| fungi | clarkb: though the more i'm playing around with it, the more it just devolves into checking the current namespace | 00:02 |
|---|---|---|
| fungi | akin to "'foo' not in dir()" | 00:02 |
| *** senk1 has joined #openstack-infra | 00:03 | |
| clarkb | fungi: let me push what I have, it seems to work, but yeah that is basically what it does | 00:04 |
| clarkb | which still seems fragile as it isn't checking that no instances exist but instead that no instances referred to by 'foo' exist | 00:04 |
| fungi | since get_referrers() on a class seems fall through to __main__ | 00:04 |
| openstackgerrit | Clark Boylan proposed a change to openstack-infra/zuul: Make merger.Repo more context manager like. https://review.openstack.org/55943 | 00:05 |
| clarkb | fungi: ^ | 00:05 |
| *** jhesketh_ has joined #openstack-infra | 00:05 | |
| clarkb | it is transitive and that is where __main__ comes from | 00:06 |
| *** masayukig has quit IRC | 00:09 | |
| *** jsergent has quit IRC | 00:09 | |
| fungi | i see some suggestions for adding a list to the class and then appending a weakref.proxy(self) into it within the initiator | 00:09 |
| fungi | then assert len(ClassName.instances) is 0 or whatever | 00:10 |
| fungi | though that would require mocking or working with a subclass for our case i guess | 00:11 |
| fungi | well, not mocking but monkeypatchig | 00:11 |
| clarkb | fungi: actually there may be a better way | 00:13 |
| clarkb | for obj in get_referents(merger.Repo): if obj isinstance git.Repo then fail | 00:13 |
| *** dstanek has joined #openstack-infra | 00:14 | |
| clarkb | nope nevermind that crosses classes and class instances | 00:14 |
| clarkb | we could just use gc.get_objects and bruteforce | 00:15 |
| fungi | i saw one example of that, and just iterating through the results | 00:15 |
| clarkb | ya | 00:16 |
| fungi | ugly and slow in theory, but in practice probably instantaneous for this purpose | 00:16 |
| clarkb | ya | 00:16 |
| *** rnirmal has quit IRC | 00:17 | |
| *** matsuhashi has joined #openstack-infra | 00:18 | |
| *** dstanek has quit IRC | 00:19 | |
| clarkb | runtime locally is consistent with what I had prior to iterating all objects and type checking them :) | 00:20 |
| openstackgerrit | Tim Daly, Jr. proposed a change to openstack-infra/config: Add the tomograph project to stackforge. https://review.openstack.org/56115 | 00:20 |
| jog0 | FI bug 1225664 appears to be the most frequesnt cause of gate failure | 00:20 |
| jog0 | FYI, and it looks like it *may* be a real infra bug | 00:21 |
| clarkb | jog0: I think it is a devstack config bug | 00:21 |
| jog0 | does that fall under infra or devtack? | 00:21 |
| clarkb | jog0: we should bump the loopback size to 6GB so that it is larger than the swift max_file_size | 00:21 |
| clarkb | devstack | 00:21 |
| jog0 | clarkb: I'll give that a shot | 00:21 |
| clarkb | k | 00:21 |
| jog0 | https://review.openstack.org/56116 | 00:26 |
| jog0 | sdague dtroyer ^ | 00:26 |
| *** nati_ueno has joined #openstack-infra | 00:34 | |
| sdague | jog0: what's the rationale on doing it there vs. devstack-gate? | 00:34 |
| *** pcrews has quit IRC | 00:35 | |
| sdague | not actually challenging that that's the right place to do it, I just want to know your thoughts | 00:35 |
| clarkb | sdague: it should be in devstack and not devstack gate because otherwise devstack deploys broken swift | 00:35 |
| *** fifieldt has quit IRC | 00:35 | |
| sdague | clarkb: always, or just because of some tests we run? | 00:35 |
| clarkb | the other option is to have devstack set the max_file_size in swift to be much smaller (but I like sticking with upstream defaults as much as possible) | 00:36 |
| *** nati_ueno has quit IRC | 00:36 | |
| clarkb | sdague: right now we create a 4GB loopback device but swift thinks it can write 5GB objects | 00:36 |
| sdague | heh, ok | 00:36 |
| sdague | that's justification enough for me | 00:36 |
| *** nati_ueno has joined #openstack-infra | 00:36 | |
| jog0 | sdague: there was another patch that bumed it up from 1 to 4 | 00:37 |
| jog0 | there so just going with that | 00:37 |
| sdague | jog0: can you amend the commit with that justification? | 00:37 |
| *** nati_ueno has quit IRC | 00:37 | |
| sdague | "we create a 4GB device, but swift thinks it can write 5GB, hence fail" | 00:37 |
| sdague | or something equiv | 00:37 |
| *** nati_ueno has joined #openstack-infra | 00:38 | |
| jog0 | sdague: done | 00:38 |
| sdague | heh | 00:39 |
| notmyname | sdague: I think there may be something else going on. ie what's the change here? bumping from 1 to 4 seemed to break things, but using 1 worked? | 00:39 |
| sdague | ok, any idea if dtroyer has been around recently | 00:39 |
| sdague | notmyname: there are more tests now | 00:39 |
| notmyname | sdague: in tempest or in the devstack functests? what kind of more tests? | 00:40 |
| jog0 | notmyname: we also went to using swift as default backend for glance at one point | 00:40 |
| sdague | in tempest | 00:40 |
| openstackgerrit | Clark Boylan proposed a change to openstack-infra/zuul: Make merger.Repo more context manager like. https://review.openstack.org/55943 | 00:40 |
| openstackgerrit | Clark Boylan proposed a change to openstack-infra/zuul: Clone repo if it is not on disk https://review.openstack.org/52689 | 00:40 |
| jog0 | so a lot more usage | 00:40 |
| clarkb | fungi: ^ I think that should do it, adds tests for both changes in the stack too | 00:40 |
| notmyname | jog0: sdague: ah ok. so more glance usage == more stored in "swift" per test run == disk full errors? | 00:40 |
| sdague | notmyname: that's my guess | 00:41 |
| *** rfolco has quit IRC | 00:41 | |
| notmyname | kk | 00:41 |
| jog0 | notmyname: thats my theory | 00:41 |
| jog0 | I wonder if 6GB will be enough in fact | 00:41 |
| * notmyname still thinks sticking to defaults but using a one VM+one replica "cluster" is funny | 00:41 | |
