on rpc | 00:34 |
| lbragstad | I *think* in this case, if we fixed the context adapter issue with log.py https://github.com/openstack/oslo-incubator/blob/master/openstack/common/log.py#L258 in oslo then I think we could try and apply it to keystone separately instead of with other modules, like I did in the first patch. I figured I would bring them all in (rpc, log, local, notifier) at once since I was going to be stepping right into the notifications work right after I finishe | 00:36 |
| jamielennox | yea it looks like 2 of the drivers rely on rpc | 00:36 |
| *** bdpayne has quit IRC | 00:36 | |
| *** sleepsontheflo-1 has quit IRC | 00:37 | |
| *** msmedved has quit IRC | 00:37 | |
| jamielennox | right | 00:37 |
| jamielennox | ok | 00:37 |
| jamielennox | Sorry for talking about all the other stuff, it appears most of rpc can be irrelevant | 00:38 |
| lbragstad | after having dims and markmc explain it to me, it sometimes goes by the content in which you're bringing things into a project, and what would happen if it needed to be backed out. No worries, it all needs to be addressed sooner or later | 00:38 |
| *** ffio has quit IRC | 00:39 | |
| *** jasdeepH has quit IRC | 00:39 | |
| jamielennox | so what is the purpose of having a weak store here? | 00:39 |
| *** cliu has joined #openstack-dev | 00:39 | |
| *** noslzzp has quit IRC | 00:39 | |
| *** ffio_ has joined #openstack-dev | 00:40 | |
| jamielennox | lbragstad: why is it doing this at all? | 00:41 |
| lbragstad | something pertaining to the scope of the context | 00:42 |
| lbragstad | I don't think I truely understood problem that was solved by this | 00:42 |
| *** galstrom_zzz is now known as galstrom | 00:43 | |
| jamielennox | lbragstad: neither | 00:43 |
| *** bdpayne has joined #openstack-dev | 00:44 | |
| jamielennox | mikal: if you are around, what is the purpose of the context in message logging? why do you need to use a corolocal? | 00:44 |
| *** vipul is now known as vipul|away | 00:45 | |
| jamielennox | lbragstad: it would appear that it's just a way of storing weakrefs on eventlet | 00:45 |
| openstackgerrit | A change was merged to openstack/ceilometer: api: enable v2 tests on SQLAlchemy & HBase https://review.openstack.org/33936 | 00:45 |
| jamielennox | i guess that when you do green context switching if you have weakrefs it must mess with there scoping | 00:46 |
| lbragstad | makes sense | 00:46 |
| jamielennox | lbragstad: all that class would appear to be doing is storing the variable in a green thread local way to deal with the scoping | 00:46 |
| jamielennox | so | 00:47 |
| *** sthaha has joined #openstack-dev | 00:47 | |
| *** sthaha has joined #openstack-dev | 00:47 | |
| jamielennox | what i would suggest is to submit a patch to osol with a similar approach to what cms.py does in keystone | 00:47 |
| *** yaguang has joined #openstack-dev | 00:47 | |
| lbragstad | ok | 00:48 |
| lbragstad | looking | 00:48 |
| jamielennox | check if we are running under eventlet, if we are then have the WeakLocal class inherit from corolocal | 00:48 |
| jamielennox | otherwise define your own WeakLocal class which is simply an object with a __getattr__ stores weakrefs | 00:48 |
| jamielennox | from memory there is a weakref dictionary that would make this really easy | 00:49 |
| jamielennox | lbragstad: but try stepping through it because given that it would appear that context is simply a greenthread local global variable | 00:51 |
| lbragstad | ok | 00:52 |
| *** Mandell has quit IRC | 00:52 | |
| jamielennox | so it might work to simply replace it with a global context variable | 00:52 |
| openstackgerrit | A change was merged to openstack/nova: Refresh volume connections when starting instances https://review.openstack.org/31992 | 00:53 |
| *** cliu has quit IRC | 00:53 | |
| *** alexpilotti has quit IRC | 00:54 | |
| jamielennox | yea, local is used in a number of places throughout oslo and it does appear to just be a way of doing a request specific global variable | 00:54 |
| jamielennox | I would go the 'make your own weak_store and strong_store' route, that way it will fix all of them | 00:54 |
| lbragstad | ok, then we don't have check for running on eventlet like in cms.py? | 00:55 |
| jamielennox | yea you will, because this will be brought into other projects that will be using eventlet | 00:55 |
| lbragstad | ahh ok | 00:55 |
| *** vipul|away is now known as vipul | 00:55 | |
| jamielennox | then i would just not import the RPC log handlers into keystone at all | 00:56 |
| lbragstad | just see if we can get away with import log.py and local.py | 00:57 |
| *** lloydde_ has quit IRC | 00:57 | |
| *** bknudson has quit IRC | 00:58 | |
| jamielennox | yea, so long as we are using oslo's logging interface then we can argue with them over how rpc works later | 00:58 |
| jamielennox | that'll be a tough fight and i would be trying to get others to have it | 00:59 |
| lbragstad | yeah, Nova is already using that implementation of notifier | 00:59 |
| *** zhuadl has quit IRC | 00:59 | |
| *** Ryan_Lane has quit IRC | 01:03 | |
| openstackgerrit | A change was merged to openstack/nova: Add basic SecurityGroup model https://review.openstack.org/33494 | 01:04 |
| openstackgerrit | A change was merged to openstack/nova: Add Instance.security_groups https://review.openstack.org/33495 | 01:04 |
| openstackgerrit | A change was merged to openstack/nova: Make InstanceList filter non-column extra attributes https://review.openstack.org/33701 | 01:04 |
| lbragstad | jamielennox: one last question, where in cms.py do you actually detect eventlet? Or is that in environment? | 01:05 |
| *** Ryan_Lane has joined #openstack-dev | 01:05 | |
| jamielennox | lbragstad: ah right because i would have converted that over to use environment, try this one: https://github.com/openstack/python-keystoneclient/blob/master/keystoneclient/common/cms.py | 01:06 |
| *** SumitNaiksatam has joined #openstack-dev | 01:07 | |
| lbragstad | jamielennox: ahh, _ensure_subprocess, makes sense, I was a little confused looking at https://github.com/openstack/keystone/blob/master/keystone/common/cms.py | 01:10 |
| lbragstad | thanks | 01:10 |
| *** gyee has quit IRC | 01:10 | |
| jamielennox | lbragstad: before you go have a look at https://review.openstack.org/#/c/34484/ for me, it's breaking devstack so it will need to be fixed quickly | 01:11 |
| lbragstad | jamielennox: sure | 01:11 |
| *** mikal has quit IRC | 01:13 | |
| lbragstad | jamielennox: maybe a stupid question, but why from eventlet.green import httplib in keystone/middleware/ec2_token.py and only import httplib in keystone/middleware/s3_token.py? | 01:13 |
| jamielennox | no idea | 01:13 |
| jamielennox | i reverted it back to its pre- environment state | 01:14 |
| lbragstad | gotcha, yeah looks good | 01:14 |
| jamielennox | there are so many inconsistencies with how eventlet is used which was part of the reason for environment in the first place | 01:15 |
| *** tjones has left #openstack-dev | 01:15 | |
| *** mikal has joined #openstack-dev | 01:15 | |
| jamielennox | i'll add a comment to explain that | 01:15 |
| *** redbeard2 has joined #openstack-dev | 01:16 | |
| lbragstad | Yeah that makes sense, | 01:16 |
| *** pabelanger has joined #openstack-dev | 01:18 | |
| *** comay has quit IRC | 01:18 | |
| *** tjones has joined #openstack-dev | 01:18 | |
| *** redbeard2 has quit IRC | 01:20 | |
| *** erkules_ has joined #openstack-dev | 01:21 | |
| *** ffio_ has quit IRC | 01:22 | |
| *** erkules has quit IRC | 01:23 | |
| *** tjones1 has joined #openstack-dev | 01:24 | |
| *** tjones1 has left #openstack-dev | 01:25 | |
| *** torandu has quit IRC | 01:26 | |
| *** tjones has quit IRC | 01:28 | |
| *** yuanz has quit IRC | 01:34 | |
| *** jasdeepH has joined #openstack-dev | 01:34 | |
| *** redbeard2 has joined #openstack-dev | 01:36 | |
| *** terry7 has quit IRC | 01:37 | |
| *** sungju has joined #openstack-dev | 01:40 | |
| *** galstrom is now known as galstrom_zzz | 01:40 | |
| *** bdpayne has quit IRC | 01:42 | |
| *** adalbas has quit IRC | 01:47 | |
| *** Tedster has quit IRC | 01:52 | |
| *** wenjianhn has joined #openstack-dev | 02:00 | |
| openstackgerrit | A change was merged to openstack/oslo-incubator: rpc: remove some unused serialization code https://review.openstack.org/34316 | 02:04 |
| *** zigo has quit IRC | 02:09 | |
| *** zigo has joined #openstack-dev | 02:10 | |
| *** shang has joined #openstack-dev | 02:11 | |
| *** ctracey` has quit IRC | 02:11 | |
| *** noslzzp has joined #openstack-dev | 02:12 | |
| *** jbresnah has quit IRC | 02:12 | |
| *** ctracey|away has joined #openstack-dev | 02:13 | |
| *** ctracey|away is now known as ctracey | 02:13 | |
| openstackgerrit | A change was merged to openstack/oslo-incubator: Add IpFilter, IPNetnsExecFilter and EnvFilter https://review.openstack.org/29636 | 02:14 |
| *** abhisri has joined #openstack-dev | 02:19 | |
| *** nati_ueno has quit IRC | 02:21 | |
| *** abhisri has quit IRC | 02:21 | |
| *** shang has quit IRC | 02:23 | |
| *** sdake has quit IRC | 02:23 | |
| *** bdpayne has joined #openstack-dev | 02:25 | |
| *** stevemar has joined #openstack-dev | 02:26 | |
| *** yanglyy has joined #openstack-dev | 02:30 | |
| *** portante_ has joined #openstack-dev | 02:30 | |
| *** portante has quit IRC | 02:34 | |
| *** matiu has quit IRC | 02:35 | |
| *** shang has joined #openstack-dev | 02:36 | |
| *** rcleere has joined #openstack-dev | 02:37 | |
| *** bing_bu has joined #openstack-dev | 02:42 | |
| *** jasdeepH has quit IRC | 02:43 | |
| *** melwitt has quit IRC | 02:43 | |
| *** jbresnah has joined #openstack-dev | 02:48 | |
| *** novas0x2a|laptop has quit IRC | 02:49 | |
| *** shang has quit IRC | 02:50 | |
| *** WindowsFanboi has joined #openstack-dev | 02:50 | |
| *** woodspa has quit IRC | 02:52 | |
| *** sleepsonthefloor has joined #openstack-dev | 02:53 | |
| *** cmark has joined #openstack-dev | 02:53 | |
| *** maheshp has joined #openstack-dev | 02:54 | |
| *** Tedster has joined #openstack-dev | 02:55 | |
| openstackgerrit | A change was merged to openstack/nova: Adds v3 API extension discovery filtering https://review.openstack.org/31953 | 02:56 |
| openstackgerrit | A change was merged to openstack/nova: port agent API into v3 part1 https://review.openstack.org/31771 | 02:57 |
| jkyle | is there an official or semi-official source for daily builds of ubuntu packages? (a ppa or something) | 03:01 |
| *** ytwu has joined #openstack-dev | 03:02 | |
| *** shang has joined #openstack-dev | 03:03 | |
| *** ytwu has quit IRC | 03:04 | |
| clarkb | jkyle: https://launchpad.net/~openstack-ubuntu-testing | 03:04 |
| jkyle | clarkb: thanks | 03:04 |
| jkyle | grizzly trunk would be 'master'? | 03:06 |
| clarkb | no havana is master | 03:08 |
| *** leizhang has joined #openstack-dev | 03:08 | |
| jkyle | k, looking up repo policy. branch -a only showed master and stable/* | 03:09 |
| jkyle | I don't see a reference to testing branches, only milestones. | 03:11 |
| *** markvoelker_ has quit IRC | 03:12 | |
| clarkb | jkyle: we do all current development on master | 03:12 |
| clarkb | jkyle: once every six months a stable/* branch is forked off of master | 03:13 |
| clarkb | stable/havana will be the next branch forked off of master so https://launchpad.net/~openstack-ubuntu-testing/+archive/havana is the PPA to use to test current master | 03:13 |
| *** litong has quit IRC | 03:15 | |
| *** timello_ has joined #openstack-dev | 03:16 | |
| jkyle | got it, so what does the grizzly trunk represent? I'd like to get at the most current stable grizzly, though testing might be sufficient. I've noticed the ubuntu cloud archive debs are not current | 03:18 |
| *** timello has quit IRC | 03:18 | |
| clarkb | grizzly is the current release. it gets periodic point releases which should end up on the cloud archive | 03:18 |
| *** stevemar has quit IRC | 03:19 | |
| *** jasdeepH has joined #openstack-dev | 03:19 | |
| *** alexpilotti has joined #openstack-dev | 03:19 | |
| clarkb | https://launchpad.net/~openstack-ubuntu-testing/+archive/grizzly-trunk-testing should be tracking the development between those point releases | 03:20 |
| jkyle | gotcha | 03:20 |
| jkyle | that ought to work. the bug I just tracked was commited to stable/grizzly on may 10th. so probably need something tracking a bit better | 03:21 |
| *** WindowsFanboi has quit IRC | 03:21 | |
| clarkb | keep in mind they are testing PPAs. I don't know how stable the grizzly test stuff is, but I know the one that tracks master can apparently be an adventure :) | 03:22 |
| *** alexpilotti has quit IRC | 03:25 | |
| jkyle | it may be more prudent to set up my own build server that tracks stable/grizzly | 03:29 |
| *** tjones has joined #openstack-dev | 03:31 | |
| *** danjared has quit IRC | 03:34 | |
| *** danjared has joined #openstack-dev | 03:34 | |
| *** neelashah has joined #openstack-dev | 03:43 | |
| *** gongysh has joined #openstack-dev | 03:51 | |
| *** martine has joined #openstack-dev | 03:58 | |
| *** nati_ueno has joined #openstack-dev | 04:04 | |
| *** nati_ueno has quit IRC | 04:05 | |
| *** nati_ueno has joined #openstack-dev | 04:05 | |
| *** jbresnah has quit IRC | 04:07 | |
| *** bdpayne has quit IRC | 04:11 | |
| *** zhiyan has joined #openstack-dev | 04:11 | |
| *** martine_ has joined #openstack-dev | 04:14 | |
| *** cliu has joined #openstack-dev | 04:14 | |
| *** Aarti has joined #openstack-dev | 04:15 | |
| *** ffio has joined #openstack-dev | 04:16 | |
| *** ffio is now known as Guest15408 | 04:17 | |
| zhiyan | hi folks, i meet a strange issue about local variable scope issue, can somebody give a hand? ? | 04:17 |
| *** prekarat has joined #openstack-dev | 04:17 | |
| *** Guest15408 has quit IRC | 04:18 | |
| lifeless | zhiyan: just ask :) | 04:18 |
| *** ffio_ has joined #openstack-dev | 04:18 | |
| *** martine has quit IRC | 04:18 | |
| *** ffio_ is now known as fire | 04:19 | |
| zhiyan | yes, i have change the 'cinder/volume/manager.py' attach_host function to give some new function. add a new argument: 'host_name'. and run some testcase, but failed. the code like this: | 04:19 |
| zhiyan | def attach_volume(self, context, volume_id, instance_uuid, host_name, | 04:19 |
| zhiyan | mountpoint): | 04:19 |
| zhiyan | """Updates db to show volume is attached""" | 04:19 |
| zhiyan | print "EXT VARS==>", locals() | 04:19 |
| zhiyan | a=1 | 04:19 |
| zhiyan | @utils.synchronized(volume_id, external=True) | 04:19 |
| zhiyan | def do_attach(): | 04:19 |
| zhiyan | print "INT VARS==>", locals() | 04:19 |
| zhiyan | when i run the test case: './run_tests.sh cinder.tests.api.contrib.test_admin_actions.AdminActionsTest.test_attach_in_used_volume_by_host' | 04:20 |
| zhiyan | i meet this situation: | 04:20 |
| zhiyan | EXT VARS==> {'instance_uuid': None, 'self': <cinder.volume.manager.VolumeManager object at 0x3c99c90>, 'host_name': 'fake_host', 'context': <cinder.context.RequestContext object at 0x3e4d7d0>, 'volume_id': 'fec306b5-cfbb-4041-b025-3b308c86e8a7', 'mountpoint': '/dev/vbd'} | 04:20 |
| zhiyan | INT VARS==> {'instance_uuid': None, 'mountpoint': '/dev/vbd', 'self': <cinder.volume.manager.VolumeManager object at 0x3c99c90>, 'context': <cinder.context.RequestContext object at 0x3e4d7d0>, 'volume_id': 'fec306b5-cfbb-4041-b025-3b308c86e8a7'} | 04:20 |
| zhiyan | so, 'host_name' argument lost in inner 'do_attach' function...it's strange to me....thanks | 04:21 |
| zhiyan | lifeless: can you give some input about this? thanks | 04:25 |
| lifeless | zhiyan: host_name isn't a local of do_attach. | 04:26 |
| *** shang has quit IRC | 04:26 | |
| *** zaitcev has quit IRC | 04:26 | |
| lifeless | zhiyan: it's in the closure, but not a local. Whats strange about that ? | 04:26 |
| *** rcleere has quit IRC | 04:27 | |
| zhiyan | lifeless: but other vars can be accessed in internal, like 'volume_id', 'mountpoint'... | 04:27 |
| zhiyan | and 'instance_uuid'.. | 04:27 |
| zhiyan | why only 'host_name' can not be | 04:27 |
| *** tjones has quit IRC | 04:27 | |
| *** afazekas has joined #openstack-dev | 04:29 | |
| *** vartom has joined #openstack-dev | 04:29 | |
| *** lloydde has joined #openstack-dev | 04:31 | |
| *** Mandell has joined #openstack-dev | 04:31 | |
| *** Hien has quit IRC | 04:32 | |
| *** tjones has joined #openstack-dev | 04:36 | |
| openstackgerrit | A change was merged to openstack/cinder: Update kombu requirement https://review.openstack.org/33717 | 04:39 |
| openstackgerrit | A change was merged to openstack/swift: Local write affinity for object PUT requests. https://review.openstack.org/33333 | 04:39 |
| *** tjones has left #openstack-dev | 04:42 | |
| *** Guest99955 has joined #openstack-dev | 04:44 | |
| *** benner has quit IRC | 04:44 | |
| *** martine_ has quit IRC | 04:44 | |
| *** martine has joined #openstack-dev | 04:45 | |
| *** maheshp has quit IRC | 04:46 | |
| *** Guest99955 has quit IRC | 04:47 | |
| *** Hien has joined #openstack-dev | 04:47 | |
| *** benner has joined #openstack-dev | 04:50 | |
| *** vartom has quit IRC | 04:50 | |
| *** koolhead11|away has joined #openstack-dev | 04:51 | |
| *** shang has joined #openstack-dev | 04:52 | |
| *** nati_uen_ has joined #openstack-dev | 04:53 | |
| *** nati_uen_ has joined #openstack-dev | 04:54 | |
