Tuesday, 2017-02-28

*** mudpuppy has quit IRC00:00
*** palendae has quit IRC00:00
*** mancdaz has quit IRC00:00
*** pwnall1337 has quit IRC00:01
*** mhayden has quit IRC00:02
*** toan has quit IRC00:02
*** Guest68651 has joined #craton00:03
*** mancdaz has joined #craton00:07
*** pwnall1337 has joined #craton00:07
*** toan has joined #craton00:07
*** mudpuppy has joined #craton00:10
*** mhayden has joined #craton00:11
*** Guest68651 has quit IRC00:14
*** palendae_ has joined #craton00:17
*** palendae_ is now known as Guest8571800:17
*** VW has joined #craton00:23
*** VW has quit IRC00:27
*** pwnall1337 is now known as zz_pwnall133700:42
*** Syed__ has quit IRC01:05
*** valw has joined #craton02:25
*** VW has joined #craton02:47
*** VW has quit IRC02:52
*** VW has joined #craton03:02
*** zz_pwnall1337 has quit IRC03:04
*** zz_pwnall1337 has joined #craton03:06
*** VW has quit IRC03:09
*** VW has joined #craton03:09
*** tojuvone has quit IRC04:04
*** VW has quit IRC04:35
*** VW has joined #craton04:36
*** VW has quit IRC04:40
*** zz_pwnall1337 is now known as pwnall133705:07
*** valw has quit IRC05:12
*** pwnall1337 is now known as zz_pwnall133705:55
*** tojuvone has joined #craton06:14
*** tojuvone has quit IRC06:56
*** tojuvone has joined #craton06:56
*** zz_pwnall1337 has quit IRC07:58
*** zz_pwnall1337 has joined #craton07:59
*** acabot has joined #craton08:03
suloo/10:00
*** wirehead_ has joined #craton10:18
*** tojuvone has quit IRC11:41
*** tojuvone has joined #craton11:41
*** tojuvone has quit IRC12:02
*** tojuvone has joined #craton12:03
*** johnthetubaguy has quit IRC12:04
*** johnthetubaguy has joined #craton12:14
thomasemo/12:16
-openstackstatus- NOTICE: restarting gerrit to address performance problems13:04
*** ChanServ changes topic to "restarting gerrit to address performance problems"13:04
*** git-harry has quit IRC13:11
*** VW has joined #craton13:12
*** VW has quit IRC13:12
*** VW has joined #craton13:13
*** git-harry has joined #craton13:16
*** VW has quit IRC13:20
sigmavirusfarid: no I have a few keyboards13:31
sigmavirusPoker II, KBv60, and an Ergodox13:31
-openstackstatus- NOTICE: ok gerrit is back to normal13:33
*** ChanServ changes topic to "ok gerrit is back to normal"13:33
*** ChanServ changes topic to "Summit talk: https://www.youtube.com/watch?v=Q-sf12SDR3M || Logs: http://eavesdrop.openstack.org/irclogs/%23craton/latest.log.html || Mon 1500 UTC in #openstack-meeting-4 || Client/ecosystem Tues 1700 UTC || Core Thur 1700 UTC || Resources: https://etherpad.openstack.org/p/Fleet_Management"13:39
-openstackstatus- NOTICE: gerrit is back to normal and I don't know how to use the openstackstaus bot13:39
*** tojuvone has quit IRC13:41
*** tojuvone has joined #craton13:45
*** wirehead_ has quit IRC13:48
*** wirehead_ has joined #craton13:52
*** openstackgerrit has quit IRC14:03
* sigmavirus is so tempted to approve his own work in cratonclient14:03
* thomasem understands14:08
thomasemBy the way... do we expect to never have variables with an underscore prefix to the key?14:08
thomasemLike, one would never have a _key="foo" variable?14:09
*** VW has joined #craton14:23
*** valw has joined #craton14:35
*** VW has quit IRC14:37
*** VW has joined #craton14:38
faridhi all14:39
faridsigmavirus: hadn't seen that ergodox, looks interesting14:40
suloi only have apple keyboard :(14:42
*** git-harry has quit IRC14:42
suloholy crap ... 150 dollars for a keyboard .. this keyboard better type its own python ;)14:44
*** zz_pwnall1337 is now known as pwnall133714:50
sigmavirussulo: better than that, it seriously helps with my RSI14:53
thomasemSeems worthwhile!14:55
*** Guest85718 is now known as palendae14:55
sigmavirusI use the ergodox the most14:55
sigmavirusbut when travelling I use the poker or kbv60 depending on distance travelling14:55
sigmavirusmac keyboards are awful for me14:55
sigmaviruslenovo keyboards are better14:55