| sdague | jog0: that does mean that we're just going to keep bumping that number, right? | 00:41 |
| clarkb | jog0: good question. it fixes the world from max_file_size perspective but not necessarily if there are many objects | 00:41 |
| clarkb | notmyname: it is "funny" but so is the large ops test | 00:42 |
| clarkb | turns out you can still find many valid bugs this way :) | 00:42 |
| sdague | notmyname: patches welcome for a better set of devstack defaults / config :) | 00:42 |
| jog0 | I figure if this doesn't break things we merge and watch the trends | 00:42 |
| *** nati_ueno has quit IRC | 00:42 | |
| jog0 | and if the bug still occurs but less often (thank you elastic-recheck) we know its at least partially working | 00:42 |
| notmyname | sure sure. I understand the functional usefulness and the importance of the integration testing (absolutely essential). | 00:43 |
| clarkb | gah I ran tox -epy27 but not tox -epep8, I fail | 00:43 |
| notmyname | sdague: always :-). I ran out of time again today to talk to lifeless about a hardware cluster. maybe tomorrow | 00:43 |
| sdague | I guess the question is, is there a way to figure out if it's a space issue? Like query the loop device on fail. | 00:43 |
| jog0 | sdague: swift logs disk is full | 00:44 |
| clarkb | should be able to do a df on the device as it is mounted | 00:44 |
| openstackgerrit | Clark Boylan proposed a change to openstack-infra/zuul: Make merger.Repo more context manager like. https://review.openstack.org/55943 | 00:45 |
| openstackgerrit | Clark Boylan proposed a change to openstack-infra/zuul: Clone repo if it is not on disk https://review.openstack.org/52689 | 00:45 |
| clarkb | fungi: ^ should pass flake8 now | 00:45 |
| jog0 | sdague: the log http://logstash.openstack.org/#eyJmaWVsZHMiOltdLCJzZWFyY2giOiJtZXNzYWdlOlwiTm8gc3BhY2UgbGVmdCBvbiBkZXZpY2VcIiBBTkQgZmlsZW5hbWU6XCJsb2dzL3N5c2xvZy50eHRcIiBBTkQgYnVpbGRfcmVmOlwicmVmcy96dXVsL21hc3Rlci9aZTAyNTA4NDZkN2E2NDE0YzgwMDRlNTVhOWQ2ZDk2OWVcIiIsInRpbWVmcmFtZSI6IjYwNDgwMCIsImdyYXBobW9kZSI6ImNvdW50Iiwib2Zmc2V0IjowLCJ0aW1lIjp7InVzZXJfaW50ZXJ2YWwiOjB9LCJzdGFtcCI6MTM4NDMwMzQ5MzUwOSwibW9kZSI6IiIsImFuYWx5emVfZmllbGQiOiIifQ== | 00:45 |
| jlk | I don't suppose monty is IRCing from a plane somewhere? | 00:49 |
| jlk | mordred: you around by chance? | 00:50 |
| *** sarob has quit IRC | 00:51 | |
| mordred | jlk: whazzup? | 00:51 |
| openstackgerrit | Michael Still proposed a change to openstack-infra/elastic-recheck: Tell people to do a recheck https://review.openstack.org/56118 | 00:51 |
| jlk | mordred: rackspace-auth-openstack -- I believe you're the only one to have ever contributed to it... | 00:52 |
| mordred | jlk: yes. that is correct | 00:52 |
| jlk | mordred: don't suppose you recall what the process is for doing new releases of it | 00:52 |
| jlk | or if you should still be the gateway for it | 00:52 |
| jlk | but I haz a fix for it: https://github.com/rackerlabs/rackspace-auth-openstack/pull/1 | 00:52 |
| mordred | jlk: sure do - python setup.py sdist upload should do the trick | 00:52 |
| jog0 | haha | 00:53 |
| mordred | jlk: you know - if you would like to take it over, I would not be opposed | 00:53 |
| jog0 | mikal: just saw your patch | 00:53 |
| mordred | jlk: I can add you as owner to the pypi entry | 00:53 |
| jog0 | mikal: perhaps pointing to https://wiki.openstack.org/wiki/GerritJenkinsGit#Test_Failures is better since sometimes you need to do a reverify | 00:53 |
| mikal | jog0: wow, that was fast? The recheck patch? | 00:54 |
| jog0 | mikal: yup | 00:54 |
| jlk | mordred: ho boy. I suppose, lack for a better person at the moment, I could do that. | 00:54 |
| jog0 | or maybe something saying we don't want to close the loop on this so a human is still needed | 00:54 |
| mikal | jog0: fair enough, I shall tweak | 00:54 |
| sdague | jog0: so is there nothing in the swift logs, or is it just that they aren't in logstash because they are a different format? | 00:54 |
| jlk | hopefully it won't have to be touched for another year | 00:54 |
| jog0 | sdague: the latter | 00:55 |
| jlk | really hopefully someday it can go the eff away | 00:55 |
| *** wenlock has quit IRC | 00:55 | |
| *** senk1 has left #openstack-infra | 00:55 | |
| jog0 | sdague: or at least that was true at one point | 00:55 |
| mordred | jlk: I could also just take your PR and release it if you want | 00:55 |
| mordred | jlk: I have no personal use for it anymore | 00:55 |
| jlk | mordred: whichever you prefer, I'm easy. | 00:55 |
| clarkb | sdague: jog0: swift logs are sort of mostly in logstash | 00:55 |
| clarkb | the ones in apache combined format are parsed and indexed and whatever goes to syslog is there too | 00:56 |
| clarkb | there is one service that just splats to a file with no timestamps or apparently structure that is ignored | 00:56 |
| clarkb | the proxy service | 00:56 |
| clarkb | everything else is indexed | 00:57 |
| jog0 | clarkb: that may have been in syslog has timestamps so used taht | 00:57 |
| mordred | jlk: let's add you to pypi - that PR is against a rackerlabs repo, which I don't have access to | 00:57 |
| jog0 | and just got in the habbit of doing swift with syslog | 00:57 |
| mordred | jlk: or is there someone else at rax that would be a better pypi owner? | 00:57 |
| jlk | mordred: gotcha. I think whomever was re-arranging some stuff in github decided to make the repo in racker labs. | 00:57 |
| jlk | mordred: I'm fine for now | 00:57 |
| jlk | I can hand it off if somebody better comes along | 00:57 |
| clarkb | jog0: if it is going to syslog today (which it seems like maybe it is) then we are capturing it | 00:57 |
| mordred | jlk: great. what's your pypi username? | 00:58 |
| *** sandywalsh has quit IRC | 00:58 | |
| jlk | great question! | 00:58 |
| jlk | I may not have one. | 00:58 |