| *** nati_ueno has quit IRC | 04:56 | |
| *** koolhead11|away is now known as koolhead17 | 04:57 | |
| *** kaushikc has joined #openstack-dev | 04:58 | |
| *** Hien has quit IRC | 04:59 | |
| *** fire has quit IRC | 04:59 | |
| *** maheshp has joined #openstack-dev | 04:59 | |
| *** noslzzp has quit IRC | 05:00 | |
| *** ffio_ has joined #openstack-dev | 05:01 | |
| *** ffio_ is now known as fire | 05:01 | |
| *** Hien has joined #openstack-dev | 05:01 | |
| *** colinmcnamara has joined #openstack-dev | 05:01 | |
| *** neelashah has quit IRC | 05:02 | |
| *** jasdeepH has quit IRC | 05:04 | |
| *** abhishekkr_ has joined #openstack-dev | 05:07 | |
| *** alexxu has joined #openstack-dev | 05:10 | |
| *** SergeyLukjanov has joined #openstack-dev | 05:11 | |
| *** gongysh has quit IRC | 05:13 | |
| *** maheshp has quit IRC | 05:13 | |
| *** jasdeepH has joined #openstack-dev | 05:13 | |
| *** cmark has quit IRC | 05:15 | |
| *** martine has quit IRC | 05:18 | |
| lifeless | zhiyan: it should be accessible, but not via locals() | 05:24 |
| *** anfrolov has quit IRC | 05:25 | |
| *** colinmcnamara has quit IRC | 05:26 | |
| *** anfrolov has joined #openstack-dev | 05:27 | |
| *** cdub_ has quit IRC | 05:28 | |
| *** maheshp has joined #openstack-dev | 05:30 | |
| zhiyan | lifeless: thanks for you response. i just resolved this issue :) . also, locals() can access it ('host_name'). | 05:31 |
| *** fire has quit IRC | 05:32 | |
| zhiyan | this situation seems because some python feature (py27 i used) | 05:35 |
| *** kaushikc has quit IRC | 05:41 | |
| *** lloydde has quit IRC | 05:41 | |
| *** kaushikc has joined #openstack-dev | 05:41 | |
| *** giroro_ has quit IRC | 05:42 | |
| *** Ruetobas has joined #openstack-dev | 05:44 | |
| *** Ruetobas has joined #openstack-dev | 05:47 | |
| *** nati_ueno has joined #openstack-dev | 05:48 | |
| *** vartom has joined #openstack-dev | 05:48 | |
| *** nati_ueno has quit IRC | 05:49 | |
| *** nati_ueno has joined #openstack-dev | 05:51 | |
| *** nati_uen_ has quit IRC | 05:51 | |
| *** zhuadl has joined #openstack-dev | 05:51 | |
| *** devvesa has joined #openstack-dev | 05:52 | |
| *** nati_ueno has quit IRC | 05:52 | |
| *** jcoufal has joined #openstack-dev | 05:52 | |
| *** nati_ueno has joined #openstack-dev | 05:53 | |
| *** vartom has quit IRC | 05:53 | |
| *** zhuadl has quit IRC | 05:56 | |
| *** anfrolov has quit IRC | 05:59 | |
| *** jasdeepH has quit IRC | 06:01 | |
| *** jasdeepH has joined #openstack-dev | 06:01 | |
| *** lloydde has joined #openstack-dev | 06:02 | |
| *** zhuadl has joined #openstack-dev | 06:04 | |
| *** SergeyLukjanov has quit IRC | 06:04 | |
| *** kushal has joined #openstack-dev | 06:05 | |
| *** kushal has quit IRC | 06:08 | |
| *** cmark has joined #openstack-dev | 06:09 | |
| *** kushal has joined #openstack-dev | 06:09 | |
| *** cmark has joined #openstack-dev | 06:09 | |
| *** devvesa has quit IRC | 06:10 | |
| *** devvesa has joined #openstack-dev | 06:16 | |
| *** lsmola has joined #openstack-dev | 06:19 | |
| yanglyy | Do you know, are there any document links of heat api? | 06:21 |
| *** bashok has joined #openstack-dev | 06:23 | |
| *** aeperezt has quit IRC | 06:23 | |
| *** xqueralt-afk is now known as xqueralt | 06:24 | |
| *** maheshp has quit IRC | 06:25 | |
| *** cmark has quit IRC | 06:28 | |
| *** psedlak has joined #openstack-dev | 06:28 | |
| *** xga has joined #openstack-dev | 06:30 | |
| *** xga_ has joined #openstack-dev | 06:30 | |
| openstackgerrit | A change was merged to openstack/nova: Port quota API into v3 part2 https://review.openstack.org/34401 | 06:30 |
| *** egallen has joined #openstack-dev | 06:31 | |
| *** xga__ has joined #openstack-dev | 06:32 | |
| *** navid_ has joined #openstack-dev | 06:33 | |
| *** xga has quit IRC | 06:34 | |
| *** xga_ has quit IRC | 06:34 | |
| *** xga__ has quit IRC | 06:37 | |
| openstackgerrit | A change was merged to openstack/swift: Deleted account respond as non existing accounts https://review.openstack.org/32566 | 06:38 |
| *** yanglyy has quit IRC | 06:39 | |
| *** maheshp has joined #openstack-dev | 06:40 | |
| *** xga has joined #openstack-dev | 06:41 | |
| *** yanglyy has joined #openstack-dev | 06:42 | |
| openstackgerrit | A change was merged to openstack/swift: updated changelog and authors update for 1.9.0 release https://review.openstack.org/34496 | 06:43 |
| *** xga has quit IRC | 06:45 | |
| openstackgerrit | A change was merged to openstack/swift: version bump to 1.9.1 https://review.openstack.org/34497 | 06:46 |
| *** eglynn has joined #openstack-dev | 06:46 | |
| ianw | zhiyan: it's because the variable isn't bound. printing it or something like that would make it turn up in locals | 06:47 |
| *** xga has joined #openstack-dev | 06:51 | |
| *** anfrolov has joined #openstack-dev | 06:51 | |
| *** kpavel has joined #openstack-dev | 06:52 | |
| *** odyssey4me has joined #openstack-dev | 06:53 | |
| *** Ryan_Lane has quit IRC | 06:54 | |
| *** eglynn has quit IRC | 06:54 | |
| *** Ryan_Lane has joined #openstack-dev | 06:54 | |
| *** vsergeyev has quit IRC | 06:55 | |
| zhiyan | ianw: thanks for you replay. but you see the 'host_name' not represent in the internal print output. | 06:55 |
| *** abhishekkr_ has quit IRC | 06:57 | |
| *** torandu has joined #openstack-dev | 06:58 | |
| *** yolanda has joined #openstack-dev | 06:58 | |
| *** reidrac has joined #openstack-dev | 07:02 | |
| *** tkammer_ has joined #openstack-dev | 07:02 | |
| *** zoresvit has joined #openstack-dev | 07:07 | |
| *** mmagr has joined #openstack-dev | 07:08 | |
| *** xga has quit IRC | 07:08 | |
| *** timello_ has quit IRC | 07:09 | |
| *** timello_ has joined #openstack-dev | 07:10 | |
| *** gongysh has joined #openstack-dev | 07:12 | |
| *** cliu has quit IRC | 07:13 | |
| *** jtomasek has joined #openstack-dev | 07:13 | |
| gongysh | garyk: ping | 07:13 |
| *** abhishekkr_ has joined #openstack-dev | 07:14 | |
| *** pnavarro has joined #openstack-dev | 07:15 | |
| *** xga has joined #openstack-dev | 07:16 | |
| *** egallen has quit IRC | 07:17 | |
| *** ffio has joined #openstack-dev | 07:18 | |
| *** egallen has joined #openstack-dev | 07:18 | |
| *** lloydde has quit IRC | 07:18 | |
| *** lloydde has joined #openstack-dev | 07:19 | |
| *** READ10 has joined #openstack-dev | 07:19 | |
| *** mrunge has joined #openstack-dev | 07:20 | |
| *** xga__ has joined #openstack-dev | 07:20 | |
| *** xga_ has joined #openstack-dev | 07:20 | |
| *** pasquier-s has joined #openstack-dev | 07:21 | |
| *** xga has quit IRC | 07:22 | |
| *** Alexei_987 has joined #openstack-dev | 07:23 | |
| *** koolhead17 has quit IRC | 07:23 | |
| *** sushils has joined #openstack-dev | 07:23 | |
| *** lloydde has quit IRC | 07:24 | |
| *** freedomhui has joined #openstack-dev | 07:24 | |
| *** ffio has quit IRC | 07:25 | |
| *** corXi has joined #openstack-dev | 07:27 | |
| *** kaushikc has quit IRC | 07:27 | |
| *** zhuadl has quit IRC | 07:29 | |
| *** kaushikc has joined #openstack-dev | 07:29 | |
| *** sthaha has quit IRC | 07:34 | |
| *** JordanP has joined #openstack-dev | 07:36 | |
| openstackgerrit | A change was merged to openstack/quantum: Update to use OSLO db https://review.openstack.org/27265 | 07:36 |
| *** zhuadl has joined #openstack-dev | 07:37 | |
| *** Ryan_Lane has quit IRC | 07:37 | |
| *** jgallard has joined #openstack-dev | 07:43 | |
| *** Ryan_Lane has joined #openstack-dev | 07:44 | |
| *** maheshp has quit IRC | 07:44 | |
| *** iartarisi has joined #openstack-dev | 07:45 | |
| *** JordanP has quit IRC | 07:47 | |
| *** JordanP has joined #openstack-dev | 07:48 | |
| *** lloydde has joined #openstack-dev | 07:49 | |
| *** jgallard has quit IRC | 07:51 | |
| *** jgallard has joined #openstack-dev | 07:52 | |
| *** Ryan_Lane has quit IRC | 07:53 | |
| *** Ryan_Lane has joined #openstack-dev | 07:54 | |
| *** johnthetubaguy has joined #openstack-dev | 07:54 | |
| enikanorov | wow. oslo db landed into quantum | 07:57 |
| *** lloydde has quit IRC | 07:58 | |
| *** erkules_ is now known as erkules | 08:02 | |
| *** Aarti has quit IRC | 08:04 | |
| *** huats has joined #openstack-dev | 08:05 | |
| *** huats has quit IRC | 08:05 | |
| *** huats has joined #openstack-dev | 08:05 | |
| *** maheshp has joined #openstack-dev | 08:06 | |
| *** xga__ has quit IRC | 08:08 | |
| *** xga_ has quit IRC | 08:08 | |
| *** eglynn has joined #openstack-dev | 08:08 | |
| *** Mandell has quit IRC | 08:09 | |
| *** bashok has quit IRC | 08:10 | |
| *** bashok has joined #openstack-dev | 08:10 | |
| *** negronjl has quit IRC | 08:11 | |
| *** athomas has joined #openstack-dev | 08:11 | |
| *** lucasagomes has joined #openstack-dev | 08:12 | |
| *** huats has quit IRC | 08:12 | |
| *** huats has joined #openstack-dev | 08:13 | |
| *** huats has joined #openstack-dev | 08:13 | |
| *** Ryan_Lane has quit IRC | 08:13 | |
| *** xga has joined #openstack-dev | 08:16 | |
| *** xga_ has joined #openstack-dev | 08:16 | |
| *** Ryan_Lane has joined #openstack-dev | 08:16 | |
| *** abhishekkr__ has joined #openstack-dev | 08:17 | |
| *** derekh has joined #openstack-dev | 08:17 | |
| *** nati_ueno has quit IRC | 08:17 | |
| *** pixelbeat has joined #openstack-dev | 08:17 | |
| *** abhishekkr_ has quit IRC | 08:20 | |
| *** ffio has joined #openstack-dev | 08:22 | |
| *** afazekas_ has joined #openstack-dev | 08:23 | |
| *** afazekas_ has quit IRC | 08:23 | |
| *** afazekas_ has joined #openstack-dev | 08:24 | |
| *** kaushikc has quit IRC | 08:24 | |
| *** nati_ueno has joined #openstack-dev | 08:24 | |
| *** reed has joined #openstack-dev | 08:25 | |
| *** Ryan_Lane has quit IRC | 08:26 | |
| *** jclift has quit IRC | 08:29 | |
| *** negronjl has joined #openstack-dev | 08:31 | |
| *** jpich has joined #openstack-dev | 08:31 | |
| *** bswartz has quit IRC | 08:32 | |
| *** martyntaylor has joined #openstack-dev | 08:37 | |
| *** SergeyLukjanov has joined #openstack-dev | 08:49 | |
| *** jgallard has quit IRC | 08:49 | |
| *** jgallard has joined #openstack-dev | 08:49 | |
| *** lloydde has joined #openstack-dev | 08:49 | |
| *** salv-orlando_ has joined #openstack-dev | 08:54 | |
| *** salv-orlando has quit IRC | 08:54 | |
| *** salv-orlando_ is now known as salv-orlando | 08:54 | |
| *** lloydde has quit IRC | 08:54 | |
| *** mikal has quit IRC | 08:57 | |
| *** zodiak has quit IRC | 08:58 | |
| *** bashok has quit IRC | 08:58 | |
| *** mikal has joined #openstack-dev | 08:58 | |
| *** mkollaro has joined #openstack-dev | 09:02 | |
| *** bashok has joined #openstack-dev | 09:03 | |
| *** vartom has joined #openstack-dev | 09:04 | |
| *** romcheg has joined #openstack-dev | 09:06 | |
| *** WindowsFanboi has joined #openstack-dev | 09:07 | |
| *** bashok has quit IRC | 09:08 | |
| *** johnthetubaguy has quit IRC | 09:08 | |
| *** johnthetubaguy has joined #openstack-dev | 09:08 | |
| *** vartom has quit IRC | 09:09 | |
| *** bashok has joined #openstack-dev | 09:11 | |
| *** danpb has joined #openstack-dev | 09:11 | |
| *** xga has quit IRC | 09:16 | |
| *** xga_ has quit IRC | 09:16 | |
| *** nati_ueno has quit IRC | 09:17 | |
| *** mrunge has quit IRC | 09:21 | |
| *** romcheg has left #openstack-dev | 09:23 | |
| *** bing_bu has quit IRC | 09:24 | |
| *** athomas has quit IRC | 09:26 | |
| *** mrunge_ has joined #openstack-dev | 09:28 | |
| *** navid_ has quit IRC | 09:31 | |
| agorodnev | lifeless: hi! =) I was trying to use MonkeyPatch from fixtures instead of ReplaceModule from my patch, but didn't have luck. It seems that MonkeyPatch doesn't work for modules. Any ideas? | 09:31 |
| lifeless | agorodnev: well, the main point was that you don't need to couple your thing to the test case. | 09:32 |
| lifeless | agorodnev: there are several ways you could do that. | 09:32 |
| *** shang has quit IRC | 09:32 | |
| *** mrunge_ has quit IRC | 09:34 | |
| *** mrunge_ has joined #openstack-dev | 09:34 | |
| agorodnev | lifeless: ok =) I separated all that things from the testcase class. And I've read your article. Now I try to create more "concentrated" version of the patch | 09:36 |
| openstackgerrit | A change was merged to openstack/horizon: Fix Further Pyflakes warnings https://review.openstack.org/31383 | 09:36 |
| *** athomas has joined #openstack-dev | 09:37 | |
| openstackgerrit | A change was merged to openstack/horizon: python3: Introduce py33 to tox.ini https://review.openstack.org/31398 | 09:37 |
| *** ffio has quit IRC | 09:38 | |
| *** boris-42 has quit IRC | 09:39 | |
| *** navid_ has joined #openstack-dev | 09:39 | |
| *** mihgen has joined #openstack-dev | 09:40 | |
| *** maheshp has quit IRC | 09:41 | |
| *** zhiyan has left #openstack-dev | 09:41 | |
| *** maheshp has joined #openstack-dev | 09:41 | |
| *** jgallard has quit IRC | 09:41 | |
| *** jgallard has joined #openstack-dev | 09:42 | |
| *** jgallard has quit IRC | 09:42 | |
| *** mrunge_ has quit IRC | 09:43 | |
| *** jgallard has joined #openstack-dev | 09:43 | |
| *** drusskikh has quit IRC | 09:43 | |
| *** lloydde has joined #openstack-dev | 09:50 | |
| *** alexisT has joined #openstack-dev | 09:51 | |
| *** dabo has joined #openstack-dev | 09:51 | |
| *** crazed has joined #openstack-dev | 09:51 | |
| *** briancline has joined #openstack-dev | 09:51 | |
| *** devx has joined #openstack-dev | 09:51 | |
| *** jjmb has joined #openstack-dev | 09:51 | |
| *** cyeoh has joined #openstack-dev | 09:52 | |
| *** lloydde has quit IRC | 09:55 | |
| *** READ10 has quit IRC | 09:55 | |
| *** READ10 has joined #openstack-dev | 09:59 | |
| *** yanglyy has quit IRC | 09:59 | |
| *** maheshp has quit IRC | 10:01 | |
| *** WindowsFanboi has quit IRC | 10:03 | |
| *** maheshp has joined #openstack-dev | 10:03 | |
| *** pcm___ has joined #openstack-dev | 10:04 | |
| *** pcm___ has quit IRC | 10:04 | |
| *** psedlak has quit IRC | 10:04 | |
| *** pcm__ has joined #openstack-dev | 10:05 | |
| *** markmc has joined #openstack-dev | 10:08 | |
| *** gmoro_ has joined #openstack-dev | 10:10 | |
| *** drusskikh_ has quit IRC | 10:13 | |
| *** mihgen has quit IRC | 10:13 | |
| *** fifieldt has joined #openstack-dev | 10:14 | |
| *** maheshp1 has joined #openstack-dev | 10:15 | |
| *** maheshp has quit IRC | 10:16 | |
| *** freedomhui has quit IRC | 10:17 | |
| *** shang has joined #openstack-dev | 10:18 | |
| *** martyntaylor has quit IRC | 10:20 | |
| *** yaguang has quit IRC | 10:21 | |
| *** xga_ has joined #openstack-dev | 10:25 | |
| *** xga has joined #openstack-dev | 10:25 | |
| *** romcheg1 has joined #openstack-dev | 10:26 | |
| *** zul has quit IRC | 10:27 | |
| *** zul has joined #openstack-dev | 10:29 | |
| *** maheshp1 has quit IRC | 10:32 | |
| *** psedlak has joined #openstack-dev | 10:33 | |
| *** leizhang has quit IRC | 10:42 | |
| *** martyntaylor has joined #openstack-dev | 10:48 | |
| *** gongysh has quit IRC | 10:50 | |
| *** lloydde has joined #openstack-dev | 10:50 | |
| *** martyntaylor has quit IRC | 10:52 | |
| *** martyntaylor has joined #openstack-dev | 10:53 | |
| *** jgallard has quit IRC | 10:53 | |
| *** portante_ is now known as portante | 10:53 | |
| *** rpodolyaka has joined #openstack-dev | 10:54 | |
| *** maheshp has joined #openstack-dev | 10:55 | |
| *** lloydde has quit IRC | 10:55 | |
| *** ffio has joined #openstack-dev | 10:58 | |
| *** timello_ has quit IRC | 10:58 | |
| *** timello_ has joined #openstack-dev | 10:59 | |
| *** pnavarro has quit IRC | 10:59 | |
| *** kaushikc has joined #openstack-dev | 11:01 | |
| *** sungju_ has joined #openstack-dev | 11:02 | |
| *** sungju_ has quit IRC | 11:02 | |
| *** ndipanov has quit IRC | 11:03 | |
| *** anfrolov has quit IRC | 11:09 | |
| *** anfrolov has joined #openstack-dev | 11:14 | |
| *** jruzicka has joined #openstack-dev | 11:14 | |
| *** zhuadl has quit IRC | 11:14 | |
| *** ifarkas has quit IRC | 11:16 | |
| *** ndipanov has joined #openstack-dev | 11:16 | |
| *** ifarkas has joined #openstack-dev | 11:16 | |
| *** bashok has quit IRC | 11:18 | |
| *** dkranz has quit IRC | 11:18 | |
| *** fifieldt has quit IRC | 11:19 | |
| *** HenryG has joined #openstack-dev | 11:20 | |
| *** prekarat has quit IRC | 11:20 | |
| *** romcheg has joined #openstack-dev | 11:21 | |
| *** boris-42 has joined #openstack-dev | 11:23 | |
| *** kaushikc has quit IRC | 11:24 | |
| *** romcheg1 has quit IRC | 11:25 | |
| *** sdake has joined #openstack-dev | 11:26 | |
| *** sdake has joined #openstack-dev | 11:26 | |
| *** maheshp has quit IRC | 11:28 | |
| *** maheshp has joined #openstack-dev | 11:28 | |
| *** jjmb has quit IRC | 11:29 | |
| *** rscottcoyle has joined #openstack-dev | 11:30 | |
| *** garyk has joined #openstack-dev | 11:30 | |
| *** abhishekkr__ has quit IRC | 11:31 | |
| *** jclift has joined #openstack-dev | 11:38 | |
| openstackgerrit | A change was merged to openstack/oslo-incubator: Make AMQP based RPC consumer threads more robust https://review.openstack.org/32235 | 11:40 |
| *** abhishekkr__ has joined #openstack-dev | 11:40 | |
| *** eglynn has quit IRC | 11:40 | |
| *** eglynn has joined #openstack-dev | 11:41 | |
| openstackgerrit | A change was merged to openstack/oslo-incubator: Don't reconnect to exclusive fanout consumers. https://review.openstack.org/31718 | 11:41 |
| *** vkmc has joined #openstack-dev | 11:41 | |
| *** vkmc has quit IRC | 11:41 | |
| *** vkmc has joined #openstack-dev | 11:41 | |