thomasemI tried a TypeHeaven at one point... does that count?14:56
thomasemTopre switches14:56
* sigmavirus hasn't heard of that14:56
sigmavirusOne of these? https://www.amazon.com/Topre-Type-Heaven-104-key-Keyboard/dp/B00DGJALYW14:56
thomasemYerp14:57
thomasemEnded up giving it away to a writer friend of mine.14:57
thomasemIt felt fine for regular typing. I hated it for programming.14:57
thomasemWound up missing my $50 Logitech keyboard, lol.14:58
sigmavirusAt $lastjob a coworker had an extra http://www.coolermaster.com/peripheral/keyboards/quickfirerapidred/ that I started using14:59
sigmavirusIt was oddly satisfying but given our open office floor plan, really obnoxious14:59
thomasemHow loud is that?14:59
sigmavirusVERY14:59
thomasemYeah... that's the biggest turn off for me.14:59
thomasemI went with topre in an attempt to avoid that.15:00
sigmavirusMX Cherry Reds are also on the lower end of the force scale and very loud15:00
thomasemGotcha15:00
sigmavirusI did feel better though, typing on that, instead of the apple keyboards we had15:00
sigmavirusSo I did investigation and found quieter switches15:00
sigmavirusNo mechanical switch will be as quiet as a rubberdome, but they're much more comfortable for me15:00
thomasemI've been debating just doing a round of trying out keyboards and taking them back if I don't like them.15:01
sigmavirusThere's also a brand new line of *quiet* cherry switches that are quieter than what I have15:01
thomasemOh wild. I'd like to give that a go.15:01
thomasemGot a link?15:01
sigmavirushttp://www.digitaltrends.com/computing/cherry-americas-cherry-mx-silent-switches-all-mechanical-keyboard-makers/15:01
sigmavirusAlso, you'll need to try the different colors15:02
sigmavirusBecause soem are great for gaming (less force required to activate)15:02
sigmavirusWhile others are better for typing (more force required)15:02
thomasemLol, I'm going to have to rig something up for fast swapping keyboards.15:03
thomasem5pm switch to gaming keyboard.15:03
sigmavirusAlso https://www.youtube.com/user/RhinoFeed is good because he does sound tests15:03
thomasemExcellent15:03
*** git-harry has joined #craton15:03
sigmavirusmost of the keyboards he ends up reviewing are not with silent switches though15:03
sigmavirushttps://www.youtube.com/watch?v=x_4ngJzKY_s is the switch I use in my ergodox15:05
sigmavirushttps://www.youtube.com/watch?v=C_LDmUs5jH015:05
sigmavirus^ sound test with silent keycaps15:05
sigmavirus(not rhinofeed though)15:05
sigmavirusthomasem: keep in mind, this is a deep dark hole15:06
sigmavirusit led to me soldering my own keyboard together15:07
thomasemLOL15:07
sigmavirusI have another unbuilt keyboard in my basement at the moment waiting for me to find time to solder it together too15:08
thomasemI have my doubts I'll go that far. But, that's commendable determination to get exactly what you want. :P15:08
thomasemHahahaha15:08
thomasemThat's awesome15:08
thomasemjimbaker: would you have time to chat about your get/set/delete vars CLI patch this morning?15:26
thomasemI'm struggling with context here on why we're doing this clever sort of [(k,v) for k,v in self.__dict__.items() if key[0] != '_' and key != 'manager]15:27
sigmavirusvery weird15:27
sigmavirusI don't know why we would want to ignore any var15:27
sigmavirusalso15:27
thomasemas opposed to something that doesn't have the limitations where a variable key cannot start with an underscore of be "manager"15:27
sigmavirusI hate that we're using self.__dict__15:28
thomasemYeah, me too.15:28
sigmavirusMaybe we're trying to not accidentally upload implementation attribtues?15:28
thomasemWell, it is a Python class15:28
thomasemIt's trying to be clever using the Python class as the construct for representing a series of variables.15:28
* sigmavirus wonders if that's too clever15:29
sigmavirusWhat if we just had a Variable class to represent a variable?15:29