| mordred | heh | 00:58 |
| clarkb | https://github.com/thiderman/doge we should run that on all of our servers >_> | 00:59 |
| openstackgerrit | Michael Still proposed a change to openstack-infra/elastic-recheck: Tell people to do a recheck https://review.openstack.org/56118 | 00:59 |
| *** rcleere has quit IRC | 00:59 | |
| sdague | clarkb: :) | 00:59 |
| jlk | arg, now I have to remember my gpg key | 01:00 |
| clarkb | I think putting that one a server that we expect jeblair to log into when he gets back would be lolzy | 01:01 |
| jlk | and remember where I stashed my private key | 01:01 |
| fungi | clarkb: i also had never seen https://github.com/busyloop/lolcat | 01:03 |
| fungi | scary | 01:03 |
| *** dcramer_ has joined #openstack-infra | 01:04 | |
| *** salv-orlando has quit IRC | 01:05 | |
| jlk | mordred: j2solutions | 01:06 |
| jlk | WAT | 01:06 |
| dstufft | if it's for pypi the gpg field there is basically useless | 01:07 |
| *** salv-orlando has joined #openstack-infra | 01:08 | |
| *** gyee has quit IRC | 01:08 | |
| sdague | clarkb: you could get him hyper paranoid that things were hacked :) | 01:08 |
| clarkb | sdague: symlink lolcat to cat and put doge in the motd | 01:09 |
| clarkb | what could go wrong | 01:09 |
| sdague | heh | 01:09 |
| *** markwash has quit IRC | 01:09 | |
| *** nosnos has joined #openstack-infra | 01:13 | |
| *** nosnos has quit IRC | 01:13 | |
| *** nosnos has joined #openstack-infra | 01:14 | |
| jlk | clarkb: evil. | 01:14 |
| jlk | dstufft: oh ok, so I won't have to sign the archives I upload | 01:14 |
| dstufft | you can sign them, but that doesn't validate against that field | 01:14 |
| dstufft | and nothing really bothers to check the sigs are valid | 01:15 |
| dstufft | lolpackaging | 01:15 |
| *** pcrews has joined #openstack-infra | 01:17 | |
| clarkb | dstufft: we should pipe all pip output through lolcat | 01:17 |
| openstackgerrit | Arx Cruz proposed a change to openstack-infra/devstack-gate: Adding support to Fedora and Red Hat https://review.openstack.org/55827 | 01:17 |
| openstackgerrit | Clark Boylan proposed a change to openstack-infra/zuul: Make merger.Repo more context manager like. https://review.openstack.org/55943 | 01:17 |
| openstackgerrit | Clark Boylan proposed a change to openstack-infra/zuul: Clone repo if it is not on disk https://review.openstack.org/52689 | 01:17 |
| dstufft | clarkb: lol | 01:18 |
| clarkb | lifeless: ^ I have addressed your comments, now time for dinner. back later | 01:18 |
| mrodden | i feel like know the person who wrote lolcat without even looking | 01:18 |
| mrodden | i showed someone cowsay at work once, and he was entertained for HOURS after | 01:19 |
| clarkb | mrodden: I got someone to write an IRC bot in haskell that did cowsay | 01:19 |
| jlk | wow... | 01:19 |
| clarkb | it was the greatest thing ever | 01:19 |
| jlk | !! | 01:19 |
| openstack | jlk: Error: "!" is not a valid command. | 01:19 |
| mrodden | clarkb: awesome... | 01:19 |
| mrodden | is that project source available :) | 01:20 |
| clarkb | I think it was rewritten in ruby once he realized execing cowsay in haskell and reading back the stdout was silly | 01:20 |
| clarkb | mrodden: I am not sure | 01:20 |
| * clarkb asks | 01:20 | |
| mrodden | whatever works | 01:20 |
| mrodden | its not like its a terribly complex use case :) | 01:20 |
| openstackgerrit | Arx Cruz proposed a change to openstack-infra/devstack-gate: Adding an option to use qpid instead of rabbit or zeromq https://review.openstack.org/55829 | 01:21 |
| jlk | no, just a terrible use case | 01:21 |
| mrodden | lol | 01:21 |
| clarkb | it had great channel anti flood buffering | 01:21 |
| *** talluri has joined #openstack-infra | 01:21 | |
| *** tian has joined #openstack-infra | 01:22 | |
| *** guohliu has joined #openstack-infra | 01:22 | |
| mrodden | wow the version is even 42.0.x | 01:22 |
| mrodden | awesome | 01:23 |
| *** dcramer_ has quit IRC | 01:23 | |
| mrodden | the license is pretty sweet too | 01:24 |
| jlk | WTFPL ? | 01:24 |
| jlk | love running across that one | 01:24 |
| mrodden | lol | 01:24 |
| mrodden | no | 01:24 |
| mrodden | its the "do what the f**k you want" license | 01:25 |
| mrodden | version 2 | 01:25 |
| mrodden | pretty much MIT/BSD | 01:25 |
| mrodden | :) | 01:25 |
| *** UtahDave has quit IRC | 01:27 | |
| vipul | if you put Author or Co-Authored-By in a commit message.. is that meaningful to any tool? | 01:29 |
| *** guohliu has quit IRC | 01:32 | |
| *** dcramer_ has joined #openstack-infra | 01:37 | |
| *** HenryG has quit IRC | 01:37 | |
| *** dcramer_ has quit IRC | 01:46 | |
| *** talluri has quit IRC | 01:48 | |
| *** melwitt1 has quit IRC | 01:50 | |
| *** dstanek has joined #openstack-infra | 01:53 | |
| *** loq_mac has joined #openstack-infra | 01:54 | |
| *** loq_mac has quit IRC | 01:54 | |
| *** nati_ueno has joined #openstack-infra | 01:55 | |
| lifeless | mordred: https://review.openstack.org/#/c/54628/2/pbr/packaging.py | 01:58 |
| *** yaguang has joined #openstack-infra | 01:58 | |
| jlk | mrodden: that's awfully close to the wtfpl | 02:00 |
| jlk | http://www.wtfpl.net/ | 02:00 |
| clarkb | vipul: I think some tools do, I don't think ours do though | 02:00 |
| vipul | clarkb: ok cool - noticed some patches with that | 02:01 |
| fungi | specifically, i think the activity board may take it into account, but mostly we settled on a standard so that we *can* try to make use of it in the future if we want | 02:02 |
| *** sdake_ has quit IRC | 02:02 | |
| dims | mrodden, ran into the "Don't be evil license"? (http://blog.sonatype.com/people/2012/03/use-json-well-youd-better-not-be-evil/#.UoLdvKpu7mc) | 02:03 |
| *** rcleere has joined #openstack-infra | 02:04 | |
| mrodden | so how do you legally define Good vs Evil | 02:05 |
| mrodden | would be my question :) | 02:05 |