| openstackgerrit | A change was merged to openstack/oslo-incubator: Add network_utils.urlsplit https://review.openstack.org/33634 | 11:42 |
| *** vkmc has quit IRC | 11:44 | |
| *** vkmc has joined #openstack-dev | 11:45 | |
| *** vkmc has quit IRC | 11:45 | |
| *** vkmc has joined #openstack-dev | 11:45 | |
| *** dkranz has joined #openstack-dev | 11:49 | |
| *** lloydde has joined #openstack-dev | 11:51 | |
| *** rnirmal has joined #openstack-dev | 11:51 | |
| openstackgerrit | A change was merged to openstack/oslo-incubator: Don't use mixture of cfg.Opt() deprecated args https://review.openstack.org/34259 | 11:54 |
| *** tkammer_ has quit IRC | 11:55 | |
| *** jcoufal has quit IRC | 11:55 | |
| *** afazekas_ has quit IRC | 11:55 | |
| *** martine has joined #openstack-dev | 11:56 | |
| *** noslzzp has joined #openstack-dev | 11:59 | |
| *** eharney has joined #openstack-dev | 11:59 | |
| *** eharney has quit IRC | 11:59 | |
| *** eharney has joined #openstack-dev | 11:59 | |
| *** flaper87 has joined #openstack-dev | 11:59 | |
| *** drewlander has joined #openstack-dev | 12:01 | |
| *** psedlak_ has joined #openstack-dev | 12:01 | |
| *** mkollaro has quit IRC | 12:02 | |
| *** jcoufal has joined #openstack-dev | 12:02 | |
| *** mkollaro has joined #openstack-dev | 12:02 | |
| *** psedlak has quit IRC | 12:02 | |
| *** dprince has joined #openstack-dev | 12:05 | |
| *** lloydde has quit IRC | 12:05 | |
| *** freedomhui has joined #openstack-dev | 12:06 | |
| *** SergeyLukjanov has quit IRC | 12:07 | |
| *** lucasagomes is now known as lucas-hungry | 12:08 | |
| *** CaptTofu_ has quit IRC | 12:10 | |
| *** CaptTofu has joined #openstack-dev | 12:11 | |
| *** CaptTofu has quit IRC | 12:12 | |
| *** CaptTofu has joined #openstack-dev | 12:12 | |
| *** maheshp has quit IRC | 12:13 | |
| *** afazekas_ has joined #openstack-dev | 12:13 | |
| *** ndipanov has quit IRC | 12:13 | |
| *** maheshp has joined #openstack-dev | 12:14 | |
| zoresvit | Hi. According to https://wiki.openstack.org/wiki/UnifiedCLI/Mapping openstackclient should already support "user role list" command, but it's not: ERROR: cliff.app Unknown command ['user', 'role', 'add'] | 12:14 |
| zoresvit | Is the doc outdated? | 12:14 |
| *** galstrom_zzz is now known as galstrom | 12:15 | |
| *** SergeyLukjanov has joined #openstack-dev | 12:17 | |
| rpodolyaka | zoresvit: openstackclient is so mysterious that it's hard find the answer for your question even on openstack-dev channel :) | 12:17 |
| *** rpodolyaka has quit IRC | 12:20 | |
| *** tkammer_ has joined #openstack-dev | 12:21 | |
| *** afazekas_ has quit IRC | 12:21 | |
| *** rpodolyaka has joined #openstack-dev | 12:23 | |
| *** maheshp has quit IRC | 12:24 | |
| garyk | markmc: i was chatting with HenryG and a question arose regarding the lowercase section names. would puppet, chef etc support this? | 12:24 |
| *** ndipanov has joined #openstack-dev | 12:25 | |
| *** maheshp has joined #openstack-dev | 12:28 | |
| *** freedomhui has quit IRC | 12:29 | |
| *** maheshp has quit IRC | 12:29 | |
| *** morazi_ has quit IRC | 12:29 | |
| *** maheshp has joined #openstack-dev | 12:30 | |
| *** gordc has quit IRC | 12:33 | |
| *** zhuadl has joined #openstack-dev | 12:34 | |
| JordanP | sdague, hi. A moment about the id vs UUID in tempest/quantum ? In fact the check that validate the identifier (whether it is a uuid or an id) is done at nova-api side (see api/OS/compute/contrib/security_groups*, which calls validate_id()). Quantum (nor the quantum-client) is not implied at all in the validation. It's just that the implementation of validate_id (in nova/network/security_group/quantum_drive.py for Quantum and nova/co | 12:37 |
| JordanP | mpute/api.py for Nova-Net) is different | 12:37 |
| JordanP | so 2 weeks ago, you recommended to do some sort of exception translation not to raise the default/raw exception raised by quantum-client. This recommendation doesn't stand, i think | 12:39 |
| *** alexxu has quit IRC | 12:39 | |
| markmc | garyk, that's a very broad question | 12:39 |
| markmc | garyk, the case we found yesterday was where a default value using the new name was taking precedence over a user-provided value using the deprecated name | 12:40 |
| markmc | garyk, the fix for that is to make sure default values are commented out in default config files | 12:40 |
| markmc | garyk, lowercase vs uppercase should work fine, it's actually more about the deprecated option names | 12:41 |
| *** vsergeyev has joined #openstack-dev | 12:41 | |
| garyk | markmc: understood. gongysh is already in that task. thanks for the clarifications. | 12:41 |
| *** pnavarro has joined #openstack-dev | 12:42 | |
| *** galstrom is now known as galstrom_zzz | 12:42 | |
| *** dhellmann-away is now known as dhellmann | 12:42 | |
| HenryG | markmc: Thanks, looks like I should be fine then | 12:42 |
| *** lbragstad has quit IRC | 12:43 | |
| *** sandywalsh has quit IRC | 12:45 | |
| *** SergeyLukjanov has quit IRC | 12:48 | |
| *** leizhang has joined #openstack-dev | 12:49 | |
| openstackgerrit | A change was merged to openstack/tempest: Remove dependency on MySQL-python. https://review.openstack.org/34152 | 12:51 |
| openstackgerrit | A change was merged to openstack/glance: Do not raise NEW exceptions https://review.openstack.org/33267 | 12:51 |
| *** noslzzp has quit IRC | 12:51 | |
| *** lloydde has joined #openstack-dev | 12:51 | |
| *** noslzzp has joined #openstack-dev | 12:56 | |
| *** lloydde has quit IRC | 12:56 | |
| *** jjmb has joined #openstack-dev | 12:58 | |
| *** gordc has joined #openstack-dev | 12:58 | |
| *** bing_bu has joined #openstack-dev | 12:58 | |
| *** kaushikc has joined #openstack-dev | 12:59 | |
| *** sandywalsh has joined #openstack-dev | 12:59 | |
| *** portante has quit IRC | 12:59 | |
| *** kaushikc has quit IRC | 13:00 | |
| sandywalsh | markmc, what's the scoop on oslo-config 1.2 ... I was hoping to merge oslo-common into ceilometer, but there's a 1.2 dependency in master it seems. | 13:00 |
| *** kaushikc has joined #openstack-dev | 13:00 | |
| *** redbeard2 has quit IRC | 13:00 | |
| *** msmedved has joined #openstack-dev | 13:01 | |
| *** psedlak__ has joined #openstack-dev | 13:02 | |
| *** gargya has joined #openstack-dev | 13:02 | |
| *** bknudson has joined #openstack-dev | 13:03 | |
| *** litong has joined #openstack-dev | 13:03 | |
| *** psedlak_ has quit IRC | 13:03 | |
| *** mkollaro has quit IRC | 13:03 | |
| *** martine_ has joined #openstack-dev | 13:03 | |
| *** martine has quit IRC | 13:04 | |
| *** bdpayne has joined #openstack-dev | 13:05 | |
| *** tzumainn has joined #openstack-dev | 13:05 | |
| *** gongysh has joined #openstack-dev | 13:06 | |
| *** jtomasek has quit IRC | 13:06 | |
| *** jtomasek has joined #openstack-dev | 13:07 | |
| *** jergerber has joined #openstack-dev | 13:08 | |
| Alexei_987 | russellb: Hello, is there any list of extensions that should be ported to nova-v3-api? I would like to choose one to work on it. | 13:08 |
| *** SergeyLukjanov has joined #openstack-dev | 13:09 | |
| *** maheshp has quit IRC | 13:09 | |
| *** rharwood has joined #openstack-dev | 13:10 | |
| *** ayoung has joined #openstack-dev | 13:11 | |
| *** lbragstad has joined #openstack-dev | 13:11 | |
| *** krtaylor has quit IRC | 13:12 | |
| openstackgerrit | A change was merged to openstack/tempest: Revert "Update a test to work with Quantum SecGroup" https://review.openstack.org/34428 | 13:13 |
| openstackgerrit | A change was merged to openstack/glance: Import sql_connection option before using it. https://review.openstack.org/33721 | 13:13 |
| openstackgerrit | A change was merged to openstack/tempest: Skip list_vifs test with Quantum https://review.openstack.org/34439 | 13:13 |
| markmc | sandywalsh, https://bugs.launchpad.net/oslo/+bug/1189889 should give you some pointers | 13:13 |
| *** Aarti has joined #openstack-dev | 13:13 | |
| *** kbringard has joined #openstack-dev | 13:14 | |
| *** jonesld has joined #openstack-dev | 13:14 | |
| markmc | sandywalsh, basically, include the full url in requirements.txt and make sure to require latest pbr | 13:14 |
| *** olaph has joined #openstack-dev | 13:14 | |
| *** prad has joined #openstack-dev | 13:15 | |
| sandywalsh | markmc, looking ... thanks | 13:16 |
| *** egallen has quit IRC | 13:16 | |
| sdague | JordanP: as long as nova returns the same status code for the same request regardless of whether nova-network or quantum is the backend, I don't really care what the intermediate implementation is. The important thing is the surface doesn't change when the backend changes. | 13:16 |
| *** cp16net is now known as cp16net|away | 13:16 | |
| *** egallen has joined #openstack-dev | 13:16 | |
| *** egallen has quit IRC | 13:17 | |
| *** jjmb has quit IRC | 13:17 | |
| *** olaph_ has quit IRC | 13:18 | |
| *** jayg|g0n3 is now known as jayg | 13:19 | |
| JordanP | sdague, so I think the only solution is to remove the validation, don't check whether the id (be it a regular ID or a uuid) is valid, because the moment you try to validate, as the validation method is different, you introduce a difference based on the backend | 13:19 |
| *** gargya has quit IRC | 13:19 | |
| *** gargya has joined #openstack-dev | 13:20 | |
| *** joesavak has joined #openstack-dev | 13:20 | |
| *** dkranz has quit IRC | 13:21 | |
| sdague | JordanP: there is a difference between validating, and validating and returning the validation error result to the user. In nova net we're just returning 404 in an invalid id, right? | 13:22 |
| JordanP | no | 13:22 |
| sdague | we do this in the db layer already | 13:22 |
| JordanP | qe are returning BadRquest | 13:23 |
| JordanP | no it is done before the DB is even hit | 13:23 |
| sdague | JordanP: link? | 13:23 |
| JordanP | a sec | 13:23 |
| sdague | JordanP: no, I don't mean for this case, I mean in other cases | 13:23 |
| sdague | where pg and mysql act differently on validation | 13:23 |
| *** lucas-hungry is now known as lucasagomes | 13:24 | |
| *** afazekas_ has joined #openstack-dev | 13:25 | |
| *** rpodolyaka has left #openstack-dev | 13:25 | |
| JordanP | i am not aware of what is done in for mysql vs db. In my case (uuid vs id) validation is done by nova-api before the call to the backend (nova net vs quantum) is even done : https://github.com/openstack/nova/blob/master/nova/api/openstack/compute/contrib/security_groups.py#L258 | 13:26 |
| *** tkammer_ has quit IRC | 13:26 | |
| sdague | JordanP: ok, I have to go run to some meetings for the next 1.5 hrs, I'll look as soon as I get out | 13:27 |
| *** abhishekkr__ has quit IRC | 13:27 | |
| JordanP | ok thanks | 13:27 |
| *** AGiardini has joined #openstack-dev | 13:28 | |
| *** dolphm has joined #openstack-dev | 13:28 | |
| *** AGiardini has quit IRC | 13:28 | |
| *** neelashah has joined #openstack-dev | 13:28 | |
| *** tkammer has joined #openstack-dev | 13:29 | |
| openstackgerrit | A change was merged to openstack/tempest: Use os.path.join to form the cli command https://review.openstack.org/34265 | 13:30 |
| *** dhellmann is now known as dhellmann-away | 13:31 | |
| *** yaguang has joined #openstack-dev | 13:31 | |
| *** yaguang has quit IRC | 13:34 | |
| zoresvit | Alright, is there at least a way to validate policy.json file for keystone, anyone? | 13:34 |
| *** afazekas_ has quit IRC | 13:34 | |
| dolphm | zoresvit: validate against what? | 13:35 |
| openstackgerrit | A change was merged to openstack/tempest: Skip test that is not relevant to Quantum https://review.openstack.org/34406 | 13:35 |
| *** mkollaro has joined #openstack-dev | 13:35 | |
| zoresvit | dolphm: validate that for example this syntax is meaningful for Keystone: | 13:36 |
| zoresvit | "admin_required": "(role:tenant_admin and tenant_id:%(target_tenant_id)s)" | 13:36 |
| zoresvit | Because iven if I put an invalid json into policy.json file, I can't see any complains from Keystone whatsoever. So just wondered if there are any checks made on that. | 13:36 |
| *** dkranz has joined #openstack-dev | 13:36 | |
| dolphm | zoresvit: that should cause a backtrace as soon as it's parsed | 13:37 |
| openstackgerrit | A change was merged to openstack/tempest: Skip fixed_ip_details test with Quantum https://review.openstack.org/34438 | 13:38 |
| openstackgerrit | A change was merged to openstack/tempest: CLI doc missing colon for code block https://review.openstack.org/34311 | 13:40 |
| bknudson | ayoung dolphm: can I get a review of https://review.openstack.org/#/c/34475/ -- it should be quick because it re-does a fix we had earlier. | 13:40 |
| dolphm | bknudson: looking | 13:41 |
| *** morazi has joined #openstack-dev | 13:41 | |
| zoresvit | dolphm: Is that correct that to change policy, I modify /etc/keystone/policy.json file and than restart Keystone? | 13:41 |
| bknudson | zoresvit: you shouldn't have to restart Keystone | 13:41 |
| dolphm | bknudson: does it re-read the policy file per request? | 13:42 |
| bknudson | dolphm: that's what I think happens. | 13:42 |
| bknudson | I'll try it. | 13:42 |
| ayoung | bknudson, you move the code around, into a different function. Why is that? | 13:43 |
| zoresvit | dolphm: okay, so it's read each time some action is performed, not on service startup. Got it, thanks :) | 13:44 |
| bknudson | the code in original at 872 didn't do anything. users.append(user_dn) couldn't raise UserNotFound | 13:44 |
| bknudson | it's just adding a string to a list. | 13:44 |
| *** jecarey_ has joined #openstack-dev | 13:44 | |
| bknudson | ayoung: It's "self.user.get(user_id)" at line 373 in original that could raise UserNotFound | 13:45 |
| ayoung | bknudson, ah, so users.append(user_dn) did nothing | 13:46 |
| *** noslzzp has quit IRC | 13:48 | |
| bknudson | ayoung: it doesn't check if the user exists, that's done when users.append(self.user.get(user_id)) | 13:48 |
| *** stevemar has joined #openstack-dev | 13:48 | |
| garyk | gongysh: ping | 13:48 |
| *** andrewbogott_afk is now known as andrewbogott | 13:49 | |
| *** jjmb has joined #openstack-dev | 13:49 | |
| *** kevinconway has left #openstack-dev | 13:50 | |
| *** portante has joined #openstack-dev | 13:50 | |
| ayoung | dolphm, one issue from last nights release meeting you should be aware of: | 13:50 |
| ayoung | 21:28:04 <ayoung> but are we OK with saying that all of the current clients change over to using keystoneclient as a dependency for auth? | 13:50 |
| ayoung | 21:28:20 <ttx> ayoung: I think that would be a sane statement | 13:50 |
| ayoung | dolphm, a little background | 13:51 |
| ayoung | Is coming to you in an email | 13:51 |
| *** krtaylor has joined #openstack-dev | 13:51 | |
| *** mtreinish has joined #openstack-dev | 13:51 | |
| *** lloydde has joined #openstack-dev | 13:52 | |
| *** noslzzp has joined #openstack-dev | 13:52 | |
| *** drewland_ has joined #openstack-dev | 13:53 | |
| russellb | Alexei_987: talk to cyeoh, he's leading the v3 api effort | 13:54 |
| russellb | Alexei_987: oh nm, i see that you got the answer in another channel | 13:54 |
| Alexei_987 | russellb: Yep, thanks | 13:55 |
| *** jimfehlig has joined #openstack-dev | 13:55 | |
| *** drewlander has quit IRC | 13:55 | |
| *** kevinconway has joined #openstack-dev | 13:56 | |
| *** lloydde has quit IRC | 13:57 | |
| openstackgerrit | A change was merged to openstack/python-quantumclient: python3: Introduce py33 to tox.ini https://review.openstack.org/31410 | 13:58 |
| dolphm | ayoung: what's the issue? that's just a goal lol | 13:59 |
| *** rkukura has joined #openstack-dev | 13:59 | |
| ayoung | dolphm, heh, still geting the email ready. Launching it in 3. 2. 1. | 13:59 |
| *** xga__ has joined #openstack-dev | 14:00 | |
| ayoung | dolphm, it started when a couple people on my team started looking into doing additional forms of authentication to Nova and such. For people looking for Kerberos and or X509 authentication. | 14:01 |
| *** xga_ has quit IRC | 14:01 | |
| *** xga has quit IRC | 14:01 | |
| *** xga has joined #openstack-dev | 14:01 | |
| ayoung | dolphm, Turns out, that making it work for Keystone is the easy part. | 14:01 |
| *** garyk has quit IRC | 14:01 | |
| *** redbeard2 has joined #openstack-dev | 14:02 | |
| *** shang has quit IRC | 14:04 | |
| *** markmcclain has joined #openstack-dev | 14:06 | |
| *** mrodden has joined #openstack-dev | 14:07 | |
| *** shang has joined #openstack-dev | 14:09 | |
| *** odyssey4me has quit IRC | 14:10 | |
| *** cdub_ has joined #openstack-dev | 14:10 | |
| *** michchap has quit IRC | 14:12 | |
| *** zhuadl has quit IRC | 14:13 | |
| *** timello_ has quit IRC | 14:15 | |
| *** crobertsrh has joined #openstack-dev | 14:15 | |
| *** crobertsrh has left #openstack-dev | 14:15 | |
| *** timello_ has joined #openstack-dev | 14:15 | |
| *** tkammer has quit IRC | 14:15 | |
| *** pschaef has joined #openstack-dev | 14:16 | |
| *** tkammer has joined #openstack-dev | 14:16 | |
| *** aeperezt has joined #openstack-dev | 14:22 | |
| *** andrew_plunk has joined #openstack-dev | 14:22 | |
| *** rcleere has joined #openstack-dev | 14:23 | |
| *** zodiak has joined #openstack-dev | 14:23 | |
| sandywalsh | markmc, I have the link in requirements.txt and it seems to build the .tox env fine, but when the tests are running it complains it can't find the oslo module. http://paste.openstack.org/show/39250/ any ideas what I'm missing? | 14:26 |