thomasemSo, now our variables have the same limitations that a Python class does in terms of conventions with private attributes and such.15:29
sigmavirusThen we could do isinstance checks15:29
sigmavirus(which isn't better, but might be less clever)15:29
thomasemYeah, I think that's a lot of the context I'm missing. Just seems awfully roundabout to accomplish this task.15:30
* sigmavirus hasn't looked at that patch yet15:30
thomasemWell, it's essentially applying crud.Resource to Variables also15:30
thomasemSo, you'd have a crud.Resource that has a 'variables' attribute referencing a class Variables(crud.Resource)15:30
thomasemThe thing is, I have resource.variables anyway as a result of getting variables back on a GET /<resource>/<id>15:31
thomasemBut, to set/delete them, it's trying to override the build_url method to add /variables15:31
thomasemhttps://review.openstack.org/#/c/427032/1415:33
*** openstackgerrit has joined #craton15:34
openstackgerritThomas Maddox proposed openstack/python-cratonclient master: WIP: CLI and client support for get/set/delete of resource vars  https://review.openstack.org/42703215:34
thomasemsigmavirus: https://review.openstack.org/#/c/427032/15/cratonclient/v1/variables.py15:35
faridsulo: around to chat with tim ?15:35
sulofarid: i was in the room, waited for 5 mins, then jumped out ;)15:35
sulofarid: joining now15:35
faridwhat15:35
faridok, thanks sulo15:35
*** valw has quit IRC15:36
*** Syed__ has joined #craton15:38
*** valw has joined #craton15:40
*** ediardo has quit IRC15:44
*** ediardo has joined #craton15:46
thomasemSo, yeah. This issue is all fundamentally because we don't know what the keys might be and that doesn't map well to the way we interact with resources in the CLI.15:51
thomasemSince for all resources we get back from the API, we use a class to represent it and use getattr and a white-list of fields to pull the attributes we want.15:51
thomasemIn the same way that we can't begin a variable key with "_" or use "manager" as a key, we also can't do that with any of the resources from our API. We couldn't display a "manager" attribute on a Cloud resource, for instance, because "manager" is already taken up. :)15:53
thomasemI'm not so bothered by not being able to display "_key", but the "manager" one is a bit bothersome.15:53
thomasemAnd all of it is a bit obtuse.15:53
*** wirehead_ has quit IRC15:59
*** wirehead_ has joined #craton16:05
sigmavirusthomasem: yeah I don't like this at all16:15
sigmavirusAlso, really don't like that there's a class that hosts/cells/regions all now inherit from16:16
sigmavirusthat kind of many layers of inheritance will bite us in the future in all likelihood16:16
thomasemYeah, that's what I'm most worried about.16:17
thomasemThe thing is - sure, we can ship it with these limitations short-term, but we _really_ need to refactor this into something more future-proof.16:18
*** jovon has joined #craton16:18
thomasemAnd, of course, my worry is the deadline fast approaching.16:18
jovonhey all, i won't be in attendance for today's meeting because i have a training course to attend16:23
thomasemjovon: thanks for letting us know. Hope training goes well!16:23
jovonthanks thomasem :)16:25
sigmavirusjovon: any updates for us?16:26
jovoni believe RAML with RAMLfication is a good fit craton. However there will be a slight time overhead converting existing docs into RAML16:30
jovonand i stress slight because its mostly a change to the denotation of params16:31
thomasemI think that's okay. Eating a little overhead to get us to automated and better maintained docs is perfectly reasonable.16:40
* sigmavirus also now realizes why a change he made yesterday didn't break the world16:52
sigmavirusthomasem: I get the sense that dusty doesn't want to say much in our meetings =)16:56
thomasemLol "nope."16:58
sulois my clock fast or everyone taking time to get into the next meeting16:58