| dstufft | mrodden: hence the problem with that license :] | 02:05 |
| dstufft | well one of the problems | 02:05 |
| fungi | debian has (infamously) decided it's non-free as a result | 02:05 |
| clarkb | the php json package/module/whatever suffers from that | 02:06 |
| clarkb | debian was going to drop it | 02:06 |
| mrodden | opensource is called evil sometimes because, well you know, its sharing | 02:06 |
| mrodden | and sharing is bad | 02:06 |
| fungi | the trollish counterargument being that debian wants to allow its users to be evil if that is their choice | 02:06 |
| mrodden | and stuff | 02:06 |
| mrodden | like ubuntu and cannonical | 02:06 |
| *** senk has joined #openstack-infra | 02:06 | |
| mrodden | not to start a war... | 02:06 |
| dstufft | I argued with cockford about this before | 02:06 |
| fungi | i think he really just wants to argue, and that's the entire point of the license. it's metaphysically transcendent or something | 02:08 |
| *** senk1 has joined #openstack-infra | 02:08 | |
| *** dcramer_ has joined #openstack-infra | 02:08 | |
| mrodden | that makes my head hurt | 02:08 |
| dstufft | I think he either just likes the attention, or he feels like he's somehow making the world better by preventing evil people from using JSON | 02:08 |
| fungi | (paraphrasing) "i'll be happy to grant you a license exemption allowing you to do evil with this software as long as you request it in writing:" | 02:09 |
| *** rcleere has quit IRC | 02:09 | |
| dstufft | even though if you're willing to torture a person i'm not sure why you'd care what a license says | 02:09 |
| *** toddmorey has quit IRC | 02:09 | |
| dstufft | crimes against humanity? A-OK, crimes against intellectual property? That's where we draw the line | 02:09 |
| mrodden | The Open Source Definition. Clause #6 is “No Discrimination Against Fields of Endeavor” | 02:09 |
| *** nati_ueno has quit IRC | 02:09 | |
| mrodden | lol | 02:09 |
| mrodden | so if torturers can't use it then its not open source | 02:10 |
| mrodden | sorry | 02:10 |
| fungi | i sow evil through all my fields of endeavor | 02:10 |
| mrodden | heh | 02:10 |
| *** senk has quit IRC | 02:10 | |
| fungi | it grows surprisingly well | 02:10 |
| *** nati_ueno has joined #openstack-infra | 02:10 | |
| mrodden | anything the plants can do to get ahead | 02:11 |
| *** bingbu has joined #openstack-infra | 02:11 | |
| dims | :) | 02:11 |
| mrodden | i'd better stop before i get in trouble | 02:11 |
| *** toddmorey has joined #openstack-infra | 02:12 | |
| mrodden | oh yeah i had a question | 02:13 |
| mrodden | how does the -infra stuff store its secret pieces (passwords, private keys, etc.) outside of the configurations and scripts | 02:14 |
| mrodden | puppet magic? | 02:14 |
| *** toddmorey has quit IRC | 02:14 | |
| *** toddmorey has joined #openstack-infra | 02:15 | |
| clarkb | hiera | 02:15 |
| *** Ryan_Lane has quit IRC | 02:15 | |
| mrodden | k thats what i thought | 02:15 |
| mrodden | need to read up on that | 02:16 |
| mrodden | i have a need to keep secret pieces in some of my chef recipes, but the encrypted databags still require a secret key, either on system or somewhere else unencrypted | 02:16 |
| dstufft | mrodden: eh | 02:17 |
| mrodden | ? | 02:17 |
| dstufft | mrodden: here's the thing, you're trusting the chef server not to send you arbitrary code | 02:17 |
| dstufft | so trusting them to also not read some data from the unencrypted data bags | 02:17 |
| dstufft | isn't that big of an addiitonal trust | 02:17 |
| *** toddmorey has joined #openstack-infra | 02:17 | |
| mrodden | well the clients are validated... | 02:18 |
| dstufft | but the clients just download cookbooks from the server and then execute them | 02:18 |
| mrodden | right | 02:18 |
| mrodden | but they validate that the server is teh right server i believe | 02:18 |
| mrodden | validate its identity | 02:18 |
| mrodden | and likewise | 02:18 |
| mrodden | client validated to server through its private key | 02:19 |
| mrodden | its more of an ACL problem | 02:19 |
| mrodden | i dont want certain users to be able to get my passwords for another set of systems | 02:19 |
| mrodden | well | 02:20 |
| mrodden | services on those systems | 02:20 |
| dstufft | ah you're worried about other users, not opscode (or whoever is running your chef server) | 02:20 |
| mrodden | right | 02:20 |
| mrodden | i'm running my server... | 02:20 |
| mrodden | its super easy to set one up | 02:20 |
| mrodden | IMO | 02:20 |
| *** mriedem has joined #openstack-infra | 02:21 | |
| mrodden | i guess i could just set a server up for each set | 02:21 |
| mrodden | seems like there should be a better way to do that though | 02:21 |
| *** dcramer_ has quit IRC | 02:22 | |
| *** toddmorey has quit IRC | 02:23 | |
| *** toddmorey has joined #openstack-infra | 02:24 | |
| *** changbl has joined #openstack-infra | 02:24 | |
| *** nosnos_ has joined #openstack-infra | 02:27 | |
| *** guohliu has joined #openstack-infra | 02:28 | |
| mrodden | so apparently hiera isn't much better than the encypted data bags | 02:28 |
| mrodden | :( | 02:28 |
| mrodden | maybe i'm just paranoid | 02:28 |
| *** nati_uen_ has joined #openstack-infra | 02:29 | |
| sdague | mrodden: what's the concern? | 02:29 |
| clarkb | mrodden: it isn't | 02:29 |
| *** nosnos has quit IRC | 02:30 | |
| clarkb | though pupept clients don't send back stuff to be exec'd | 02:30 |
| clarkb | just facts iirc | 02:30 |
| clarkb | so the master is much more isolated | 02:30 |
| sdague | right the master is pretty issolated, and it compiles out the secrets to the clients. Though if you have root on the clients, you get all the passwords that were handed to the client at hand. | 02:31 |
| *** toddmorey has quit IRC | 02:31 | |
| mrodden | that seems bad | 02:32 |
| clarkb | correct, buyt you can;t get other clients' secrets without their keys | 02:32 |