| markmc | sandywalsh, there could be a problem here - apparently there was an issue with nova unit tests and tripleo are seeing an issue too | 14:27 |
| markmc | sandywalsh, could you confirm that 1.2.0a2 is actually installed in the venv? | 14:27 |
| markmc | sandywalsh, e.g. pkg_resources.working_set.find(pkg_resources.Requirement.parse('oslo.config>=1.2.0a2')) | 14:27 |
| *** dolphm has quit IRC | 14:28 | |
| sandywalsh | markmc, yep, sec | 14:28 |
| markmc | sandywalsh, after .tox/py27/bin/activate of course | 14:28 |
| *** wenjianhn has quit IRC | 14:28 | |
| *** egallen has joined #openstack-dev | 14:28 | |
| *** adalbas has joined #openstack-dev | 14:29 | |
| *** kaushikc has quit IRC | 14:31 | |
| *** Aarti has quit IRC | 14:31 | |
| openstackgerrit | A change was merged to openstack/swift: Remove dead code and tests for staticweb. https://review.openstack.org/34499 | 14:32 |
| *** Aarti has joined #openstack-dev | 14:32 | |
| *** dolphm has joined #openstack-dev | 14:32 | |
| sandywalsh | markmc, yes, it's there (along with 1.1 packages) http://paste.openstack.org/show/39254/ | 14:32 |
| markmc | sandywalsh, ok, I've no idea why the unit tests are getting 1.1 then :( | 14:35 |
| sandywalsh | markmc, ok, I'll dig deeper ... I'm sure there's some other CM dependency in there (perhaps nova or client) that's pulling it in | 14:36 |
| *** woodspa has joined #openstack-dev | 14:37 | |
| markmc | sandywalsh, I suspect it's some horrendous python packaging issue, like there not being .pth file to activate the egg or some other horror | 14:38 |
| sandywalsh | markmc, I removed 1.1 and it's still complaining that it can't find 1.2.0a2 | 14:39 |
| markmc | sandywalsh, it's probably mordred's fault | 14:39 |
| zoresvit | Probably, final question, is there a way to implement "and" rule for Keystone policy? Is there a specific syntax for that? Thanks. | 14:39 |
| sandywalsh | markmc, perhaps I'll take another tact and only pull over the oslo.rpc stuff and leave the DeprecatedOpt stuff out for now | 14:39 |
| markmc | sandywalsh, selective syncing should be what most people do IMHO https://wiki.openstack.org/wiki/Oslo#Syncing_Code_from_Incubator | 14:40 |
| *** topol has joined #openstack-dev | 14:41 | |
| sandywalsh | markmc, cool ... I'll try that | 14:41 |
| *** psedlak__ has quit IRC | 14:41 | |
| *** aelkikhia has joined #openstack-dev | 14:42 | |
| *** jtomasek has quit IRC | 14:42 | |
| *** jtomasek has joined #openstack-dev | 14:42 | |
| *** michchap has joined #openstack-dev | 14:42 | |
| *** kenperkins has quit IRC | 14:43 | |
| bknudson | zoresvit: in old-style, outer-list is ors, inner-lists are ands; empty string is "@". | 14:43 |
| *** beagles is now known as beagles_brb | 14:44 | |
| *** carl_baldwin has joined #openstack-dev | 14:44 | |
| *** terryh has joined #openstack-dev | 14:45 | |
| *** spzala has joined #openstack-dev | 14:46 | |
| *** _TheDodd_ has joined #openstack-dev | 14:46 | |
| zoresvit | bknudson: Thanks a lot for the tip, works! | 14:46 |
| bknudson | zoresvit: not sure where this is documented... at one point I went through the code and wrote down how I thought it worked. | 14:47 |
| *** vkmc has quit IRC | 14:47 | |
| zoresvit | bknudson: Yeah, I had troubles finding the format. | 14:47 |
| *** maheshp has joined #openstack-dev | 14:48 | |
| *** maheshp has quit IRC | 14:49 | |
| *** datsun180b has joined #openstack-dev | 14:49 | |
| *** maheshp has joined #openstack-dev | 14:49 | |
| *** FunnyLookinHat has joined #openstack-dev | 14:50 | |
| *** galstrom_zzz is now known as galstrom | 14:51 | |
| *** ifarkas has quit IRC | 14:51 | |
| *** romcheg has left #openstack-dev | 14:52 | |
| dolphm | annegentle: is diane fleming on irc? | 14:52 |
| *** michchap has quit IRC | 14:52 | |
| *** devvesa has quit IRC | 14:53 | |
| *** ffio_ has joined #openstack-dev | 14:53 | |
| *** ffio has quit IRC | 14:53 | |
| *** locke105 has quit IRC | 14:54 | |
| *** ifarkas has joined #openstack-dev | 14:54 | |
| *** mrodden has quit IRC | 14:54 | |
| *** xga_ has joined #openstack-dev | 14:55 | |
| *** aelkikhia has left #openstack-dev | 14:55 | |
| *** sneezezhang has joined #openstack-dev | 14:56 | |
| *** locke105 has joined #openstack-dev | 14:56 | |
| *** xga has quit IRC | 14:56 | |
| *** xga has joined #openstack-dev | 14:57 | |
| *** xga__ has quit IRC | 14:57 | |
| *** mkerrin has joined #openstack-dev | 14:57 | |
| *** leizhang has quit IRC | 14:58 | |
| openstackgerrit | A change was merged to openstack/python-swiftclient: Remove explicit distribute depend. https://review.openstack.org/32606 | 14:58 |
| *** dhellmann-away is now known as dhellmann | 14:58 | |
| *** jvrbanac has joined #openstack-dev | 14:58 | |
| *** Samos123 has quit IRC | 14:59 | |
| *** Samos123 has joined #openstack-dev | 15:00 | |
| *** portante has quit IRC | 15:01 | |
| *** carl_baldwin has quit IRC | 15:02 | |
| spzala | dolphm: good morning! | 15:02 |
| *** sleepsonthefloor has quit IRC | 15:03 | |
| *** drewland_ has quit IRC | 15:03 | |
| *** drewlander has joined #openstack-dev | 15:04 | |
| sdague | JordanP: ok, finally back to https://github.com/openstack/nova/blob/master/nova/api/openstack/compute/contrib/security_groups.py#L258, why doesn't that address the issue? | 15:04 |
| sdague | because validate id is different on different implementations? | 15:04 |
| JordanP | yes | 15:05 |
| *** psedlak has joined #openstack-dev | 15:05 | |
| JordanP | one checks it's an ID, the other checks it's an UUID | 15:05 |
| JordanP | thus same input returns 2 differents results | 15:05 |
| sdague | so that's sort of broken | 15:05 |
| *** jasdeepH has joined #openstack-dev | 15:05 | |
| *** Samos123 has quit IRC | 15:06 | |
| JordanP | i am not sure wa can call this broken, quatum decided to use UUID instead of ID, it's the origin of the issue | 15:06 |
| *** reidrac has quit IRC | 15:07 | |
| sdague | JordanP: so when quantum does that UUID check, does it return an exception that we can ignore or translate? | 15:07 |
| *** mrodden has joined #openstack-dev | 15:08 | |
| JordanP | its not quantum that does the check, it's nova in : nova/network/sec_group/quantum_driver.py | 15:08 |
| JordanP | before quantum (quantum server) is even called | 15:08 |
| JordanP | euh no | 15:09 |
| *** Samos123 has joined #openstack-dev | 15:09 | |
| JordanP | the validate_id() method is implemented in nova/network/sec_group/quantum_driver.py for Quantum | 15:09 |
| JordanP | but the check, ie the call to validate_id() is done by nova (see the github link you pasted) | 15:10 |
| *** lloydde has joined #openstack-dev | 15:10 | |
| *** prekarat has joined #openstack-dev | 15:10 | |
| *** gargya has quit IRC | 15:12 | |
| *** jcoufal has quit IRC | 15:12 | |
| *** pmathews has joined #openstack-dev | 15:12 | |
| sdague | JordanP: so is the crux of the issue that we just need to remove the validate_id() code in nova for the quantum driver, and it's totally self inflicted because nova code is returning differently for different back ends? | 15:13 |
| *** nati_ueno has joined #openstack-dev | 15:13 | |
| *** pmathews1 has joined #openstack-dev | 15:13 | |
| *** armax has joined #openstack-dev | 15:13 | |
| *** rcrit has quit IRC | 15:13 | |
| sdague | validate_id really seems like that needs to be nova generic code, and shouldn't be specific to secgroup implementation | 15:13 |
| *** Samos123 has quit IRC | 15:14 | |
| *** jtomasek has quit IRC | 15:14 | |
| JordanP | i am not sure I fully understand the sentence before the last one :) crux == heart ? and what do you mean by self-inflicted ? | 15:15 |
| *** Samos123 has joined #openstack-dev | 15:15 | |
| *** otherwiseguy has joined #openstack-dev | 15:16 | |
| JordanP | imo, validate_id can't be generic because the nature of an id depends on the backend. The validation should be done by the backend and not before (ie by the api) | 15:16 |
| sdague | JordanP: that's the issue | 15:16 |
| sdague | it really can't | 15:16 |
| JordanP | yep | 15:16 |
| *** navid_ has quit IRC | 15:17 | |
| sdague | so a good instance on the db side | 15:17 |
| *** pmathews has quit IRC | 15:17 | |
| *** iartarisi has quit IRC | 15:17 | |
| *** xga__ has joined #openstack-dev | 15:17 | |
| *** andrewbogott is now known as andrewbogott_afk | 15:17 | |
| sdague | mysql and pg have different max int lengths | 15:17 |
| sdague | so it turns out that sending really big ints down act differently | 15:17 |
| *** tkammer has quit IRC | 15:18 | |
| sdague | they return different errors | 15:18 |
| *** tkammer has joined #openstack-dev | 15:18 | |
| sdague | but we turn them all back into 404s because we didn't want to enforce on the top side max length for those fields | 15:18 |
| sdague | so I think in this case, it's time for the same approach | 15:18 |
| _TheDodd_ | How do I modify associate my real name and email with my IRC nick? So that the information will show up for `whois`. | 15:19 |
| *** bing_bu has quit IRC | 15:19 | |
| *** cliu has joined #openstack-dev | 15:19 | |
| *** michchap has joined #openstack-dev | 15:20 | |
| _TheDodd_ | * How do I associate my real name and email with my IRC nick? So that the information will show up for `whois`. | 15:20 |
| *** maheshp has quit IRC | 15:20 | |
| lucasagomes | sandywalsh, I got stuck in this quantum problem as well, what seems to be causing it is the .pth file from the old version of quantum: http://fpaste.org/21056/ | 15:20 |
| sdague | https://github.com/openstack/nova/blob/master/nova/compute/api.py#L1436 | 15:20 |
| lucasagomes | old version of oslo* | 15:20 |
| *** xga_ has quit IRC | 15:20 | |
| sdague | if you trace how we use InvalidID, you'll see it's thrown by a DataError exception in PG | 15:20 |
| *** danwent has joined #openstack-dev | 15:20 | |
| *** xga has quit IRC | 15:20 | |
| JordanP | yeah I believe i see how it works | 15:21 |
| *** xga has joined #openstack-dev | 15:21 | |
| *** kaushikc has joined #openstack-dev | 15:22 | |
| _TheDodd_ | * How do I associate my real name and email with my IRC nick? So that the information will show up for `whois`. | 15:22 |
| *** Samos123 has quit IRC | 15:23 | |
| *** kmartin has joined #openstack-dev | 15:25 | |
| JordanP | sdague, in fact i don't see why the id validation is necessary, quantum deals fine with id that is not UUID | 15:25 |
| *** kaushikc1 has joined #openstack-dev | 15:25 | |
| *** jasdeepH has quit IRC | 15:26 | |
| JordanP | the validate_id call in nova-api is just a way to bypass the call the to quantum backend, as an optimization | 15:26 |
| sdague | JordanP: that's probably even a better solution then, just remove that | 15:26 |
| *** rcrit has joined #openstack-dev | 15:26 | |
| *** jasdeepH has joined #openstack-dev | 15:27 | |
| sdague | yeh, that seems like the wrong place to put the optimization. I'd say just feed it to quantum, don't do the extra step, which is basically putting quantum policy in nova | 15:27 |
| JordanP | exactly ! | 15:27 |
| sdague | will be interesting if that solves the issue, or there still needs to be a translation of errors to get them to line up | 15:27 |
| sdague | will look forward to the patch :) | 15:27 |
| JordanP | nova should not be aware of what is concidered valid by Quantum, and if quantum wants to validate data, it should do in in its own code | 15:28 |
| *** _TheDodd_ has quit IRC | 15:28 | |
| *** ifarkas has quit IRC | 15:28 | |
| *** mmagr has quit IRC | 15:28 | |
| *** kaushikc1 has quit IRC | 15:28 | |
| *** kaushikc has quit IRC | 15:29 | |
| *** rwsu has joined #openstack-dev | 15:29 | |
| *** _TheDodd_ has joined #openstack-dev | 15:29 | |
| *** Ritz-Laptop has quit IRC | 15:29 | |
| bknudson | dolphm: I finally got around to trying changing policy.json without restarting keystone. It does parse policy.json on every request. | 15:30 |
| bknudson | {"error": {"message": "An unexpected error prevented the server from fulfilling your request. Expecting , delimiter: line 29 column 35 (char 1272)", "code": 500, "title": "Internal Server Error"}} | 15:30 |
| bknudson | I changed it so it's not valid. | 15:30 |
| *** bing_bu has joined #openstack-dev | 15:31 | |
| *** michchap has quit IRC | 15:31 | |
| *** martyntaylor has quit IRC | 15:31 | |
| dolphm | bknudson: finally got to your review, +2/approved | 15:33 |
| bknudson | dolphm: thank you | 15:33 |
| mirrorbox | lifeless: maybe you could share your thoughts on that subject in this bp: https://blueprints.launchpad.net/oslo/+spec/common-unit-tests ? | 15:33 |
| mirrorbox | err | 15:34 |
| dolphm | bknudson: cool, i know i had seen policy.json syntax errors raised to the api, but wasn't sure if it reloaded on every request | 15:34 |
| mirrorbox | lifeless: sorry, please ignore | 15:34 |
| openstackgerrit | A change was merged to openstack/python-keystoneclient: python3: Introduce py33 to tox.ini https://review.openstack.org/31411 | 15:34 |
| bknudson | dolphm: probably would be better if the text of the error wasn't sent to clients. Not sure if this is because of --debug. | 15:35 |
| *** mihgen has joined #openstack-dev | 15:36 | |
| *** Ritz-Laptop has joined #openstack-dev | 15:37 | |
| dolphm | bknudson: details of 500's aren't suppressed from the api in --debug | 15:38 |
| dolphm | bknudson: could make UnexpectedError extend SecurityError though to get that for free | 15:39 |
| mordred | markmc: I've sent a patch for the namespace_packages bug upstream, and have pinged the author directly to see if we can get a new release cut asap | 15:39 |
| mordred | markmc: https://github.com/iguananaut/d2to1/pull/32 if you want to track | 15:39 |
| mordred | markmc: if I don't hear back from him soonish, I'll see if I can sort a workaround in pbr itself | 15:40 |
| *** anfrolov has quit IRC | 15:40 | |
| *** otherwiseguy has quit IRC | 15:40 | |
| *** Aarti has quit IRC | 15:41 | |
| *** maheshp has joined #openstack-dev | 15:42 | |
| openstackgerrit | A change was merged to openstack/nova: Refactor db.security_group_get() instance join behavior https://review.openstack.org/34000 | 15:42 |
| *** corXi has quit IRC | 15:42 | |
| openstackgerrit | A change was merged to openstack/swift: Adding unit tests for direct client https://review.openstack.org/33405 | 15:42 |
| *** Aarti has joined #openstack-dev | 15:42 | |
| openstackgerrit | A change was merged to openstack/oslo-incubator: Refactor KombuStubs fixture. https://review.openstack.org/33804 | 15:43 |
| *** beagles_brb is now known as beagles | 15:43 | |
| *** wenjianhn has joined #openstack-dev | 15:45 | |
| *** bing_bu has quit IRC | 15:45 | |
| *** yidclare has joined #openstack-dev | 15:45 | |
| *** bing_bu has joined #openstack-dev | 15:45 | |
| *** matiu has joined #openstack-dev | 15:46 | |
| *** maheshp has quit IRC | 15:46 | |
| *** maheshp has joined #openstack-dev | 15:46 | |
| *** martyntaylor has joined #openstack-dev | 15:47 | |
| *** cliu has quit IRC | 15:48 | |
| *** Samos123 has joined #openstack-dev | 15:48 | |
| *** maheshp has quit IRC | 15:48 | |
| *** sandeepr has quit IRC | 15:49 | |
| JordanP | sdague, I know you're against having different input (id or uuid) depending on the backend, in a Tempest test. But'is already the case in nova unit test : https://github.com/openstack/nova/blob/master/nova/tests/api/openstack/compute/contrib/test_security_groups.py#L127 | 15:49 |
| *** michchap has joined #openstack-dev | 15:50 | |
| *** sandeepr has joined #openstack-dev | 15:50 | |
| *** bing_bu has quit IRC | 15:50 | |
| markmc | mordred, excellent, thanks | 15:50 |
| *** michchap_ has joined #openstack-dev | 15:51 | |
| markmc | mordred, proxying that to the bug | 15:51 |
| *** Ritz-Laptop has quit IRC | 15:51 | |
| *** edmund has joined #openstack-dev | 15:51 | |
| *** pabelanger_ has joined #openstack-dev | 15:52 | |
| *** dprince has quit IRC | 15:52 | |
| mordred | JordanP: yeah, but that's a unit test ... tempest is black-box testing what the api experience is | 15:52 |
| *** michcha__ has joined #openstack-dev | 15:52 | |
| markmc | mordred, I see you've some pull requests sitting there for some months :( | 15:52 |
| *** enikanorov has quit IRC | 15:52 | |
| *** pabelanger_ has quit IRC | 15:52 | |
| *** pabelanger_ has joined #openstack-dev | 15:52 | |
| mordred | JordanP: if the end user needs to know something about the implementation choice of the provider, something is broken in the interface | 15:52 |
| mordred | markmc: yeah, that's why I pinged him directly | 15:53 |
| *** pabelanger has quit IRC | 15:53 | |
| markmc | mordred, cool | 15:53 |
| *** pabelanger_ is now known as pabelanger | 15:53 | |
| mordred | markmc: I spoke with him when we were looking at using d2to1 about whether I should make pbr a fork of d2to1 or use it, and he said he was interested in continuing to maintain it as a library | 15:53 |
| *** pabelanger_ has joined #openstack-dev | 15:53 | |
| *** pabelanger has quit IRC | 15:53 | |
| JordanP | mordred, good point | 15:53 |
| *** michch___ has joined #openstack-dev | 15:53 | |
| markmc | mordred, ok | 15:53 |
| *** pabelanger has joined #openstack-dev | 15:54 | |
| *** michchap has quit IRC | 15:54 | |
| mordred | markmc: but, tbh, there isn't _that_ much code in there. If he goes awol, I'll just merge the functionality directly into pbr and be done with it | 15:54 |