sigmavirussulo: fast16:59
sulolooks like my clock16:59
sigmavirusMy clock has us at 10:5916:59
suloits odd .. looks like my laptop syncs with time.euro.apple.com and its seems to be few mins off17:01
sigmavirusjimbaker: https://review.openstack.org/#/c/437658/17:05
sigmavirushttps://review.openstack.org/#/c/437659/717:06
sigmavirushttps://review.openstack.org/#/c/438693/117:06
*** valw has quit IRC17:24
openstackgerritIan Cordasco proposed openstack/python-cratonclient master: Update cells commands to use pluggable formatters  https://review.openstack.org/43906717:25
sigmavirusthomasem: https://review.openstack.org/#/q/status:open+(project:openstack/craton+OR+project:openstack/python-cratonclient)17:30
thomasemsigmavirus: perfect. Thank you!17:30
sigmavirusThere's also https://github.com/openstack/gerrit-dash-creator17:31
sigmavirusMakes it easy to build the dashboard https://github.com/openstack/gerrit-dash-creator/blob/master/dashboards/glance.dash17:31
suloso open question17:55
sulowhy dont we resolve vars for cells regions etc17:55
sulonow that we have project -> cloud -> region -> cell17:56
sulomight make sense to follow the pattern to resolve vars everywhere17:56
sigmavirussulo: *shrug*17:58
* sigmavirus wonders if we need that for Friday17:58
openstackgerritIan Cordasco proposed openstack/python-cratonclient master: Update clouds commands to use pluggable formatters  https://review.openstack.org/43908718:03
*** valw has joined #craton18:07
openstackgerritMerged openstack/python-cratonclient master: Reorganize shell based integration tests  https://review.openstack.org/43765818:22
sulois the docker registry hosed due to S3 ?18:26
sigmavirusseems plausible18:29
openstackgerritMerged openstack/python-cratonclient master: Add --format to the client shell  https://review.openstack.org/43765918:31
thomasemfood18:35
sulook it looks like s318:42
suloso i cant do tests right now ... how awesome18:42
sulowith s3 goes half the internet18:43
jimbakersulo, yeah, i switched to direct mode, but now are gates probably will fail18:43
jimbakershould fail18:43
sigmavirusWell it's Amazon. I'm sure it'll be fixed shortly =P18:44
openstackgerritMerged openstack/python-cratonclient master: Update hosts commands to use pluggable formatters  https://review.openstack.org/43869318:47
*** valw has quit IRC18:52
openstackgerritMerged openstack/python-cratonclient master: Update cells commands to use pluggable formatters  https://review.openstack.org/43906718:53
openstackgerritMerged openstack/python-cratonclient master: Update clouds commands to use pluggable formatters  https://review.openstack.org/43908718:54
* sigmavirus isn't taking lunch today btw18:58
*** valw has joined #craton18:58
sigmavirusI need to cut out at 1400 central18:58
openstackgerritIan Cordasco proposed openstack/python-cratonclient master: Update projects commands to use pluggable formatters  https://review.openstack.org/43911619:15
jimbakersigmavirus, ok, all of your patches in the pipeline have been approved. good stuff!19:25
sigmavirusjimbaker: working on the last one now :)19:27
sigmavirusjust ensuring it'll pass tests19:27
jimbakeryep, sounds good19:27
openstackgerritIan Cordasco proposed openstack/python-cratonclient master: Update regions commands to use pluggable formatters  https://review.openstack.org/43912819:31
jimbakerthomasem, i'm going to assign this to you, since it unfortunately is in the variables patch you are working on; https://bugs.launchpad.net/python-cratonclient/+bug/165942819:38
openstackLaunchpad bug 1659428 in Craton's Python Client "CLI requires region id for commands" [Undecided,New]19:38
jimbakermaybe split it out? that would seem to be the best course of action19:38
*** valw has quit IRC19:54
pwnall1337hey guys is cloud is now a required param for creating a host or variable?19:54
pwnall1337cloud_id *19:54
pwnall1337sure is19:59