| sdague | yep | 02:32 |
| *** nati_ueno has quit IRC | 02:33 | |
| mrodden | i'm pretty sure with chef, if you have access (and a private key) to a data bag, you can access anything in that databag | 02:33 |
| mrodden | haven't tested it yet | 02:33 |
| sdague | mrodden: so in puppet the client only gets the secrets it specifically needs | 02:34 |
| mrodden | right | 02:34 |
| mrodden | can arbitrarily request stuff from the server | 02:34 |
| mrodden | can't* | 02:34 |
| sdague | right | 02:34 |
| sdague | because the master defines what you get | 02:34 |
| sdague | so it's about as good as you are going to get if your software actually needs passwords to do stuff | 02:35 |
| mrodden | right | 02:35 |
| *** dcramer_ has joined #openstack-infra | 02:35 | |
| mrodden | might be time for some chef-server hacking... | 02:35 |
| sdague | man, just become a cool kid and switch to ansible :) | 02:36 |
| mrodden | haven't played with that one yet | 02:36 |
| mrodden | i have a salt master running 50% of my jenkins slave nodes though | 02:36 |
| *** krotscheck has quit IRC | 02:36 | |
| mrodden | its fast | 02:36 |
| mrodden | but i don't like the way it does configuration management | 02:36 |
| clarkb | ansible and salt seem similar from a cfg mgmt perspective | 02:37 |
| clarkb | data structure based using yaml with strict ordering and so on | 02:37 |
| mrodden | yeah | 02:37 |
| mrodden | its all yaml data | 02:37 |
| sdague | yeh, honestly, I like puppet, even with it's sometimes odd dsl | 02:37 |
| mrodden | and salt has the state files, and then you put stuff in the pillars... | 02:37 |
| *** ljjjustin has joined #openstack-infra | 02:37 | |
| mrodden | brain hurt | 02:38 |
| clarkb | sdague: I really like puppet agent --noop | 02:38 |
| mrodden | actually i gave up because i found it to be very very slow when compiling the state files when they were templates | 02:38 |
| clarkb | which nothing else seems to do as well | 02:38 |
| sdague | yeh | 02:38 |
| mrodden | stub run? | 02:38 |
| sdague | they all have their warts, but at least none of them are cfengine :) | 02:39 |
| mrodden | lol | 02:39 |
| clarkb | mrodden: its a noop run it basically shows you a diff | 02:39 |
| mrodden | yeah | 02:39 |
| mrodden | chef has that | 02:39 |
| clarkb | and it doesn't suffer from cascading failures as much as the others do | 02:39 |
| mrodden | chef-client --why-run | 02:39 |
| sdague | clarkb: https://review.openstack.org/#/q/status:open+project:openstack-infra/devstack-gate+branch:master+topic:devstack_gate_refactor,n,z | 02:39 |
| clarkb | if for example a file is missing or doesn't have the right permissions | 02:39 |
| sdague | so before other devstack-gate changes land, it would be good to process the rest of that series | 02:39 |
| clarkb | sdague: will review after dinner | 02:39 |
| sdague | as that moves the reexec way up to the top | 02:39 |
| sdague | cool | 02:40 |
| sdague | I'm hoping that means we self test much better | 02:40 |
| clarkb | I have salad nd a beer | 02:40 |
| sdague | I made pear and goat cheese crepes, very tasty | 02:40 |
| * mrodden is now hungry | 02:41 | |
| openstackgerrit | A change was merged to openstack-dev/pbr: Avoid using 'which' executable for finding git https://review.openstack.org/55476 | 02:41 |
| *** mrodden has quit IRC | 02:43 | |
| clarkb | sdague: for the first one in the stack it looks like the major changes are lowering case on varaible names and when cd $DEST happens? | 02:45 |
| clarkb | I wish gerrit would let me line up diffs | 02:45 |
| sdague | which review? | 02:46 |
| clarkb | sdague: https://review.openstack.org/#/c/55028/3 | 02:46 |
| sdague | right, so by convention in devstack we use lower case vars when they are local | 02:47 |
| clarkb | gotcha | 02:47 |
| sdague | so you know if you are adjusting global state easier | 02:47 |
| clarkb | I like that | 02:47 |
| sdague | it's also in the google shell style guide IIRC | 02:47 |
| clarkb | sdague: I am not going to approve anything tonight as these changes are not self tested yet | 02:48 |
| clarkb | but I will +2 them and plan on approving first thing tomorrow | 02:48 |
| openstackgerrit | Michael Still proposed a change to openstack-infra/elastic-recheck: Tell people to do a recheck https://review.openstack.org/56118 | 02:48 |
| sdague | sounds good | 02:48 |
| clarkb | assuming mordred/fungi don't beat me to it | 02:48 |
| jog0 | https://review.openstack.org/#/c/55914/ | 02:48 |
| jog0 | looking for a quick review of that e-r patch | 02:48 |
| jog0 | (queries file change) | 02:48 |
| sdague | yep, by the time you get to the end, the exec moves to the top, so as a series they are pretty safe (as they test the one before, until you move the re-exec to the top, then it basically tests it's self) | 02:49 |
| clarkb | jog0: done | 02:49 |
| sdague | ok, eyes getting droopy.... tv then bed | 02:49 |
| clarkb | good night | 02:50 |
| openstackgerrit | A change was merged to openstack-infra/elastic-recheck: Remove fixed bugs from queries.yaml https://review.openstack.org/55914 | 02:51 |
| openstackgerrit | Joe Gordon proposed a change to openstack-infra/elastic-recheck: Add query for bug 1249065 https://review.openstack.org/56138 | 02:54 |
| jog0 | clarkb: not to patch bomb you but ^ there is another query patch | 02:54 |
| *** annegentle has quit IRC | 02:55 | |
| *** annegentle has joined #openstack-infra | 02:55 | |
| *** bnemec has quit IRC | 02:57 | |
| clarkb | jog0: looking | 02:58 |
| *** fifieldt has joined #openstack-infra | 02:59 | |
| clarkb | and approved | 02:59 |
| *** krotscheck has joined #openstack-infra | 02:59 | |
| jog0 | clarkb: thanks | 03:00 |
| clarkb | I like those small query updates, very easy to review :) | 03:01 |
| openstackgerrit | A change was merged to openstack-infra/elastic-recheck: Add query for bug 1249065 https://review.openstack.org/56138 | 03:01 |