| markmc | mordred, funny with this new github world, it's upload access to pypi being restricted to a single maintainer would be the bigger worry than commit access :) | 15:54 |
| markmc | mordred, yeah, that's an option too | 15:54 |
| mordred | markmc: I would prefer to find a path forward for sanity though | 15:55 |
| *** michchap has joined #openstack-dev | 15:55 | |
| markmc | mordred, yep | 15:55 |
| mordred | markmc: actually, you can add uploaders on pypi | 15:55 |
| markmc | mordred, right, you could ask him to add you :) | 15:55 |
| *** sleepsonthefloor has joined #openstack-dev | 15:55 | |
| *** michchap_ has quit IRC | 15:55 | |
| *** michchap_ has joined #openstack-dev | 15:55 | |
| *** lloydde has quit IRC | 15:55 | |
| mordred | markmc: just got a response, he's been on vacation for a few weeks and is working through post-vacation email backlog | 15:56 |
| *** michcha__ has quit IRC | 15:56 | |
| mordred | markmc: but he's going to bump that patch to the top of the list | 15:56 |
| markmc | mordred, great | 15:56 |
| *** Alexei_987 has quit IRC | 15:56 | |
| *** redondos has quit IRC | 15:56 | |
| *** michcha__ has joined #openstack-dev | 15:57 | |
| ayoung | dolphm, https://review.openstack.org/#/c/32510/ OK to merge? | 15:57 |
| *** jgallard has joined #openstack-dev | 15:57 | |
| *** michch___ has quit IRC | 15:58 | |
| *** yidclare has quit IRC | 15:58 | |
| *** avishay has joined #openstack-dev | 15:58 | |
| *** michchap has quit IRC | 15:59 | |
| *** armax has left #openstack-dev | 15:59 | |
| *** electrichead has joined #openstack-dev | 16:00 | |
| *** jsavak has joined #openstack-dev | 16:00 | |
| *** michchap_ has quit IRC | 16:00 | |
| *** jdurgin1 has joined #openstack-dev | 16:00 | |
| *** michcha__ has quit IRC | 16:01 | |
| *** michchap has joined #openstack-dev | 16:01 | |
| *** Alla has joined #openstack-dev | 16:02 | |
| *** carl_baldwin has joined #openstack-dev | 16:03 | |
| *** michchap_ has joined #openstack-dev | 16:03 | |
| *** joesavak has quit IRC | 16:03 | |
| *** joesavak has joined #openstack-dev | 16:04 | |
| *** jasondotstar has joined #openstack-dev | 16:04 | |
| *** pcm__ has quit IRC | 16:04 | |
| *** jsavak has quit IRC | 16:05 | |
| *** michchap has quit IRC | 16:06 | |
| *** michchap_ has quit IRC | 16:07 | |
| *** garyk has joined #openstack-dev | 16:09 | |
| *** JordanP has quit IRC | 16:10 | |
| *** alop has joined #openstack-dev | 16:10 | |
| *** SumitNaiksatam has quit IRC | 16:11 | |
| *** devoid has joined #openstack-dev | 16:11 | |
| *** xga__ has quit IRC | 16:13 | |
| *** xga has quit IRC | 16:14 | |
| *** eglynn is now known as eglynn-afk | 16:14 | |
| sandywalsh | lucasagomes, thanks ... I'll try that (I wonder how we would patch that?) | 16:15 |
| lucasagomes | sandywalsh, it's more like a workaround for now | 16:16 |
| *** kbrierly has joined #openstack-dev | 16:17 | |
| markmc | lucasagomes, sorry, missed your comment earlier | 16:17 |
| openstackgerrit | A change was merged to openstack/keystone: LDAP list group users not fail if user entry deleted https://review.openstack.org/34475 | 16:18 |
| markmc | lucasagomes, we'll need to understand why the version in the venv doesn't override that | 16:18 |
| markmc | lucasagomes, nasty | 16:18 |
| *** psedlak has quit IRC | 16:18 | |
| markmc | mordred, see this? http://fpaste.org/21056/ | 16:18 |
| lucasagomes | markmc, yea it's pretty nasty :/ | 16:18 |
| markmc | mordred, oslo.config 1.1 in /usr/lib, 1.2 in a venv, 1.1 somehow gets loaded | 16:18 |
| markmc | mordred, it's a --system-site-packages venv, but that shouldn't be an issue | 16:19 |
| markmc | mordred, presumably this was the problem with nova too | 16:19 |
| lucasagomes | in my case I don't even have the oslo-config 1.1 in my venv, the pth file making the venv to leak and use the system one | 16:20 |
| markmc | mordred, see https://review.openstack.org/33888 for the nova revert | 16:20 |
| markmc | lucasagomes, yep | 16:20 |
| markmc | dansmith, got more details on that nova revert ? | 16:21 |
| dansmith | markmc: actually clarkb understood it more than I did | 16:21 |
| markmc | dansmith, ok | 16:21 |
| dansmith | markmc: but it was breaking tox runs for everyone, but working in CI | 16:21 |
| markmc | clarkb, more details ? | 16:21 |
| *** markwash has joined #openstack-dev | 16:21 | |
| mordred | yeek. how the heck does it work in CI but break locally? | 16:22 |
| * markmc thinks we need a hadoop based irc -> launchpad proxy | 16:22 | |
| markmc | it's a Big Data problem | 16:22 |
| dansmith | mordred: yeah, it was scary.. I think he said some sort of test ordering thing was making it pass erroneously | 16:22 |
| russellb | markmc: wat | 16:22 |
| markmc | useful info on irc, need to extract it into launchpad | 16:22 |
| *** lloydde has joined #openstack-dev | 16:22 | |
| *** electrichead has quit IRC | 16:22 | |
| mordred | markmc: is it in scrollback? | 16:22 |
| *** _TheDodd_ has quit IRC | 16:23 | |
| markmc | mordred, no idea, I only saw the revert commit days after it was proposed | 16:23 |
| dansmith | it was in infra, and a week ago | 16:23 |
| * markmc has no context other than "nova broke" | 16:23 | |
| mordred | oy. I'll never find that in my scrollback | 16:23 |
| markmc | which isn't in eavesdrop either | 16:23 |
| *** nati_ueno has quit IRC | 16:23 | |
| russellb | was also in -nova | 16:23 |
| russellb | (not logged either afaik) | 16:23 |
| mordred | I'll see if clarkb can fill me in on his understanding when he gets back online | 16:23 |
| mordred | we should maybe log more channels | 16:23 |
| markmc | can we file bugs people, please? :) | 16:23 |
| * markmc pleads | 16:23 | |
| russellb | i don't remember, people were just like "omg it broke and if i revert it works, let's revert and we'll figure it out" | 16:23 |
| *** jruzicka has quit IRC | 16:24 | |
| markmc | russellb, I'm totally down with that | 16:24 |
| dansmith | markmc: yeah, my bad, sorry, it was right after merge, and major breakage | 16:24 |
| russellb | with a bug report? :) | 16:24 |
| markmc | dansmith, anything you can dig up from scrollback or whatever would be great | 16:24 |
| russellb | re-apply patch and see what happens now? | 16:25 |
| *** otherwiseguy has joined #openstack-dev | 16:25 | |
| markmc | russellb, yeah | 16:25 |
| dansmith | well, | 16:25 |
| dansmith | there was some analysis by clarkb | 16:25 |
| dansmith | I'll get some logs, just a sec | 16:25 |
| markmc | russellb, in other words, "STFU and reproduce it yourself" :) | 16:25 |
| russellb | i meant that to dansmith | 16:25 |
| *** electrichead has joined #openstack-dev | 16:25 | |
| *** xqueralt is now known as xqueralt-afk | 16:25 | |
| markmc | russellb, heh, ok | 16:25 |
| * markmc can just reproduce it | 16:26 | |
| *** martyntaylor has quit IRC | 16:26 | |
| *** markwash_ has joined #openstack-dev | 16:27 | |
| *** garyk has quit IRC | 16:27 | |
| *** mkollaro has quit IRC | 16:28 | |
| *** boris-42 has quit IRC | 16:28 | |
| *** anfrolov has joined #openstack-dev | 16:29 | |
| dansmith | markmc: http://paste.openstack.org/show/39258/ | 16:29 |
| dansmith | markmc: that's from when I started the conversation through the end of the day | 16:29 |
| *** markwash has quit IRC | 16:29 | |
| *** markwash_ is now known as markwash | 16:29 | |
| markmc | dansmith, heh, ok - thanks | 16:29 |
| *** kenperkins has joined #openstack-dev | 16:30 | |
| markmc | <clarkb>this is my gripe with oslo.config the way we use it makes import order super important rather than putting the config stuff right at the beginning of the import order we mix it around | 16:30 |
| bknudson | ayoung: would be nice to get https://review.openstack.org/#/c/33102/ merged with https://review.openstack.org/#/c/32510/ , since both are fixing MyISAM. | 16:30 |
| *** anfrolov has quit IRC | 16:30 | |
| * markmc can't parse that | 16:31 | |
| *** tkammer has quit IRC | 16:31 | |
| markmc | <clarkb>geekinutah: instead this is a test run order conflict. run tests in one order and they pass. Run them in a different order and they fail | 16:31 |
| markmc | that doesn't sound right | 16:31 |
| *** Guest32596 is now known as med_ | 16:32 | |
| *** med_ has quit IRC | 16:32 | |
| *** med_ has joined #openstack-dev | 16:32 | |
| *** gyee has joined #openstack-dev | 16:32 | |
| *** shang_ has joined #openstack-dev | 16:32 | |
| *** epim has joined #openstack-dev | 16:33 | |
| *** jgallard has quit IRC | 16:34 | |
| *** galstrom is now known as galstrom_zzz | 16:34 | |
| *** shang has quit IRC | 16:36 | |
| *** Aarti has quit IRC | 16:36 | |
| *** jcoufal has joined #openstack-dev | 16:36 | |
| *** michchap has joined #openstack-dev | 16:36 | |
| *** Aarti has joined #openstack-dev | 16:36 | |
| *** SergeyLukjanov has quit IRC | 16:36 | |
| *** eglynn-afk is now known as eglynn | 16:37 | |
| *** derekh has quit IRC | 16:37 | |
| *** leizhang has joined #openstack-dev | 16:37 | |
| *** galstrom_zzz is now known as galstrom | 16:37 | |
| *** sulrich has joined #openstack-dev | 16:38 | |
| *** Mandell has joined #openstack-dev | 16:38 | |
| *** dprince has joined #openstack-dev | 16:38 | |
| mordred | markmc: I wonder if the problem is related to the namespace package problem | 16:39 |
| *** martyntaylor has joined #openstack-dev | 16:39 | |
| *** gongysh has quit IRC | 16:39 | |
| *** martyntaylor has quit IRC | 16:39 | |
| *** sneezezhang has quit IRC | 16:39 | |
| *** SumitNaiksatam has joined #openstack-dev | 16:39 | |
| *** martyntaylor has joined #openstack-dev | 16:39 | |
| markmc | mordred, maybe, can't really see how though | 16:40 |
| markmc | mordred, if it was the other way around, maybe - i.e. if there was a oslo/__init__.py in /usr/lib | 16:40 |
| mordred | yeah. good point | 16:40 |
| markmc | mordred, anything is possible with this stuff though | 16:40 |
| mordred | yup. also, the nova site-packages/libvirt bit makes the isolation we shoot for harder | 16:41 |
| *** egallen has quit IRC | 16:42 | |
| *** colinmcnamara has joined #openstack-dev | 16:43 | |
| *** ajyounge_ has joined #openstack-dev | 16:43 | |
| *** colinmcnamara has quit IRC | 16:45 | |
| *** garyk has joined #openstack-dev | 16:45 | |
| *** _TheDodd_ has joined #openstack-dev | 16:45 | |
| *** michchap has quit IRC | 16:48 | |
| iccha | annegentle: any leads on glance documentation ? (based on our previous conversation) | 16:49 |
| *** freedomhui has joined #openstack-dev | 16:51 | |
| *** jyhc has joined #openstack-dev | 16:53 | |
| *** freedomhui has quit IRC | 16:54 | |
| *** romcheg has joined #openstack-dev | 16:55 | |
| *** nati_ueno has joined #openstack-dev | 16:55 | |
| sandywalsh | markmc, the update.py file seems to want to copy everything all the time. I've moved oslo-common.conf and issued http://paste.openstack.org/show/39260/, but it replaces all modules. | 16:56 |
| sandywalsh | markmc, is there a different incantation I should be using? | 16:56 |
| markmc | sandywalsh, ugh, try --nodeps | 16:57 |
| *** devoid has quit IRC | 16:57 | |
| * markmc knew that feature was a bad idea | 16:57 | |
| sandywalsh | markmc, thanks | 16:57 |
| *** kushal has quit IRC | 17:03 | |
| *** jyhc has quit IRC | 17:06 | |
| *** comay has joined #openstack-dev | 17:09 | |
| *** lucasagomes has quit IRC | 17:09 | |
| *** mikal has quit IRC | 17:09 | |
| *** danpb has quit IRC | 17:10 | |
| *** mikal has joined #openstack-dev | 17:11 | |
| *** athomas has quit IRC | 17:11 | |
| *** esheffield1 has joined #openstack-dev | 17:13 | |
| *** terriyu has joined #openstack-dev | 17:13 | |
| *** esheffield has quit IRC | 17:14 | |
| *** michchap has joined #openstack-dev | 17:15 | |
| *** athomas has joined #openstack-dev | 17:17 | |
| *** wenjianhn has quit IRC | 17:17 | |
| mordred | markmc: speaking of ... | 17:17 |
| mordred | markmc: when pip 1.4 comes out | 17:17 |
| mordred | markmc: I'd like to have a discussion with you about re-thinking a few things that we do | 17:17 |
| mordred | markmc: because I think the machinery will be different/better | 17:18 |
| mordred | markmc: it might not be possible to do _anything_ differently- but I think it's worth a mental exercise | 17:18 |
| *** rkukura has quit IRC | 17:18 | |
| mordred | markmc: (this is all around the new behavior of not installing pre-release things unless you explicitly ask for them) | 17:19 |
| *** athomas has quit IRC | 17:20 | |
| *** pschaef has quit IRC | 17:20 | |
| *** xBytez has quit IRC | 17:21 | |
| *** jyhc has joined #openstack-dev | 17:21 | |
| *** jyhc has quit IRC | 17:21 | |
| *** jyhc has joined #openstack-dev | 17:21 | |
| *** sandywalsh has quit IRC | 17:21 | |
| *** sarob has joined #openstack-dev | 17:22 | |
| openstackgerrit | A change was merged to openstack/swift: Change ring partition size for SAIO https://review.openstack.org/34574 | 17:22 |
| *** andrewbogott_afk is now known as andrewbogott | 17:22 | |
| *** sarob has quit IRC | 17:23 | |
| *** sarob has joined #openstack-dev | 17:23 | |
| *** jbresnah has joined #openstack-dev | 17:24 | |
| *** sarob has quit IRC | 17:24 | |
| *** sarob has joined #openstack-dev | 17:25 | |
| *** Aarti has quit IRC | 17:27 | |
| *** michchap has quit IRC | 17:27 | |
| *** Farooque has joined #openstack-dev | 17:27 | |
| *** sarob_ has joined #openstack-dev | 17:27 | |
| *** rods has joined #openstack-dev | 17:28 | |
| *** xBytez has joined #openstack-dev | 17:30 | |
| *** sarob__ has joined #openstack-dev | 17:30 | |
| *** ssurana has joined #openstack-dev | 17:31 | |
| *** sarob has quit IRC | 17:31 | |
| *** sarob_ has quit IRC | 17:31 | |
| *** kushal has joined #openstack-dev | 17:32 | |
| *** prekarat1 has joined #openstack-dev | 17:33 | |
| *** prekarat has quit IRC | 17:34 | |
| mordred | markmc: also - workaround for namespace package issue until a fix is in to d2to1: | 17:35 |
| mordred | markmc: add namespace_packages = ['oslo'], into setup.py directly | 17:36 |
| mordred | markmc: it's not how we want to do that- but it will work until a new fix is released upstream | 17:36 |
| *** mihgen has quit IRC | 17:36 | |
| *** GheRiver1 is now known as GheRivero | 17:38 | |
| *** sandywalsh has joined #openstack-dev | 17:38 | |
| *** johnthetubaguy has quit IRC | 17:38 | |
| *** sgordon has quit IRC | 17:38 | |
| *** sgordon has joined #openstack-dev | 17:38 | |
| *** prekarat1 has quit IRC | 17:39 | |
| *** zaitcev has joined #openstack-dev | 17:39 | |
| *** Ryan_Lane has joined #openstack-dev | 17:41 | |
| lbragstad | mikal: jamielennox and I were discussing earlier what the purpose of the context being stored in a WeakLocal object was? We are trying to implement a unified logging solution into keystone from Oslo-incubator but Keystone is hesitant on bringing in dependencies on eventlet (corolocal.local). Just wondering if you had a take on what we could do on this? | 17:42 |
| *** Ryan_Lane1 has joined #openstack-dev | 17:43 | |
| *** tmclaugh[work] has joined #openstack-dev | 17:43 | |
| *** Ryan_Lane1 has quit IRC | 17:44 | |
| *** Ryan_Lane1 has joined #openstack-dev | 17:44 | |
| *** Ryan_Lane has quit IRC | 17:44 | |
| *** Ryan_Lane1 is now known as Ryan_Lane | 17:44 | |
| *** rnirmal has quit IRC | 17:44 | |
| *** sandywalsh has quit IRC | 17:44 | |
| *** ndipanov is now known as ndipanov_gone | 17:45 | |
| *** otherwiseguy has quit IRC | 17:46 | |
| *** xmltok has joined #openstack-dev | 17:48 | |
| *** xmltok has quit IRC | 17:49 | |
| *** Tross has quit IRC | 17:50 | |
| *** xmltok_ has quit IRC | 17:50 | |
| *** galstrom is now known as galstrom_zzz | 17:50 | |
| *** jcoufal has quit IRC | 17:50 | |
| *** xmltok has joined #openstack-dev | 17:50 | |
| *** michchap has joined #openstack-dev | 17:53 | |
| *** romcheg has left #openstack-dev | 17:54 | |
| *** mikal has quit IRC | 17:55 | |
| *** mrodden has quit IRC | 17:56 | |
| *** mikal has joined #openstack-dev | 17:57 | |
| *** pcm___ has joined #openstack-dev | 17:58 | |
| *** sandywalsh has joined #openstack-dev | 17:58 | |
| *** pcm___ has quit IRC | 17:58 | |
| *** timello_ has quit IRC | 17:58 | |
| *** pcm__ has joined #openstack-dev | 17:58 | |
| *** timello_ has joined #openstack-dev | 17:59 | |
| *** lbragstad has quit IRC | 17:59 | |
| *** mrodden has joined #openstack-dev | 17:59 | |
| *** sdake has quit IRC | 17:59 | |
| *** ydacheville has joined #openstack-dev | 18:00 | |
| *** lbragstad has joined #openstack-dev | 18:00 | |
| *** ydacheville_ has quit IRC | 18:04 | |
| *** esheffield1 has quit IRC | 18:05 | |
| *** michchap has quit IRC | 18:06 | |
| *** _TheDodd_ has quit IRC | 18:06 | |
| *** Mandell has quit IRC | 18:07 | |
| *** _TheDodd_ has joined #openstack-dev | 18:09 | |
| *** otherwiseguy has joined #openstack-dev | 18:11 | |
| *** anfrolov has joined #openstack-dev | 18:13 | |
| *** anniec has joined #openstack-dev | 18:14 | |
| *** ctlaugh has quit IRC | 18:14 | |
| *** jpich has quit IRC | 18:16 | |
| *** redondos has joined #openstack-dev | 18:16 | |
| *** Farooque has left #openstack-dev | 18:16 | |
| *** sulrich has quit IRC | 18:17 | |
| openstackgerrit | A change was merged to openstack/cinder: Content-length missing in put_object https://review.openstack.org/33639 | 18:17 |
| openstackgerrit | A change was merged to openstack/nova: Add basic InstanceFault model https://review.openstack.org/33551 | 18:17 |