pwnall1337i guess this is super new, but docs need updated :)19:59
jimbakerpwnall1337, it's required for a host, yes19:59
jimbakeryes, we need that. also we should relax this as a req of the client - just because we denormalize in our schema shouldn't force the client to do the same20:00
jimbakerthomasem, ^^^20:01
jimbakerthomasem, not saying this is work for this week, but one more implication of what you disliked about that schema design. but again, something we can readily hide20:01
pwnall1337updated import script to ensure compatibility with juno -> newton and added cloud_id20:02
pwnall1337for openstack-ansible20:02
jimbakercool20:02
pwnall1337hmm ran into something interesting here20:06
pwnall1337http://10.22.226.230:808020:06
pwnall1337cratan api hanging after querying hosts with import script20:06
pwnall1337started to fail when getting hosts for removal20:07
pwnall1337--truncate*20:07
pwnall1337root@craton-dev:~# curl 127.0.0.1:808020:10
pwnall1337^C20:10
jimbakerpwnall1337, can you put that a gist so it's public?20:10
jimbakeron a gist20:10
pwnall1337https://gist.github.com/pwnall1337/ce707fecdeaa9e1577dc65f6d5aa784e20:13
Syed__n20:14
Syed__n20:14
Syed__n20:14
Syed__n20:14
Syed__oops, sorry20:15
*** harlowja has quit IRC20:15
jimbakerpwnall1337, please file a bug that allows us to reproduce all steps to get here, and we can hopefully find & resolve20:15
pwnall1337looks like a dead tcp connection20:15
*** harlowja has joined #craton20:16
jimbakeri don't believe there's anything in cloud_id per se, but maybe one extra step?20:16
pwnall1337it's not using any kind of multi threading module is it?20:16
pwnall1337greenlet or multiprocessing?20:16
pwnall1337purely single threaded daemon?20:16
jimbakerpwnall1337, that's not how i would expect it's been deployed. in theory, someone could do multithreading, but i expect multi process20:17
jimbakeri.e., absolutely easiest way to deploy. note that we usually run single threaded in terms of the docker setup20:18
jimbakerand no greenlets. at all20:18
jimbakerwe avoided that approach to async code completely20:19
openstackgerritMerged openstack/python-cratonclient master: Update projects commands to use pluggable formatters  https://review.openstack.org/43911620:19
pwnall1337https://bugs.launchpad.net/craton/+bug/166874820:21
openstackLaunchpad bug 1668748 in craton "api service dead locked with newest update that added cloud_id" [Undecided,New]20:21
thomasemjimbaker: no worries. That's not necessarily related to the vars patch, though, is it?20:22
openstackgerritMerged openstack/python-cratonclient master: Update regions commands to use pluggable formatters  https://review.openstack.org/43912820:28
*** valw has joined #craton20:28
thomasemjimbaker: and thank you! Easy enough. :P20:30
*** valw has quit IRC20:32
*** valw has joined #craton20:40
openstackgerritSyed Ahsan Shamim Zaidi proposed openstack/craton master: Update requests for Users and Projects  https://review.openstack.org/42546320:46
openstackgerritThomas Maddox proposed openstack/python-cratonclient master: WIP: CLI and client support for get/set/delete of resource vars  https://review.openstack.org/42703220:52
jimbakerthomasem, no direct relationship. the vars work was fast tracked because we needed to demo craton functionality, and relaxing this specific requirement was helpful for that demo21:06
thomasemjimbaker: gooootcha. cool. Well, no worries, that's an easy one to fix. :)21:06
jimbakernote that we saw the same thing with the cross project stuff that sulo demo'ed yesterday21:07
jimbakerthomasem, cool21:07
sulo?21:07
jimbakersulo, this is with respect to being able to do $ craton hosts-list, and not specify region21:07
sulorgr21:08
openstackgerritThomas Maddox proposed openstack/python-cratonclient master: WIP: CLI and client support for get/set/delete of resource vars  https://review.openstack.org/42703221:08
thomasemYeppers. I'll get a patch in for that.21:08