| *** CaptTofu has quit IRC | 03:01 | |
| openstackgerrit | Arx Cruz proposed a change to openstack-infra/devstack-gate: Adding an option to use qpid instead of rabbit or zeromq https://review.openstack.org/55829 | 03:02 |
| *** CaptTofu has joined #openstack-infra | 03:02 | |
| *** matsuhashi has quit IRC | 03:02 | |
| *** matsuhashi has joined #openstack-infra | 03:03 | |
| *** toddmorey has joined #openstack-infra | 03:03 | |
| *** matsuhashi has quit IRC | 03:03 | |
| *** rcleere has joined #openstack-infra | 03:03 | |
| *** thomasem has joined #openstack-infra | 03:04 | |
| openstackgerrit | lifeless proposed a change to openstack-infra/reviewstats: Add TOCI as a TripleO review responsibility. https://review.openstack.org/56104 | 03:04 |
| *** nati_uen_ has quit IRC | 03:04 | |
| *** nati_ueno has joined #openstack-infra | 03:05 | |
| *** bnemec has joined #openstack-infra | 03:05 | |
| jog0 | clarkb: heh yeah, and so useful too | 03:05 |
| *** changbl has quit IRC | 03:06 | |
| jog0 | clarkb: is there magic incantation to run recheck x times | 03:06 |
| jog0 | https://review.openstack.org/#/c/56116/ | 03:06 |
| clarkb | jog0: there isn't | 03:07 |
| clarkb | you basically have to babysit it due to the way zuul currently works | 03:07 |
| jog0 | I should make a auto-recheck but for myself | 03:07 |
| *** thomasem has quit IRC | 03:07 | |
| clarkb | you could write a small thing that listens to the ssh event stream that does it on demand | 03:07 |
| clarkb | eg each time results come back send another recheck | 03:07 |
| jog0 | clarkb: yeah I may have half that code already | 03:07 |
| clarkb | :) I think you do have half that code | 03:08 |
| clarkb | maybe more you just need to change the comment message :) | 03:08 |
| *** pcrews has quit IRC | 03:09 | |
| jog0 | hehe | 03:09 |
| *** nati_ueno has quit IRC | 03:10 | |
| *** dcramer_ has quit IRC | 03:10 | |
| *** annegentle has quit IRC | 03:11 | |
| *** annegentle has joined #openstack-infra | 03:12 | |
| *** changbl has joined #openstack-infra | 03:12 | |
| *** matsuhashi has joined #openstack-infra | 03:13 | |
| *** senk1 has quit IRC | 03:14 | |
| *** bingbu has quit IRC | 03:18 | |
| *** bingbu has joined #openstack-infra | 03:18 | |
| *** sdake_ has joined #openstack-infra | 03:20 | |
| *** nati_ueno has joined #openstack-infra | 03:27 | |
| *** yaguang has quit IRC | 03:32 | |
| *** yaguang has joined #openstack-infra | 03:32 | |
| *** dims has quit IRC | 03:37 | |
| *** alexpilotti has joined #openstack-infra | 03:37 | |
| *** markwash has joined #openstack-infra | 03:37 | |
| *** jamesmcarthur has joined #openstack-infra | 03:38 | |
| *** sarob has joined #openstack-infra | 03:42 | |
| *** matsuhashi has quit IRC | 03:45 | |
| *** matsuhashi has joined #openstack-infra | 03:46 | |
| *** masayukig has joined #openstack-infra | 03:47 | |
| *** dcramer_ has joined #openstack-infra | 03:48 | |
| *** matsuhas_ has joined #openstack-infra | 03:49 | |
| *** nati_ueno has quit IRC | 03:49 | |
| *** matsuhashi has quit IRC | 03:50 | |
| *** mriedem has quit IRC | 03:52 | |
| openstackgerrit | A change was merged to openstack-infra/reviewstats: Add TOCI as a TripleO review responsibility. https://review.openstack.org/56104 | 03:54 |
| openstackgerrit | Arx Cruz proposed a change to openstack-infra/devstack-gate: Adding an option to use qpid instead of rabbit or zeromq https://review.openstack.org/55829 | 03:54 |
| *** dstanek has left #openstack-infra | 03:57 | |
| *** sarob has quit IRC | 04:00 | |
| *** nati_ueno has joined #openstack-infra | 04:06 | |
| *** vipul is now known as vipul-away | 04:07 | |
| *** zehicle_at_dell has joined #openstack-infra | 04:09 | |
| *** senk has joined #openstack-infra | 04:11 | |
| *** jerryz has quit IRC | 04:11 | |
| *** senk1 has joined #openstack-infra | 04:15 | |
| *** senk has quit IRC | 04:15 | |
| *** jorisroovers has joined #openstack-infra | 04:17 | |
| *** alexpilotti has quit IRC | 04:18 | |
| *** ArxCruz has quit IRC | 04:19 | |
| *** jorisroovers has quit IRC | 04:22 | |
| *** ArxCruz has joined #openstack-infra | 04:22 | |
| *** wenlock has joined #openstack-infra | 04:25 | |
| *** jamesmcarthur has quit IRC | 04:35 | |
| *** sandywalsh has joined #openstack-infra | 04:39 | |
| *** nosnos_ has quit IRC | 04:44 | |
| *** nosnos has joined #openstack-infra | 04:44 | |
| *** senk1 has quit IRC | 04:44 | |
| *** dawalbha has joined #openstack-infra | 04:47 | |
| *** dawalbha is now known as Nayana | 04:54 | |
| *** Nayana is now known as NayanaD | 04:54 | |
| *** dcramer_ has quit IRC | 04:59 | |
| *** jhesketh_ has quit IRC | 04:59 | |
| *** toddmorey has quit IRC | 05:00 | |
| mikal | Anyone who is around know stuff about jeepyb? | 05:00 |
| anteaya | I know very little | 05:01 |
| mordred | mikal: I wrote it | 05:01 |
| anteaya | but I am awake | 05:01 |
| anteaya | even better | 05:01 |
| mikal | Heh | 05:01 |
| mordred | except for the parts I didn't | 05:01 |
| mikal | So... I want to tweak notify_impact to be a bit fancier | 05:01 |
| mikal | But the tweak I want to do would require a config dile | 05:02 |
| mikal | s/dile/file/. | 05:02 |
| mikal | So, what is an idiomatic config language for those who deploy jeepyb? | 05:02 |
| mikal | And where would I put it? | 05:02 |
| mikal | (Alternatively, is there an example puppet config which deploys it that I can cargo cult?) | 05:02 |
| *** dkliban has joined #openstack-infra | 05:04 | |
| mikal | Ahhh, YAML seems to be the config language of choice? | 05:05 |
| *** bingbu has quit IRC | 05:06 | |
| clarkb | mikal: we use yaml for complicated things and python ini for simple stuff | 05:06 |
| mikal | clarkb: where would the config file live? Is it too horrible to hard code a path? | 05:07 |
| mikal | I guess I could put it in BASE_DIR as defined in that module? | 05:07 |