| openstackgerrit | A change was merged to openstack/nova: Add Instance.fault https://review.openstack.org/33552 | 18:18 |
| *** sulrich has joined #openstack-dev | 18:18 | |
| openstackgerrit | A change was merged to openstack/nova: Make NovaObject support the 'in' operator https://review.openstack.org/33871 | 18:18 |
| *** cp16net|away is now known as cp16net | 18:19 | |
| *** sarob has joined #openstack-dev | 18:19 | |
| *** Mandell has joined #openstack-dev | 18:20 | |
| *** cliu has joined #openstack-dev | 18:22 | |
| *** otherwiseguy has quit IRC | 18:22 | |
| *** sarob__ has quit IRC | 18:22 | |
| *** sushils has quit IRC | 18:23 | |
| *** dhellmann has quit IRC | 18:23 | |
| *** epim has quit IRC | 18:25 | |
| pcm__ | nati_ueno: Hi! Do you have time for a few questions on VPNaaS? | 18:25 |
| nati_ueno | pcm__: absolutely :) | 18:25 |
| openstackgerrit | A change was merged to openstack/nova: Fix instance object's use of a db query method parameter https://review.openstack.org/33872 | 18:26 |
| pcm__ | nati_ueno: I set up two devstack nodes on bare metal. Have public net connection. Trying to setup connection | 18:26 |
| nati_ueno | pcm__: Ah sorry, please use latest patch. I did same thing yesterday, and I found one bug. But newest patch works | 18:27 |
| openstackgerrit | A change was merged to openstack/nova: Make info_cache handle when network_info is None https://review.openstack.org/33873 | 18:27 |
| pcm__ | nati_ueno: Q: I was looking to see if config files are created. Doesn't look like we set, so default should be /var/lib/quantum/strongswan/, right? | 18:27 |
| nati_ueno | pcm__: /opt/stack/data/quantum/strongswan is default for devestack | 18:27 |
| *** galstrom_zzz is now known as galstrom | 18:28 | |
| pcm__ | nati_ueno: Oh, where does that get set? | 18:28 |
| nati_ueno | pcm__: In devstack, it setup state_path. | 18:28 |
| nati_ueno | pcm__: strongswan conf goes unser the state_path | 18:28 |
| nati_ueno | s/unser/under/ | 18:29 |
| *** SumitNaiksatam has quit IRC | 18:29 | |
| pcm__ | nati_ueno: checking... | 18:29 |
| nati_ueno | pcm__: here https://github.com/openstack-dev/devstack/blob/master/lib/quantum#L463 | 18:30 |
| pcm__ | nati_ueno: Yes. I see the directory - that solves one question. | 18:31 |
| pcm__ | I did ikepolicy, ipsecpolicy, service and connection create on each side. | 18:31 |
| nati_ueno | pcm__: ok | 18:31 |
| *** michchap has joined #openstack-dev | 18:32 | |
| pcm__ | nati_ueno: I see entries with UUID of router. | 18:32 |
| vishy | markmcclain, arosen: https://review.openstack.org/#/c/34589/ | 18:33 |
| *** jculp_ has joined #openstack-dev | 18:33 | |
| *** jculp_ has quit IRC | 18:33 | |
| vishy | markmcclain, arosen: i initially added that for compatibility with our nova-network install. It turns out that it isn't a show-stopper, but I think the patch still makes sense. | 18:33 |
| pcm__ | nati_ueno: Just checked and it looks like service is ACTIVE | 18:34 |
| *** markmc has quit IRC | 18:34 | |
| pcm__ | nati_ueno: However, the connection is saying "PENDING_CREATE" | 18:35 |
| pcm__ | nati_ueno: Am I missing some step? | 18:35 |
| nati_ueno | pcm__: That's part is not implemented yet. | 18:35 |
| pcm__ | ah. | 18:36 |
| nati_ueno | pcm__: I mentioned it in the commit message | 18:36 |
| nati_ueno | pcm__: so We need to update connection status based on the ipsec status command | 18:36 |
| pcm__ | OK. Didn't read the message closely. Sorry. | 18:36 |
| nati_ueno | pcm__: Noop. The problem is connection statte could be dynamically changed | 18:36 |
| nati_ueno | pcm__: so pediatric check might be needed | 18:37 |
| *** novas0x2a|laptop has joined #openstack-dev | 18:37 | |
| *** diogogmt has joined #openstack-dev | 18:37 | |
| nati_ueno | pcm__: current patch is about 1k, so I thought it good to separate for another patch | 18:37 |
| markmcclain | vishy: looking | 18:38 |
| vishy | markmcclain: I didn't think it was big enough for a blueprint and I'm not sure it is really a "bug", so i just went for straight patch on this one | 18:39 |
| pcm__ | nati_ueno: Can I manually monitor/access the ipsec status? | 18:39 |
| *** epim has joined #openstack-dev | 18:39 | |
| *** otherwiseguy has joined #openstack-dev | 18:39 | |
| nati_ueno | pcm__: yes. so please go to $state_path/quantum/strongswan/$ROUTER_ID/ | 18:40 |
| nati_ueno | then | 18:40 |
| nati_ueno | sudo ip netns exec qrouter-$ROUTER_ID | 18:40 |
| nati_ueno | sorry typo | 18:40 |
| nati_ueno | sudo ip netns exec qrouter-$ROUTER_ID bash | 18:40 |
| nati_ueno | then | 18:40 |
| nati_ueno | quantum-vpn-nswrap `pwd` ipsec status | 18:40 |
| pcm__ | nati_ueno: Ah, so that runs the wrapper to run the ipsec command under the processes area. Saw that in the wiki, but wasn't sure how it worked. | 18:42 |
| pcm__ | nati_ueno: Cool! I'll give it a try. | 18:42 |
| *** lloydde has quit IRC | 18:42 | |
| nati_ueno | pcm__: Thanks! | 18:42 |
| pcm__ | nati_ueno: Still trying to wrap my head around it all :) | 18:42 |
| pcm__ | nati_ueno: Do have a bit of confusion on the stronswan sync() method. | 18:43 |
| nati_ueno | pcm__: thank you for your review. I'm working on it | 18:43 |
| nati_ueno | pcm__: What's confusion? | 18:43 |
| markmcclain | vishy: change makes sense | 18:44 |
| pcm__ | nati_ueno: It is getting a list of vpnservices. What are they? | 18:44 |
| *** michchap has quit IRC | 18:44 | |
| pcm__ | nati_ueno: Getting a bit confused on who's talking to whom... it contacts agent_rpc to get_vpn_services_on_host(). Is that the OVS agent or other? | 18:45 |
| *** eglynn has quit IRC | 18:45 | |
| nati_ueno | pcm__: get_vpn_services_on_host is in the service side ipsec driver code | 18:45 |
| nati_ueno | pcm__: it gets assigned vpnservices for router | 18:46 |
| *** esheffield has joined #openstack-dev | 18:46 | |
| pcm__ | nati_ueno: So is that IPSec> | 18:47 |
| pcm__ | ? | 18:47 |
| nati_ueno | pcm__: yes whole configuration | 18:48 |
| *** mrodden has quit IRC | 18:48 | |
| *** lloydde has joined #openstack-dev | 18:49 | |
| pcm__ | nati_ueno: Maybe I could step through the sync() function to get a better understanding of the pieces and flow? | 18:49 |
| bnemec | Hmm, I'm running stack.sh and seeing the following error: | 18:50 |
| bnemec | ++ nova flavor-create m1.nano 42 64 0 1 | 18:50 |
| bnemec | ERROR: printt | 18:50 |
| bnemec | I get the same thing if I try to run any other nova commands. | 18:50 |
| pcm__ | nati_ueno: Could I set a pdb breakpoint, restart the quantum process and then issue a command to step through the code in that area? | 18:50 |
| bnemec | I'll look into it, but just wondering if anyone else has run seen it. | 18:50 |
| *** devoid has joined #openstack-dev | 18:51 | |
| nati_ueno | pcm__: yes it should work | 18:51 |
| nati_ueno | nn | 18:51 |
| nati_ueno | sorry typo | 18:51 |
| *** portante has joined #openstack-dev | 18:52 | |
| pcm__ | nati_ueno: I'm trying to understand how the flow goes. | 18:52 |
| *** bknudson has quit IRC | 18:53 | |
| nati_ueno | pcm__: here is bp https://docs.google.com/a/ntti3.com/presentation/d/1uoYMl2fAEHTpogAe27xtGpPcbhm7Y3tlHIw_G1Dy5aQ/edit#slide=id.gc2bc6c2e_024 This may help you to understand the code :) | 18:55 |
| pcm__ | nati_ueno: Thanks! Will look at it. I'm having a bit of a hard time figuring out who's calling who... | 18:57 |
| *** Tross has joined #openstack-dev | 18:57 | |
| nati_ueno | pcm__: yeah, it's complicated | 18:58 |
| *** jjmb has quit IRC | 18:58 | |
| nati_ueno | pcm__: basically, service_driver calls device_driver | 18:58 |
| pcm__ | nati_ueno: Looked like plugin updates DBase and then calls driver (vpnservice_updated?) which seemed to call IPSec Agent API. Is that right? | 18:58 |
| nati_ueno | pcm__: However l3_agent also calls device_driver in case of router create and delete and restart | 18:58 |
| nati_ueno | pcm__: yes | 18:59 |
| pcm__ | nati_ueno: That seems to get l3_agents and then sends RPC? | 18:59 |
| *** rnirmal has joined #openstack-dev | 18:59 | |
| pcm__ | nati_ueno: Does that then end up at the StrongSwan Driver RPC? | 19:00 |
| nati_ueno | so device driver consumes VPN_AGENT_TOPIC with host name | 19:00 |
| nati_ueno | IPSec Agent send message for the TOPIC | 19:00 |
| nati_ueno | pcm__: yes | 19:00 |
| *** avishay has quit IRC | 19:01 | |
| *** cp16net is now known as cp16net|away | 19:01 | |
| nati_ueno | pcm__: you can find topic value in the rpc call. | 19:01 |
| *** cp16net|away is now known as cp16net | 19:01 | |
| nati_ueno | pcm__: you can also find topic value for comsumer code | 19:01 |
| dolphm | ayoung: did you already start working bug 1194938? | 19:02 |
| nati_ueno | pcm__: https://review.openstack.org/#/c/33148/8/quantum/services/vpn/service_drivers/ipsec_driver.py L87 is rpc sender | 19:02 |
| ayoung | dolphm, no | 19:03 |
| nati_ueno | pcm__: https://review.openstack.org/#/c/33148/8/quantum/services/vpn/device_drivers/strongswan_driver.py L220 is consumer code | 19:03 |
| ayoung | dolphm, just want the bug in place for when we do, and to have a place to contain the discussion about it | 19:03 |
| nati_ueno | pcm__: then you can know the flow | 19:03 |
| dolphm | ayoung: i've actually already did the same last week, while playing with caching at the manager layer... happy to pull that out and propose | 19:03 |
| *** jjmb has joined #openstack-dev | 19:04 | |
| pcm__ | nati_ueno: Great! That seems to match what I have so far in my notes. | 19:04 |
| dolphm | s/did/done/ | 19:04 |
| nati_ueno | pcm__: Ok sounds good :) | 19:04 |
| dolphm | s/grammar/more better/ | 19:04 |
| ayoung | dolphm, oh, I wish I had known that before submitting this: https://review.openstack.org/#/c/34254/ | 19:05 |
| *** rkukura has joined #openstack-dev | 19:05 | |
| dolphm | ayoung: holy crap lol | 19:05 |
| lifeless | ttx: should I put forward a proposal for TripleO at this point ? | 19:05 |
| lifeless | ttx: (or perhaps 'Ops' or 'Deployments' - the other names seem to be sane rather than cute). | 19:06 |
| dolphm | ayoung: can we agree to fix bug 1194938 before doing that? | 19:06 |
| pcm__ | nati_ueno: OK, so then SS driver updates NAT and updates the config files as the last steps. | 19:06 |
| ayoung | dolphm, sure. | 19:07 |
| ttx | lifeless: it's a bit early still. We didn't approve the concept of "programs" yet. Next week ? | 19:07 |
| ayoung | lifeless, calle it Trapelo | 19:07 |
| lifeless | ayoung: heh :) | 19:07 |
| lifeless | ttx: kk | 19:07 |
| ttx | lifeless: TripleO is fine, after all tere is Oslo | 19:07 |
| lifeless | ttx: gives me time to work on some prose | 19:08 |
| nati_ueno | pcm__: no, ss will do sync with servers. Then get vpnservices, then setup nat and setup conf | 19:08 |
| dolphm | ayoung: assigning the bug to myself, i'll put up a patch in next 24 hours | 19:08 |
| ayoung | https://en.wiktionary.org/wiki/trapelare#Italian | 19:08 |
| *** dperaza has joined #openstack-dev | 19:08 | |
| *** n0ano has quit IRC | 19:08 | |
| ayoung | dolphm, can you send me an early version? I'll redo my patch ontop of what you have and rebase when the real thing comes through. | 19:09 |
| lifeless | ayoung: nice | 19:09 |
| pcm__ | nati_ueno: OK. I'll need to look through that part "sync with servers". | 19:09 |
| ayoung | https://en.wiktionary.org/wiki/trape%C5%82o | 19:09 |
| *** Tross has quit IRC | 19:09 | |
| ayoung | Gadget, Contraption. OK lifeless we need to call it Trapelo | 19:10 |
| dolphm | ayoung: what i have right now wouldn't be useful to rebase on top of, but i'll make it a priority | 19:10 |
| ayoung | dolphm, fair enough. | 19:10 |
| nati_ueno | pcm__: OK I'm following same model of current l3-agent. so l3-agent code may also helpful | 19:10 |
| dolphm | ayoung: mostly because i completely rewrote every manager, which i plan on undo-ing :) | 19:11 |
| dolphm | ayoung: i went down a long dark road and changed my mind | 19:11 |
| *** michchap has joined #openstack-dev | 19:11 | |
| pcm__ | nati_ueno: I don't know that code (yet) either :( | 19:11 |
| ayoung | dolphm, I am going to submit the "split Identity" in a scaled back version. It won;'t allow multiple LDAP instance, just allow you to mix an identity provider and a assignment provider (assignments being projects, roles etc) | 19:11 |
| ayoung | but the magix is in not breaking existing deployments | 19:11 |
| nati_ueno | pcm__: :) | 19:11 |
| ayoung | dolphm, so if there is no assignment backend specified, it will match the backend to the identity backend | 19:12 |
| *** garyk has quit IRC | 19:12 | |
| ayoung | dolphm, Does that make sense? | 19:12 |
| dolphm | ayoung: i think that makes sense | 19:13 |
| pcm__ | nati_ueno: Can you clarify a bit on the unimplemented connection part? Are you saying more needs to be done in the SS driver as part of the sync() call or higher up at the IPsec driver? | 19:13 |
| nati_ueno | pcm__: OK currently, only vpnservice status is updated | 19:14 |
| *** networkstatic has joined #openstack-dev | 19:14 | |
| nati_ueno | pcm__: https://review.openstack.org/#/c/33148/8/quantum/services/vpn/device_drivers/strongswan_driver.py L293 | 19:14 |
| nati_ueno | pcm__: It checks ipsec status result only | 19:15 |
| nati_ueno | pcm__: We need to update check connection status in future | 19:15 |
| *** boris-42 has joined #openstack-dev | 19:16 | |
| *** jasdeepH has quit IRC | 19:17 | |
| pcm__ | nati_ueno: So that update_status() talks back to the IPSec driver, right? | 19:18 |
| nati_ueno | pcm__: yes | 19:18 |
| *** otherwiseguy has quit IRC | 19:20 | |
| *** martyntaylor has quit IRC | 19:20 | |
| pcm__ | nati_ueno: Does that end up calling RPCProxy base function? What does it update? | 19:21 |
| *** woodspa has quit IRC | 19:21 | |
| nati_ueno | pcm__: yes it using cast method in base class. It update status | 19:22 |
| *** atiwari has joined #openstack-dev | 19:22 | |
| *** kushal has quit IRC | 19:23 | |
| *** kenperkins has quit IRC | 19:23 | |
| *** michchap has quit IRC | 19:23 | |
| *** jyhc_ has joined #openstack-dev | 19:23 | |
| *** vipul is now known as vipul|away | 19:24 | |
| pcm__ | nati_ueno: So after it updates the status, it needs to update the connection status or check the connection status (sorry, did quite follow you sentence) | 19:24 |
| *** jyhc_ has quit IRC | 19:24 | |
| nati_ueno | pcm__: so (1) ss driver should check ipsec status $VPNCONNECTION_ID | 19:24 |
| *** jyhc_ has joined #openstack-dev | 19:25 | |
| nati_ueno | pcm__: (2) ss driver create status_update dict including status for each connection | 19:25 |
| nati_ueno | (3) ss driver sends the status update for ipsec driver | 19:25 |
| nati_ueno | (4) ipsec driver will update DB entry | 19:25 |
| *** jjmb has quit IRC | 19:25 | |
| nati_ueno | pcm__: This is the flow | 19:25 |
| *** jyhc has quit IRC | 19:26 | |
| ayoung | atiwari, your link is not resolving | 19:27 |
| atiwari | 1 sec | 19:27 |
| stevemar | ayoung: over here now | 19:27 |
| pcm__ | nati_ueno: Great! Thanks a bunch. I'll look at the flow diagram, try the manual commands you mentioned, and give a shot at stepping through the sync function so I better understand all the pieces. | 19:27 |
| atiwari | ayoung, please try https://blueprints.launchpad.net/keystone/+spec/serviceid-binding-with-role-definition | 19:27 |
| *** jclift has quit IRC | 19:28 | |
| pcm__ | nati_ueno: It's much less fuzzy now :) | 19:28 |
| stevemar | ayoung is very popular right now :) | 19:28 |
| nati_ueno | pcm__: OK I'll add more docstrings :P | 19:28 |
| *** sdake has joined #openstack-dev | 19:28 | |
| *** sdake has joined #openstack-dev | 19:28 | |
| pcm__ | nati_ueno: Super. Helps us newbies :) | 19:28 |
| ayoung | atiwari, so you want to namesapce roles. Fine by me. | 19:29 |
| nati_ueno | pcm__: you are not newbies. Doing great review :) | 19:29 |
| atiwari | correct | 19:29 |
| pcm__ | nati_ueno: Thanks! Trying my best. Learning a bunch just reviewing and seeing how it goes together. | 19:30 |
| pcm__ | nati_ueno: Appreciate all the help! | 19:30 |
| atiwari | ayoung, did you look the API changes in etherpad ? | 19:30 |
| *** krtyyy has joined #openstack-dev | 19:30 | |
| *** jsavak has joined #openstack-dev | 19:30 | |
| *** krtyyy has left #openstack-dev | 19:31 | |
| *** olaph has quit IRC | 19:31 | |
| *** cliu has quit IRC | 19:31 | |
| *** jyhc has joined #openstack-dev | 19:33 | |
| *** jyhc has joined #openstack-dev | 19:33 | |
| *** joesavak has quit IRC | 19:34 | |
| *** ffio_ has quit IRC | 19:34 | |
| jyhc | I am writing some python scripts to translate XML instance specifications into launching an instance (via the HTTP REST API) | 19:35 |
| jyhc | I am looking for a way to determine whether an instance has started. Any ideas? | 19:36 |
| *** jjmb has joined #openstack-dev | 19:36 | |
| jyhc | Some type of communication back to a Python daemon or script that the instance is up and running | 19:38 |
| *** rony358 has joined #openstack-dev | 19:38 | |
| *** olaph has joined #openstack-dev | 19:39 | |
| *** jasdeepH has joined #openstack-dev | 19:39 | |
| _TheDodd_ | Is there a channel for 'Core Infrastructure' developers? | 19:40 |
| *** leveldoc has joined #openstack-dev | 19:40 | |
| dolphm | _TheDodd_: #openstack-infra | 19:40 |
| _TheDodd_ | dolphm: Thanks for the timely response. | 19:40 |