thomasemDon't you worry21:09
thomasem:P21:09
jimbakerawesome21:19
sigmavirusthomasem: I can tackle the region ID stuff if you want21:19
thomasemsigmavirus: Up to you. Far be it from me to hold up work others can do while I'm working on something else. :P21:20
*** jovon has quit IRC21:22
jimbakerworks for me, simple change, just needs to get in there21:24
openstackgerritThomas Maddox proposed openstack/python-cratonclient master: WIP: CLI and client support for get/set/delete of resource vars  https://review.openstack.org/42703221:33
*** acabot has quit IRC21:33
openstackgerritThomas Maddox proposed openstack/python-cratonclient master: WIP: CLI and client support for get/set/delete of resource vars  https://review.openstack.org/42703221:35
*** valw has quit IRC21:37
openstackgerritsulochan-acharya proposed openstack/craton master: Allow resolved vars in details list calls  https://review.openstack.org/43917521:41
openstackgerritThomas Maddox proposed openstack/python-cratonclient master: WIP: CLI and client support for get/set/delete of resource vars  https://review.openstack.org/42703221:45
openstackgerritThomas Maddox proposed openstack/python-cratonclient master: WIP: CLI and client support for get/set/delete of resource vars  https://review.openstack.org/42703221:46
openstackgerritThomas Maddox proposed openstack/python-cratonclient master: WIP: CLI and client support for get/set/delete of resource vars  https://review.openstack.org/42703221:46
*** acabot has joined #craton21:46
openstackgerritThomas Maddox proposed openstack/python-cratonclient master: WIP: CLI and client support for get/set/delete of resource vars  https://review.openstack.org/42703221:48
openstackgerritThomas Maddox proposed openstack/python-cratonclient master: WIP: CLI and client support for get/set/delete of resource vars  https://review.openstack.org/42703221:49
thomasemSorry for the spam... :\21:49
thomasemWish there was a git review --silent :P21:50
openstackgerritThomas Maddox proposed openstack/python-cratonclient master: WIP: CLI and client support for get/set/delete of resource vars  https://review.openstack.org/42703221:50
thomasemThere's not something like that... is there?21:50
thomasemOr some way to get it to stop spamming channel with my increments?21:50
thomasemAlright. Did some refactoring and added an implementation for projects in ^^21:52
thomasemCurious about thoughts on that approch21:52
thomasemapproach*21:52
openstackgerritThomas Maddox proposed openstack/python-cratonclient master: WIP: CLI and client support for get/set/delete of resource vars  https://review.openstack.org/42703221:59
jimbakerthomasem, ok, will take a look at that patch22:00
thomasemDanke22:12
pwnall1337https://gist.github.com/pwnall1337/661f52376064f89392211db461a1536322:40
pwnall1337following the craton setup docs22:40
pwnall1337it appears my python 2.7, collections  doesn't have chaimmap22:40
pwnall1337it's been moved to it's own module22:41
sulopwnall1337: we are py3 only22:41
pwnall1337gotcha22:41
pwnall1337i was using out of the box python on ubuntu 16.0422:42
pwnall1337ill setup a venv22:42
jimbakerthomasem, so this may be a bit much abstraction in advance, but i think we can do some degree of generalization in https://review.openstack.org/#/c/427032/2623:01
jimbakerespecially since we have v1/<device>/<id>/labels queued up behind this vars support in the client23:02
openstackgerritOpenStack Proposal Bot proposed openstack/craton master: Updated from global requirements  https://review.openstack.org/43921223:16
*** VW has quit IRC23:16
*** VW has joined #craton23:17
*** VW has quit IRC23:17
*** VW has joined #craton23:17
openstackgerritOpenStack Proposal Bot proposed openstack/python-cratonclient master: Updated from global requirements  https://review.openstack.org/43923123:25
*** VW_ has joined #craton23:57

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!