| *** NayanaD has quit IRC | 05:07 | |
| clarkb | I would put it in /etc/jeepyb or make it configurable | 05:08 |
| mikal | Configurable is hard just because there's not config parser in here at the moment | 05:08 |
| mikal | Oh, I lie! | 05:08 |
| mikal | There is. | 05:08 |
| mikal | Configurable it is. | 05:08 |
| *** NayanaD has joined #openstack-infra | 05:09 | |
| *** bingbu has joined #openstack-infra | 05:11 | |
| mordred | mikal: what clarkb said (sorry, I got distracted) | 05:12 |
| mikal | mordred: is cool | 05:12 |
| mikal | Worst case people can comment on the review when it comes | 05:12 |
| mikal | I'm just flailing around trying to understand how this stuff works | 05:12 |
| mordred | it's the best way | 05:14 |
| mordred | that's mostly how I operate | 05:14 |
| mikal | Is there somewhere I can get an example git log to run notify_impact against to test? | 05:14 |
| mordred | uhm. 'git log' ? | 05:16 |
| mikal | Oh, its not a magical internal format or somethign? | 05:16 |
| * mordred looks at the floor embarrased at the state of our testing | 05:16 | |
| mordred | hrm. actually - not sure | 05:17 |
| mikal | Ahhh, I see | 05:17 |
| mikal | No, its just a git command line | 05:17 |
| mikal | I'm lame for not noticing | 05:17 |
| fifieldt | can you update the docs in notify_impact.py to let others know the way to test? :) | 05:17 |
| mikal | fifieldt: no, I don't work on that team | 05:17 |
| mikal | fifieldt: the union would get angry | 05:17 |
| fifieldt | just inside the file :D | 05:17 |
| fifieldt | c'mon | 05:17 |
| mikal | :P | 05:17 |
| fifieldt | or paste them here | 05:17 |
| fifieldt | and I will deliberately create a merge conflict for you :) | 05:18 |
| mikal | Oh, there's no teamsters available for pasting at the moment | 05:18 |
| mikal | Also, I'm still working it out | 05:18 |
| fifieldt | :D | 05:18 |
| fifieldt | get Josh onto it | 05:18 |
| mikal | Lana is the one for this! | 05:18 |
| fifieldt | also, I can help if you need it | 05:18 |
| *** vipul-away is now known as vipul | 05:19 | |
| mikal | fifieldt: ta | 05:20 |
| mikal | I am intending to hack on this while watching a bad movie on the couch | 05:21 |
| mikal | It being after 5pm and me starting work at 7am and all | 05:21 |
| Loquacity | what have i done now? | 05:21 |
| clarkb | I am playin planetary annihilation poorly | 05:21 |
| mikal | Loquacity: I'm doing that thing for you | 05:23 |
| mikal | Loquacity: I am doing it with vigour and massive incompetence | 05:23 |
| Loquacity | and that's why we love you | 05:23 |
| *** dcramer_ has joined #openstack-infra | 05:25 | |
| mikal | Heh | 05:27 |
| mordred | I, on the other hand, am doing expense reports while sitting on the couch watching the sportball | 05:29 |
| mikal | Ugh, expense reports | 05:30 |
| mikal | Why can't I just be paid infinity money to do whatever the hell I want? | 05:30 |
| anteaya | but the fun of trolling mordred with expense reports | 05:33 |
| anteaya | how could you pass that up? | 05:33 |
| *** markwash has quit IRC | 05:34 | |
| anteaya | mikal: I'm blocked posting to the lists until ttx awakes but here is a first patchset to the governance repo: https://review.openstack.org/#/c/56150/1 | 05:35 |
| anteaya | and thanks | 05:35 |
| *** ericw has quit IRC | 05:41 | |
| *** ericw has joined #openstack-infra | 05:42 | |
| *** nicedice_ has quit IRC | 05:43 | |
| *** senk has joined #openstack-infra | 05:44 | |
| *** pcrews has joined #openstack-infra | 05:45 | |
| *** dkliban has quit IRC | 05:48 | |
| *** ericw has quit IRC | 05:52 | |
| *** matsuhas_ has quit IRC | 05:58 | |
| *** afazekas has joined #openstack-infra | 05:59 | |
| *** blamar has quit IRC | 06:00 | |
| *** matsuhashi has joined #openstack-infra | 06:00 | |
| *** blamar has joined #openstack-infra | 06:00 | |
| *** sarob has joined #openstack-infra | 06:01 | |
| *** matsuhashi has quit IRC | 06:01 | |
| *** matsuhashi has joined #openstack-infra | 06:02 | |
| *** afazekas has quit IRC | 06:04 | |
| *** talluri has joined #openstack-infra | 06:04 | |
| *** sarob has quit IRC | 06:06 | |
| *** denis_makogon has joined #openstack-infra | 06:07 | |
| *** nati_ueno has quit IRC | 06:08 | |
| *** nati_ueno has joined #openstack-infra | 06:09 | |
| *** nati_uen_ has joined #openstack-infra | 06:11 | |
| *** markwash has joined #openstack-infra | 06:12 | |
| *** nati_ueno has quit IRC | 06:13 | |
| *** nati_ueno has joined #openstack-infra | 06:13 | |
| *** nati_uen_ has quit IRC | 06:13 | |
| *** dcramer_ has quit IRC | 06:13 | |
| *** afazekas has joined #openstack-infra | 06:17 | |
| *** wenlock has quit IRC | 06:25 | |
| *** yamahata has joined #openstack-infra | 06:26 | |
| *** boris-42 has joined #openstack-infra | 06:36 | |
| *** pblaho has joined #openstack-infra | 06:38 | |
| *** sarob has joined #openstack-infra | 06:45 | |
| *** matsuhashi has quit IRC | 06:53 | |
| *** matsuhashi has joined #openstack-infra | 06:53 | |
| *** Ryan_Lane has joined #openstack-infra | 06:55 | |
| *** matsuhashi has quit IRC | 06:57 | |
| *** jhesketh has joined #openstack-infra | 07:02 | |
| *** matsuhashi has joined #openstack-infra | 07:06 | |
| *** nati_ueno has quit IRC | 07:07 | |
| *** yolanda has joined #openstack-infra | 07:07 | |
| *** flaper87|afk is now known as flaper87 | 07:08 | |
| *** nati_ueno has joined #openstack-infra | 07:08 | |
| *** nosnos_ has joined #openstack-infra | 07:11 | |
| *** nati_ueno has quit IRC | 07:12 | |
| *** nosnos has quit IRC | 07:14 | |
| *** nati_ueno has joined #openstack-infra | 07:15 | |
| *** sarob has quit IRC | 07:24 | |
| *** rcleere has quit IRC | 07:25 | |
| *** vipul is now known as vipul-away | 07:27 | |
| *** nosnos_ has quit IRC | 07:31 | |
| *** nosnos has joined #openstack-infra | 07:32 | |
| *** senk has quit IRC | 07:42 | |
| *** sjing_ has joined #openstack-infra | 07:44 | |