| leveldoc | is there support in horizon for managing ceph-based object store (radosgw)? I saw a screenshot, but that seemed to be for essex, we're deploying grizzly | 19:41 |
| _TheDodd_ | dolphm: I know that you are quite busy, but is there any chance that you might have time for a code review at: https://review.openstack.org/#/c/34266/ | 19:41 |
| _TheDodd_ | ? | 19:42 |
| *** koolhead17 has joined #openstack-dev | 19:42 | |
| *** drewlander has quit IRC | 19:44 | |
| openstackgerrit | A change was merged to openstack/quantum: Imported Translations from Transifex https://review.openstack.org/34534 | 19:46 |
| openstackgerrit | A change was merged to openstack/quantum: Require greenlet 0.3.2 (or later) https://review.openstack.org/32845 | 19:46 |
| ayoung | atiwari, so the big question I have "Is service the only namespace that makes sense for roles" | 19:46 |
| ayoung | ? | 19:46 |
| ayoung | atiwari, because I would think that a namespace approach should work for anywhere we need to distinguish between to roles named the same thing. We already sort of namespace the role by having the project name as a reference | 19:47 |
| ayoung | atiwari, domains are also treated this way | 19:48 |
| *** dprince has quit IRC | 19:48 | |
| *** eglynn has joined #openstack-dev | 19:48 | |
| ayoung | atiwari, so, maybe the problem isn't "we need to namespace roles for services" but rather "we need to add services to the set of objects by which we scope roles | 19:49 |
| atiwari | ayoung, I think serivceId make sense because roles are basically derived from service | 19:49 |
| ayoung | atiwari, no they are not | 19:49 |
| ayoung | only a subset are | 19:49 |
| ayoung | the rest are service agnostic | 19:49 |
| ayoung | roles in domain and in projects are the same regardless of which service they are executed on | 19:49 |
| *** michchap has joined #openstack-dev | 19:50 | |
| atiwari | e.g. all the nova roles are derived from nove | 19:50 |
| atiwari | same applies for other services | 19:50 |
| *** tzumainn has quit IRC | 19:50 | |
| atiwari | e.g. swift has its own roles | 19:50 |
| atiwari | so I think serviceId names pace make sense as attribute in role definition | 19:51 |
| _TheDodd_ | I've been using 'tox' for my testing before I submit my code for review, are there any other tesing suites that anyone would recommend? | 19:51 |
| _TheDodd_ | eatmydata maybe? | 19:51 |
| atiwari | ayoung, not to mention I BP talks about role definitions | 19:52 |
| ayoung | atiwari, I am not disagreeing with that statement, I am stating that we already have a mechanism for role assignments, and we should extend it to cover your cases | 19:52 |
| ayoung | atiwari, the most common role name is "admin" | 19:52 |
| ayoung | we want to be able to distinguish between "admin for nova" and "admin for demoproject" | 19:53 |
| atiwari | ok, as mentined in the BP, it would be a challenge for PaaS services | 19:53 |
| atiwari | e.g. SQLDb service and MongoDB as service | 19:53 |
| openstackgerrit | A change was merged to openstack/ceilometer: Pin the keystone client version https://review.openstack.org/34573 | 19:54 |
| atiwari | ayoung, I think you are think in role assignment | 19:54 |
| ayoung | atiwari, well, again, unless you are administering them, the role should not be on the whole database, but on the particular namespace within the DBMS | 19:54 |
| ayoung | atiwari, yes, yes I do | 19:54 |
| ayoung | atiwari, I am not saying you are wrong, I am trying to map your BP to what we have now | 19:55 |
| atiwari | we do not want to have name space in assignment | 19:55 |
| atiwari | that looks good | 19:55 |
| ayoung | atiwari, what I thin you want is to have a separate section of the token with service roles in it like this: | 19:55 |
| ayoung | {service: 'nova'; 'role','admin'} | 19:55 |
| ayoung | and then a user can specifically ask for a service token instead of a project token | 19:56 |
| atiwari | eventually | 19:56 |
| ayoung | dolphm, does my suggestion to atiwari match your view of role assignments? | 19:56 |
| atiwari | but to go there we need a name space in role def | 19:56 |
| ayoung | atiwari, not really. | 19:57 |
| ayoung | atiwari, the role is "president' | 19:57 |
| ayoung | there is a big difference between 'president of IBM' and 'president of the USofA | 19:57 |
| ayoung | ' | 19:57 |
| ayoung | but the role name is the same | 19:57 |
| atiwari | that is correct | 19:58 |
| ayoung | atiwari, but what we don't have now is the concept of scoping roles to services. Which is, I think, a valuable concept | 19:58 |
| *** aelkikhia has joined #openstack-dev | 19:58 | |
| dolphm | ayoung: wouldn't the service just be in the token's scope? | 19:58 |
| atiwari | you are adding a name space there with role | 19:58 |
| atiwari | USA and IBM | 19:58 |
| dolphm | ayoung: or even better, the token would be encrypted such that it can only be verified by service X? | 19:59 |
| ayoung | dolphm, yeah, just scoped to the service, not to the project. To distinguish between a project admin and a service admin | 19:59 |
| atiwari | and that is what we want in role def entity | 19:59 |
| ayoung | dolphm, Bridge too far | 19:59 |
| ayoung | atiwari, yes, USA and IBM. One is a Government, one is a corporation. Two different types of groups, just like the distinction between projects and services | 20:00 |
| atiwari | now the question is where to put that distinction | 20:01 |
| atiwari | ? | 20:01 |
| ayoung | atiwari, we add a type of assignement: service_role_assignemnt. I am currently splitting assignements off of identity, into a backend called assignments. It would go into the assignments backend, and a user should be able to request a token with a service scoped role in it | 20:02 |
| *** michchap has quit IRC | 20:02 | |
| ayoung | atiwari, I think that fulfills your requirement. | 20:02 |
| atiwari | ayoung, it seems I am not getting your idea | 20:04 |
| *** mlavalle has joined #openstack-dev | 20:04 | |
| atiwari | there are more issues raised in the BP | 20:04 |
| atiwari | like tooling issue how does assignment wd handle those | 20:05 |
| ayoung | "2. There is no way a middle ware can filter out unwanted roles from token while redirecting request to service (Nova, Swift, etc....), e.g. As mentioned in the bug below, token scoped to a project which has multiple services associated will have all service roles in it. A middle ware which operating on behalf of a service X has no way to filter out roles which not related to service X, unless it make some decision based on desc | 20:05 |
| ayoung | riptive role names." | 20:05 |
| *** xBytez has quit IRC | 20:06 | |
| *** xBytez has joined #openstack-dev | 20:06 | |
| ayoung | atiwari, so, aside from "admin" what roles do you really think are going to conflict? | 20:06 |
| ayoung | roles are quickly going to reflect the set of permissions they encapsulate, like "vm manager" | 20:07 |
| ayoung | or better yet | 20:07 |
| ayoung | "vm restarter" | 20:07 |
| *** joesavak has joined #openstack-dev | 20:07 | |
| atiwari | OK, suppose there are three db PaaS service on top of IaaS | 20:07 |
| ayoung | "a user with the role 'vm restarter' is allowed to restart virtual machines, but not just power them off." | 20:07 |
| atiwari | and every one want to have db-admin | 20:07 |
| ayoung | atiwari, then that role would be assigned to the service. | 20:08 |
| *** networkstatic has quit IRC | 20:08 | |
| ayoung | It is just if you think a role needs to be scoped down to both the service and the project that there is an issue | 20:08 |
| ayoung | either/or is fine | 20:08 |
| ayoung | dm-admin is still really the role 'admin' on the service 'postgres" or 'mongo' | 20:08 |
| *** jsavak has quit IRC | 20:10 | |
| atiwari | ok, but how does some one distinguish admin or PG and mongo | 20:12 |
| atiwari | specially in case of 2 in BP | 20:12 |
| *** SumitNaiksatam has joined #openstack-dev | 20:12 | |
| ayoung | atiwari, it would be the service id. so in the token: | 20:13 |
| *** rony358 has left #openstack-dev | 20:13 | |
| ayoung | {service: 'postgres'; 'role','admin'} | 20:13 |
| ayoung | or better yet | 20:13 |
| ayoung | {service: '1BDEEF145'; 'role','admin'} | 20:13 |
| ayoung | Obviously, "postgres" is too granular a term. Probably needs to be able to be scoped to an endpoint as well as service. | 20:14 |
| *** ajyounge_ has quit IRC | 20:15 | |
| *** terryh has quit IRC | 20:15 | |
| *** ajyounge has joined #openstack-dev | 20:16 | |
| ayoung | atiwari, ponder it. I think it is the right solution, and it suits your needs. It is just adapting your blueprint to the pre-existing abstractions, not fundamentally changing it | 20:17 |
| *** jbresnah has quit IRC | 20:18 | |
| *** vipul|away is now known as vipul | 20:18 | |
| *** jaybuff has joined #openstack-dev | 20:20 | |
| *** mkollaro has joined #openstack-dev | 20:22 | |
| *** jjmb has quit IRC | 20:24 | |
| atiwari | I don't think the issue raised in the BP is addressed by having abstraction in role assignment | 20:24 |
| atiwari | you are looking one part of the issue | 20:24 |
| *** sushils has joined #openstack-dev | 20:28 | |
| *** michchap has joined #openstack-dev | 20:28 | |
| *** terryh has joined #openstack-dev | 20:29 | |
| *** lbragstad has quit IRC | 20:29 | |
| *** ffio has joined #openstack-dev | 20:29 | |
| *** SergeyLukjanov has joined #openstack-dev | 20:29 | |
| *** jbresnah has joined #openstack-dev | 20:30 | |
| *** sarob_ has joined #openstack-dev | 20:31 | |
| *** adalbas has quit IRC | 20:32 | |
| gyee | ayoung, are you saying roles assigning will be a 4 party affair? i.e. assign a "role" to a "user" for a given "project" and "service"? | 20:34 |
| ayoung | gyee, no | 20:34 |
| ayoung | gyee, I was only suggesting that roles could *also* be assigned to services | 20:34 |
| *** sarob has quit IRC | 20:34 | |
| gyee | ayoung, that's what the BP's about, having a role "definition" tied to a service | 20:35 |
| gyee | what am I missing? | 20:35 |
| ayoung | gyee, it is not the role "definition" just the role assignment that needs to change | 20:35 |
| jyhc | When Nova launches an instance, a notification is generated in JSON and then sent to an AMQP queue, correct? | 20:35 |
| *** sarob_ has quit IRC | 20:36 | |
| *** jjmb has joined #openstack-dev | 20:36 | |
| ayoung | gyee, what they are suggesting is a de facto 4 way relationship, but I don;t think that is what they want | 20:36 |
| ayoung | gyee, I think a role should be either on a project, or on a service, not both | 20:36 |
| ayoung | role assignement that is | 20:36 |
| *** dkranz has quit IRC | 20:37 | |
| gyee | ayoung, right now is role assignment is role - project/domain - user/group | 20:37 |
| gyee | oh yeah, inheritance too | 20:37 |
| gyee | are we saying we need to first assign a role to a service, then to a project? | 20:37 |
| ayoung | no | 20:38 |
| *** ffio has quit IRC | 20:38 | |
| ayoung | just that the use cases they have put forth seem to all be "admin for db is different than admin for nova" | 20:38 |
| ayoung | which I agree with | 20:38 |
| gyee | right, I understand the use case | 20:38 |
| ayoung | and thus the solution is "admin on <service>" separate from "admin on <project>" | 20:38 |
| gyee | that's where I got lost | 20:39 |
| atiwari | me too | 20:39 |
| gyee | so we are saying assign a role to a user for a given service, instead of a given project? | 20:40 |
| gyee | still a 3 way affair | 20:40 |
| atiwari | and how it work with inherited role assignment | 20:40 |
| *** michchap has quit IRC | 20:41 | |
| openstackgerrit | A change was merged to openstack/tempest: Add scenario test of instance snapshot and boot https://review.openstack.org/33766 | 20:42 |
| *** otherwiseguy has joined #openstack-dev | 20:42 | |
| *** mrodden has joined #openstack-dev | 20:42 | |
| ayoung | atiwari, my gut reaction is that inheritance is irrelevant to service roles. THere is no grouping for services. BUt I guess endpoints would inherit the roles assigned on the assocaited services | 20:42 |
| *** enikanorov has joined #openstack-dev | 20:43 | |
| ayoung | atiwari, the point you have to get clear is that roles assignments are a 3 way tuple( user, project, rolename) | 20:44 |
| ayoung | services are outside of that grouping | 20:44 |
| *** tmclaugh[work] has left #openstack-dev | 20:44 | |
| gyee | project is essentially a collection of services | 20:44 |
| ayoung | you are suggesting a de factor | 20:44 |
| atiwari | I think we are complicating a simple thing here. Service owns role Def (Nova admin or Swift admin) and they are adding a name space by add service name in role def | 20:44 |
| ayoung | de facto (user, project, service, rolename) | 20:44 |
| atiwari | then what is the issue adding an official name space | 20:45 |
| ayoung | NO! | 20:45 |
| ayoung | gyee, that is wrong | 20:45 |
| ayoung | projects are not collection of services | 20:45 |
| *** jjmb has quit IRC | 20:45 | |
| ayoung | projects are collections of resources. Services operate on the project resources, but those projects span multiple services | 20:45 |
| ayoung | if I have a project named demo (from devstack) there will be aspects of 'demo' stored in nova, glance, quantum, and cinder | 20:46 |
| ayoung | not to mention keystone, of course | 20:46 |
| ayoung | Now, if I need to admin Nova, it is irrrelevant of project | 20:47 |
| openstackgerrit | A change was merged to openstack/trove: Fixes my.cnf not reverting with a resize revert. https://review.openstack.org/31515 | 20:48 |
| *** lbragstad has joined #openstack-dev | 20:48 | |
| openstackgerrit | A change was merged to openstack/nova: Change resource links when url has no project id https://review.openstack.org/31632 | 20:49 |
| openstackgerrit | A change was merged to openstack/nova: Port flavors core API to v3 tree https://review.openstack.org/28825 | 20:49 |
| atiwari | ayoung, what are the other resource a project can have ? | 20:49 |
| openstackgerrit | A change was merged to openstack/nova: Prepare fake instance stubs for objects https://review.openstack.org/33874 | 20:49 |
| atiwari | apart from service | 20:49 |
| openstackgerrit | A change was merged to openstack/nova: Make fake_network tolerant of objects https://review.openstack.org/33875 | 20:50 |
| openstackgerrit | A change was merged to openstack/nova: Imported Translations from Transifex https://review.openstack.org/34603 | 20:50 |
| ayoung | atiwari, VMs,ports, Networks, images, partitions... | 20:50 |
| openstackgerrit | A change was merged to openstack/nova: Cells: Add support for global cinder https://review.openstack.org/31561 | 20:50 |
| gyee | ayoung, those are service resources | 20:51 |
| atiwari | and why they are not a service ? | 20:51 |
| atiwari | Networks is a service | 20:51 |
| gyee | keystone does not have VMs, ports, | 20:51 |
| gyee | keytone knows about quantum, nova, swift, etc | 20:52 |
| atiwari | basically VM are managed by Nova | 20:52 |
| atiwari | and role maps to service | 20:52 |
| atiwari | images are managed by glance | 20:53 |
| *** woodspa has joined #openstack-dev | 20:53 | |
| atiwari | so projects are nothing but collection of services | 20:53 |
| gyee | a collection of service endpoints to be exact :) | 20:54 |
| atiwari | +1 | 20:54 |
| ayoung | No | 20:54 |
| ayoung | the same endpoint spans multiple projects | 20:54 |
| atiwari | the same way we describe domains | 20:54 |
| atiwari | "Domains represent collections of users, groups and projects" | 20:54 |
| gyee | ayoung, I didn't say they an endpoint cannot span multiple projects | 20:55 |
| gyee | or service in that matter | 20:55 |
| ayoung | gyee, I have work to get done. | 20:55 |
| *** spzala has quit IRC | 20:55 | |
| *** joesavak has quit IRC | 20:55 | |
| _TheDodd_ | Does anyone use `eatmydata`, or is `tox` the preferred testing orchistrator? | 20:56 |
| *** stevemar has quit IRC | 20:56 | |
| _TheDodd_ | Does anyone use `eatmydata`, or is `tox` the preferred testing orchistrator? | 20:56 |
| ayoung | I really have no desire to spend time quibbling. Roles are just names. If that role really needs to be deconflicted between what can be done on Nova and what can be done on Glance, then the organuization deploying it needs to deconflict. | 20:56 |
| ayoung | So that argument is speciaous | 20:56 |
| ayoung | if you are trying to distinguish between "admin for nova" and "admin for glance" then we scope a role to a service and or and endpoint | 20:57 |
| ayoung | Go sleep on it, and we can discuss tomorrow | 20:57 |
| *** martine_ has quit IRC | 20:58 | |
| dolphm | _TheDodd_: tox | 20:59 |
| _TheDodd_ | dolphm, that's what I thought, and that's what I've been using. I noticed that `eatmydata ./run_test.sh` was mentioned somewhere and I wasn't sure if it was pertinent. | 21:00 |
| gyee | ayoung, define "scope a role to a service or an endpoint" so I have something to sleep on | 21:00 |
| dolphm | _TheDodd_: i've never actually heard of eatmydata | 21:00 |
| *** jjmb has joined #openstack-dev | 21:00 | |
| *** sandywalsh has quit IRC | 21:01 | |
| *** jasondotstar has quit IRC | 21:01 | |
| _TheDodd_ | dolphm: I just came across it this morning. Do you know if there happens to be any doscumentation on 'tox' and out testing process? I've been trying to get a better feel of how our testing works and how to maximize it etc. | 21:02 |
| dolphm | _TheDodd_: https://wiki.openstack.org/wiki/ProjectTestingInterface | 21:02 |
| *** bknudson has joined #openstack-dev | 21:02 | |
| *** bdpayne has quit IRC | 21:02 | |
| _TheDodd_ | dolphm: Thx. | 21:03 |
| dolphm | _TheDodd_: also checkout whatever commands are available in the tox.ini of the project you're looking at | 21:03 |
| *** topol has quit IRC | 21:05 | |
| *** enikanorov has quit IRC | 21:05 | |
| dolphm | _TheDodd_: eatmydata looks to be a completely different tool http://packages.debian.org/sid/eatmydata | 21:05 |
| dolphm | _TheDodd_: looks like you could run $ eatmydata tox | 21:06 |
| *** pcm__ has quit IRC | 21:06 | |
| _TheDodd_ | dolphm: Interesting. | 21:06 |
| _TheDodd_ | dolphm: Thanks for the links. | 21:07 |
| *** michchap has joined #openstack-dev | 21:07 | |
| atiwari | ayoung, can you please add your thoughts how you want to associate roles with service ? | 21:08 |
| atiwari | may be https://etherpad.openstack.org/serviceid-binding-with-role-definition | 21:08 |
| atiwari | is the right place | 21:08 |
| *** rharwood has quit IRC | 21:11 | |
| *** enikanorov has joined #openstack-dev | 21:15 | |
| *** yolanda has quit IRC | 21:15 | |
| *** mkoderer has quit IRC | 21:16 | |
| dtroyer_zz | dolphm: got a question on identity api version strings... | 21:17 |
| *** HenryG has quit IRC | 21:17 | |
| *** sandywalsh has joined #openstack-dev | 21:17 | |
| *** cliu has joined #openstack-dev | 21:18 | |
| *** sarob has joined #openstack-dev | 21:19 | |
| *** sarob has quit IRC | 21:19 | |
| *** koolhead17 has quit IRC | 21:19 | |
| *** sarob has joined #openstack-dev | 21:19 | |
| *** michchap has quit IRC | 21:20 | |
| *** FunnyLookinHat has quit IRC | 21:20 | |
| *** torandu has quit IRC | 21:22 | |
| *** jayg is now known as jayg|g0n3 | 21:22 | |
| *** pnavarro has quit IRC | 21:23 | |
| *** pnavarro has joined #openstack-dev | 21:23 | |
| *** pnavarro has quit IRC | 21:23 | |
| *** zul has quit IRC | 21:24 | |
| *** kenperkins has joined #openstack-dev | 21:24 | |
| *** morazi has quit IRC | 21:24 | |
| *** FunnyLookinHat has joined #openstack-dev | 21:24 | |
| *** torandu has joined #openstack-dev | 21:26 | |
| *** nati_uen_ has joined #openstack-dev | 21:28 | |
| dolphm | dtroyer_zz: sure | 21:28 |
| dolphm | bknudson: you might be interested in https://review.openstack.org/#/c/34339/2/keystone/policy/backends/rules.py | 21:28 |
| bknudson | that was fast. | 21:29 |
| *** xqueralt-afk is now known as xqueralt | 21:29 | |
| dtroyer_zz | dolphm: It looks like the string 'v3' is always going to be used in a v3 URI? the version returned by curl http://10.130.50.20:35357/ has "id": "v3.0" in it. | 21:30 |
| dtroyer_zz | is that id supposed to represent the actual version being used and I should just hard-conde 'v3' into the URL? | 21:30 |
| *** rnirmal has quit IRC | 21:30 | |
| *** nati_ueno has quit IRC | 21:31 | |
| dolphm | dtroyer_zz: well that's completely awful... | 21:32 |
| *** mrodden has quit IRC | 21:33 | |
| *** lbragstad has quit IRC | 21:33 | |
| dolphm | dtroyer_zz: hmm... you do get back a link to the v3 endpoint | 21:33 |
| *** cliu has quit IRC | 21:33 | |
| dtroyer_zz | yes but two things: a) it has localhost rather than the actual IP as the v2.0 service catalog has, and b) it has 'v3' in it. I don't want to have to parse that to match up the requested version to the available versions. | 21:35 |
| dtroyer_zz | the url is this: "href": "http://localhost:35357/v3/" | 21:35 |
| *** rscottcoyle has quit IRC | 21:35 | |
| dolphm | dtroyer_zz: the 'localhost' bit is configurable via keystone.conf | 21:36 |
| dtroyer_zz | ah, ok. so it can account for proxies and so on | 21:36 |
| dolphm | dtroyer_zz: keystone can't reliably get that info from it's own service catalog | 21:36 |
| *** krtaylor has quit IRC | 21:36 | |
| dolphm | dtroyer_zz: exactly | 21:36 |
| dolphm | dtroyer_zz: i think it's keystone.conf [DEFAULT] admin_endpoint and public_endpoint | 21:37 |
| dolphm | dtroyer_zz: you don't have to parse the URL's though? | 21:38 |
| dtroyer_zz | dolphm: yup. maybe devstack should set those? | 21:38 |
| dolphm | dtroyer_zz: +1 | 21:39 |
| dtroyer_zz | dolphm: I'm trying to figure out the multi-api version stuff in openstackclient, and one of the things I'm running in to is it using v2.0 for auth then v3, is that intentional? | 21:39 |
| dtroyer_zz | and alogn the way I noticed the '3.0' and went sideways | 21:40 |
| dolphm | dtroyer_zz: we had it defaulting the other way around, but ran into a bug with nova + auth_token, so it was reversed | 21:40 |
| _TheDodd_ | I have noticed distincly that https://wiki.openstack.org/wiki/Main_Page does not, by any means, have an exaustive list of wiki pages and sources for developers to go to. Is there a particular location where one could see all of the pertinent wikis, as opposed to generating a massive - yet incomplete - bookmarks folder? | 21:40 |
| dolphm | dtroyer_zz: the URL should *always* be /v3/ (regardless of v3.0 or v3.1) | 21:40 |
| dtroyer_zz | dolphm: ok, that's good to know. This also ties in to making devstack _use_ the v3 api rather than just enable it... | 21:41 |
| dtroyer_zz | v2.0 is hardcoded in stack.sh a couple of places | 21:41 |
| jyhc | Is there any wiki page that shows me how code is organized in, say, Nova? | 21:41 |
| dolphm | dtroyer_zz: and given a version-less endpoint http://localhost:35357/ keystoneclient should figure out whether v2.0, v3.0, or v3.1 is available and operate accordingly | 21:41 |
| jyhc | I'm looking at the github master branch and there's oodles of .py files | 21:41 |
| jyhc | (frankly I am very new to coding in Python) | 21:42 |
| jyhc | I'd just like to see an overview of the code structure of Nova and then I can take a closer look at each part alone | 21:42 |
| *** cliu has joined #openstack-dev | 21:42 | |
| dolphm | jyhc: http://docs.openstack.org/developer/nova/devref/index.html#module-reference | 21:42 |
| *** zul has joined #openstack-dev | 21:42 | |
| dtroyer_zz | dolphm: exactly…it is building the proper endpoint that I'm stumbling on. so if I configure keystone properly the url in the api version structure should be usable. | 21:43 |
| dolphm | dtroyer_zz: should be! | 21:43 |
| openstackgerrit | A change was merged to openstack/nova: Retry quota_reserve on DBDeadlock https://review.openstack.org/33915 | 21:44 |
| *** timello_ has quit IRC | 21:44 | |
| *** mrodden has joined #openstack-dev | 21:44 | |
| dolphm | dtroyer_zz: have you seen gabrielhurley's work on standardizing this stuff across clients? | 21:44 |
| jyhc | dolphm: Thanks. It is kinda lacking though... | 21:44 |
| *** timello_ has joined #openstack-dev | 21:45 | |
| dolphm | jyhc: it's not documenting much more than structural organization, but that's what you were looking for, no? | 21:46 |
| dtroyer_zz | dolphm: I have but haven't dug in to detail yet. hopefully that falls below what I need to handle in osc; it should be in the libs | 21:46 |
| *** mlavalle has quit IRC | 21:46 | |
| dolphm | dtroyer_zz: definitely get his opinion if you start writing code around this :) | 21:46 |
| *** michchap has joined #openstack-dev | 21:46 | |
| dolphm | dtroyer_zz: i'm not sure what he's coded beyond the doc he wrote | 21:47 |
| dtroyer_zz | dolphm: I don't think much. I'd like that to follow alessio's changes pretty closely though | 21:48 |
| jyhc | Does Openstack use twisted in any way? | 21:48 |
| dtroyer_zz | then we only do it once | 21:48 |
| _TheDodd_ | I have noticed distincly that https://wiki.openstack.org/wiki/Main_Page does not, by any means, have an exaustive list of wiki pages and sources for developers to go to. Is there a particular location where one could see all of the pertinent wikis, as opposed to generating a massive - yet incomplete - bookmarks folder of pages that one has come across so far? | 21:49 |
| *** nachi_ has joined #openstack-dev | 21:49 | |
| *** nachi__ has joined #openstack-dev | 21:49 | |
| *** nachi__ has quit IRC | 21:49 | |
| *** mtreinish has quit IRC | 21:50 | |
| dolphm | jyhc: http://docs.openstack.org/developer/nova/devref/threading.html | 21:50 |
| openstackgerrit | A change was merged to openstack/nova: Cells: Pass object for start/stop https://review.openstack.org/33364 | 21:50 |
| openstackgerrit | A change was merged to openstack/swift: Allow SLOs to be made up of other SLOs https://review.openstack.org/30956 | 21:51 |
| openstackgerrit | A change was merged to openstack/nova: Port quota classes extension to v3 API Part 1 https://review.openstack.org/33003 | 21:51 |
| *** mkollaro has quit IRC | 21:52 | |
| dolphm | _TheDodd_: i've never seen the main page lol... googling "openstack [topic]" leads to some obscure wiki.openstack.org page 90% of the time for me | 21:52 |
| *** mlavalle has joined #openstack-dev | 21:53 | |
| *** eglynn has quit IRC | 21:53 | |
| _TheDodd_ | dolphm: Nice. I was afraid that such was the case. I think it would be extremely beneficial to have a page wich would supply a hierarchical listing of topics and pertinent wikis. | 21:55 |
| dolphm | _TheDodd_: i suppose https://wiki.openstack.org/wiki/Special:AllPages | 21:55 |
| dolphm | _TheDodd_: they're sort of paginated (those are 5 pages listed) | 21:56 |
| openstackgerrit | A change was merged to openstack/python-openstackclient: Fix py26 tests: assertDictEqual https://review.openstack.org/32453 | 21:56 |
| *** gmurphy has quit IRC | 21:56 | |
| *** michchap_ has joined #openstack-dev | 21:57 | |
| *** michchap has quit IRC | 21:58 | |
| *** michcha__ has joined #openstack-dev | 21:58 | |
| *** michchap has joined #openstack-dev | 21:59 | |
| *** redbeard2 has quit IRC | 21:59 | |
| *** cliu has quit IRC | 22:01 | |
| *** michchap_ has quit IRC | 22:01 | |
| *** michchap_ has joined #openstack-dev | 22:02 | |
| *** michcha__ has quit IRC | 22:02 | |
| *** martyntaylor has joined #openstack-dev | 22:03 | |
| *** michcha__ has joined #openstack-dev | 22:03 | |
| ayoung | dolphm, http://logs.openstack.org/34161/2/check/gate-python-keystoneclient-pep8/1252/console.html.gz adds a test dependency on httpPretty. Is that OK? | 22:03 |
| *** jergerber has quit IRC | 22:03 | |
| *** michchap has quit IRC | 22:03 | |
| *** aelkikhia has left #openstack-dev | 22:04 | |
| ayoung | http://falcao.it/HTTPretty/ | 22:04 |
| *** michchap has joined #openstack-dev | 22:04 | |
| *** cliu has joined #openstack-dev | 22:04 | |
| *** nati_uen_ has quit IRC | 22:05 | |
| openstackgerrit | A change was merged to openstack/python-keystoneclient: Fix the cache interface to use time= by default. https://review.openstack.org/33849 | 22:05 |
| *** michch___ has joined #openstack-dev | 22:05 | |
| _TheDodd_ | Any keystone pros willing to do a code review at https://review.openstack.org/#/c/34266/? | 22:05 |
| *** kbringard has quit IRC | 22:06 | |
| *** michchap_ has quit IRC | 22:06 | |
| ayoung | _TheDodd_, oooh, I like | 22:07 |
| *** michchap_ has joined #openstack-dev | 22:07 | |
| *** michcha__ has quit IRC | 22:07 | |
| *** michchap has quit IRC | 22:09 | |
| *** xqueralt is now known as xqueralt-afk | 22:09 | |
| *** michch___ has quit IRC | 22:10 | |
| *** spzala has joined #openstack-dev | 22:10 | |
| *** afazekas has quit IRC | 22:10 | |
| _TheDodd_ | ayoung, I appreciate that! It's been waiting for a final review for a while now. | 22:10 |
| ayoung | _TheDodd_, then why didn't you add me as a reviewer? | 22:10 |
| _TheDodd_ | ayoung, lol...my bad. | 22:11 |
| *** michchap_ has quit IRC | 22:11 | |
| *** _TheDodd_ has quit IRC | 22:12 | |
| ayoung | _TheDodd_, btw, I think that Dolph's comment is the more correct solution, although yours seems to work on my machine as well | 22:12 |
| *** martyntaylor has quit IRC | 22:13 | |
| *** jasdeepH has quit IRC | 22:14 | |
| *** diogogmt has quit IRC | 22:14 | |
| *** galstrom is now known as galstrom_zzz | 22:17 | |
| *** jasdeepH has joined #openstack-dev | 22:19 | |
| openstackgerrit | A change was merged to openstack/nova: Adds support for the Indigo Virtual Switch (IVS) https://review.openstack.org/31730 | 22:22 |
| *** HenryG has joined #openstack-dev | 22:23 | |
| *** _TheDodd_ has joined #openstack-dev | 22:25 | |
| *** spzala has quit IRC | 22:25 | |
| openstackgerrit | A change was merged to openstack/swift: TempAuth: don't return 401/403 if .r without .rlistings https://review.openstack.org/34331 | 22:25 |
| openstackgerrit | A change was merged to openstack/nova: Accept is_public=None when listing all flavors https://review.openstack.org/32713 | 22:26 |
| *** lloydde has quit IRC | 22:27 | |
| *** jjmb has quit IRC | 22:29 | |
| *** sarob_ has joined #openstack-dev | 22:30 | |
| *** cliu has quit IRC | 22:31 | |
| *** FunnyLookinHat has quit IRC | 22:32 | |
| *** sarob_ has quit IRC | 22:32 | |
| *** sarob_ has joined #openstack-dev | 22:33 | |
| openstackgerrit | A change was merged to openstack/trove: GuestTimeout needed to be imported. https://review.openstack.org/34621 | 22:33 |
| *** lloydde has joined #openstack-dev | 22:33 | |
| *** sarob has quit IRC | 22:34 | |
| *** SergeyLukjanov has quit IRC | 22:35 | |
| *** atiwari has quit IRC | 22:35 | |
| *** nati_ueno has joined #openstack-dev | 22:36 | |
| jaybuff | when I delete an image from glance is the resource url supposed to eventually return 404? | 22:36 |
| *** jvrbanac has quit IRC | 22:36 | |
| *** gmurphy has joined #openstack-dev | 22:37 | |
| *** michchap has joined #openstack-dev | 22:37 | |
| *** edmund has quit IRC | 22:38 | |
| openstackgerrit | A change was merged to openstack/trove: Fix few missed issues with the reddwarf -> trove rename https://review.openstack.org/34459 | 22:39 |
| jaybuff | looking at the code (and what my system does) it looks like it returns 200 with a json object that has status => DELETED in it. But the tempest.api.compute.images.test_images_oneserver:ImagesOneServerTestJSON.test_create_delete_image tempest test seems to want the resource url to be 404. | 22:39 |
| *** krtaylor has joined #openstack-dev | 22:40 | |
| *** neelashah has quit IRC | 22:44 | |
| *** jecarey_ has quit IRC | 22:44 | |
| *** gongysh has joined #openstack-dev | 22:45 | |
| *** dolphm has quit IRC | 22:45 | |
| *** zhuadl has joined #openstack-dev | 22:45 | |
| *** jjmb has joined #openstack-dev | 22:48 | |
| *** andrewbogott is now known as andrewbogott_afk | 22:48 | |
| *** michchap has quit IRC | 22:50 | |
| *** jimfehlig has quit IRC | 22:50 | |
| *** michchap has joined #openstack-dev | 22:52 | |
| *** michchap has quit IRC | 22:52 | |
| *** mlavalle has quit IRC | 22:53 | |
| *** mrodden has quit IRC | 22:53 | |
| *** michchap has joined #openstack-dev | 22:53 | |
| *** electrichead has quit IRC | 22:55 | |
| *** fifieldt has joined #openstack-dev | 22:56 | |
| *** cliu has joined #openstack-dev | 22:57 | |
| *** jjmb has quit IRC | 22:59 | |
| *** redbeard2 has joined #openstack-dev | 23:00 | |
| *** bknudson has quit IRC | 23:00 | |
| *** aelkikhia1 has joined #openstack-dev | 23:02 | |
| *** otherwiseguy has quit IRC | 23:02 | |
| *** datsun180b has quit IRC | 23:03 | |
| *** aelkikhia1 has quit IRC | 23:03 | |
| *** aelkikhia has joined #openstack-dev | 23:03 | |
| *** pmathews1 has quit IRC | 23:04 | |
| *** _TheDodd_ has quit IRC | 23:04 | |
| *** rcleere has quit IRC | 23:06 | |
| *** nati_ueno has quit IRC | 23:06 | |
| *** shang_ has quit IRC | 23:07 | |
| *** shang_ has joined #openstack-dev | 23:08 | |
| *** jjmb has joined #openstack-dev | 23:08 | |
| *** rcrit has quit IRC | 23:09 | |
| *** jjmb has quit IRC | 23:10 | |
| *** pixelbeat has quit IRC | 23:10 | |
| *** pixelbeat has joined #openstack-dev | 23:12 | |
| *** sarob_ has quit IRC | 23:12 | |
| *** lloydde has quit IRC | 23:12 | |
| *** sarob has joined #openstack-dev | 23:13 | |
| *** lbragstad has joined #openstack-dev | 23:15 | |
| *** nati_ueno has joined #openstack-dev | 23:18 | |
| *** Mandell has quit IRC | 23:19 | |
| *** aelkikhia has left #openstack-dev | 23:20 | |
| *** andrew_plunk has quit IRC | 23:21 | |
| *** noslzzp has quit IRC | 23:21 | |
| *** sdake has quit IRC | 23:23 | |
| *** rcrit has joined #openstack-dev | 23:23 | |
| *** portante has quit IRC | 23:25 | |
| *** cp16net is now known as cp16net|away | 23:28 | |
| *** Oneiroi has quit IRC | 23:28 | |
| *** cp16net|away is now known as cp16net | 23:29 | |
| *** lloydde has joined #openstack-dev | 23:31 | |
| *** lbragstad has left #openstack-dev | 23:31 | |
| *** leizhang has quit IRC | 23:34 | |
| *** devoid has quit IRC | 23:36 | |
| *** Oneiroi has joined #openstack-dev | 23:37 | |
| *** gyee has quit IRC | 23:37 | |
| *** markwash has quit IRC | 23:42 | |
| *** lbragstad has joined #openstack-dev | 23:43 | |
| *** jasdeepH has quit IRC | 23:46 | |
| *** crank_ is now known as crank | 23:49 | |
| *** melwitt has joined #openstack-dev | 23:51 | |
| *** crank is now known as crank_ | 23:51 | |
| *** pabelanger has quit IRC | 23:52 | |
| *** crank_ is now known as crank | 23:52 | |
| *** crank is now known as crank_ | 23:53 | |
| *** crank_ is now known as crank | 23:53 | |
| *** bing_bu has joined #openstack-dev | 23:54 | |
| *** dkranz has joined #openstack-dev | 23:55 | |
| *** jasdeepH has joined #openstack-dev | 23:58 | |
| *** salv-orlando has quit IRC | 23:59 | |
| *** jjmb has joined #openstack-dev | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!