| sjing_ | Hello | 07:44 |
| *** Loquacity has quit IRC | 07:53 | |
| anteaya | hi sjing_ | 07:53 |
| sjing_ | Hi anteaya, how are you? | 07:54 |
| anteaya | do we have a date for icehouse-1 yet? my wiki.o.o searching didn't turn up anything useful | 07:54 |
| anteaya | sjing_: well though jet laggged | 07:54 |
| anteaya | if you want to have a getting to know you chat we can talk in #openstack-dev | 07:54 |
| anteaya | did you have an -infra question? | 07:54 |
| *** sarob has joined #openstack-infra | 07:55 | |
| *** jhesketh__ has quit IRC | 07:55 | |
| *** vipul-away is now known as vipul | 07:55 | |
| sjing_ | I'm not sure if this is -infra, but where can I get the slides in HK openstack design summit | 07:55 |
| sjing_ | I do find some videos on openstack website, but not the slides | 07:56 |
| sjing_ | thank you in advance | 07:56 |
| anteaya | sjing_: let me see if anything is up yet | 07:57 |
| anteaya | oh umm, slides for the design summit - we use etherpads | 07:57 |
| anteaya | https://wiki.openstack.org/wiki/Summit/Icehouse/Etherpads | 07:58 |
| anteaya | so the design summit is about discussion, and the etherpads are collaborative tools | 07:58 |
| anteaya | does that help you, sjing_? | 07:58 |
| sjing_ | anteaya, sorry, I did not make my statement clear | 08:01 |
| sjing_ | I did not mean the design discussion, I meant the presentations(Y, G, related OSS, etc..) | 08:02 |
| anteaya | sjing_: the design summit sessions don't include slides and presentations | 08:03 |
| anteaya | the etherpad is on the screen and discussion centers on the etherpad | 08:03 |
| anteaya | did you attend design sessions that were actually presentations instead of discussions? | 08:04 |
| sjing_ | anteaya, hehe, I was in the ironic design sessions, so knew that we used etherpad at that time | 08:05 |
| sjing_ | I meant the sessions listed http://openstacksummitnovember2013.sched.org/ | 08:05 |
| *** Loquacity has joined #openstack-infra | 08:06 | |
| *** sarob has quit IRC | 08:06 | |
| sjing_ | I wanted the slides, but only found parts of the videos here http://www.openstack.org/summit/openstack-summit-hong-kong-2013/session-videos/ | 08:08 |
| ttx | anteaya: oh, and btw reed can serve as backup when i'm not around for the tc list moderation | 08:08 |
| *** senk has joined #openstack-infra | 08:10 | |
| *** davidhadas has quit IRC | 08:11 | |
| *** senk1 has joined #openstack-infra | 08:12 | |
| openstackgerrit | Michael Still proposed a change to openstack-infra/jeepyb: Allow automatic subscription to DocImpact bugs https://review.openstack.org/56158 | 08:13 |
| anteaya | ttx great thanks, I will bug reed too | 08:14 |
| *** senk has quit IRC | 08:15 | |
| ttx | anteaya: just approved the queue | 08:15 |
| anteaya | ttx thanks | 08:15 |
| anteaya | sjing_: ah the talks, sorry in my mind the design summit is separate from the talks, my confusion | 08:16 |
| anteaya | not sure who is co-ordinating the uploading of slides from talk presentations | 08:17 |
| anteaya | ttx would you happen to know? | 08:17 |
| anteaya | would that be claire? | 08:17 |
| ttx | anteaya: probably | 08:17 |
| anteaya | okay, so sjing_ I suggest that you email claire@openstack.org and request an estimated time of arrival for access to those slides? | 08:19 |
| openstackgerrit | Oleg Gelbukh proposed a change to openstack-infra/config: Add rubick project to stackforge https://review.openstack.org/53858 | 08:20 |
| anteaya | claire basically organized the entire summit, from a details point of view so she would both probably know and also be very tired | 08:20 |
| anteaya | but she is probably your best person | 08:20 |
| sjing_ | anteaya, thank you so much!! | 08:22 |
| anteaya | my pleasure | 08:22 |
| sjing_ | I will bother claire :) | 08:22 |
| anteaya | and if you get a response, do you mind coming back and telling the rest of us, so we know and don't have to keep bugging claire? | 08:22 |
| sjing_ | sure! | 08:24 |
| anteaya | thanks | 08:26 |
| *** hashar has joined #openstack-infra | 08:26 | |
| anteaya | just post in this channel and I will read the logs | 08:26 |
| sjing_ | ok, I will | 08:30 |
| *** Krast has joined #openstack-infra | 08:31 | |
| *** sarob has joined #openstack-infra | 08:32 | |
| *** osanchez has joined #openstack-infra | 08:33 | |
| anteaya | ttx so far I see that neutron devs can find other neutron devs, some reviewers aren't on irc so I am working to address that | 08:33 |
| anteaya | ttx also there seems to be initial resistance to the idea of idling in channels of other projects and contribution to other projects | 08:33 |
| anteaya | ttx just keeping you aprised of my findings in case the TC has something to say about it | 08:34 |
| anteaya | neutron devs can't find other neutron devs | 08:34 |
| ttx | anteaya: imho at a minimum core devs should have some presence on IRC (and PTLs should use a proxy) | 08:35 |
| anteaya | they couldn't find each other in the design summit sessions, they seem to have difficulty finding each other in channel | 08:35 |
| anteaya | I agree | 08:35 |
| ttx | that sounds like the bare minimum for what they signed up for | 08:35 |
| *** senk has joined #openstack-infra | 08:36 | |
| anteaya | the core devs may be there, I haven't done a thorough investigation | 08:36 |
| *** matsuhashi has quit IRC | 08:36 | |
| anteaya | just initial impressions, but I might be wrong | 08:37 |
| *** senk1 has quit IRC | 08:37 | |
| anteaya | will continue to observe, dig and address as I am able | 08:37 |
| *** matsuhashi has joined #openstack-infra | 08:37 | |
| *** sarob has quit IRC | 08:37 | |
| anteaya | by PTLs should use a proxy, do you mean an irc server, or a person as proxy? | 08:37 |
| *** NayanaD has quit IRC | 08:38 | |
| *** pblaho has quit IRC | 08:38 | |
| *** senk has quit IRC | 08:40 | |
| *** denis_makogon has quit IRC | 08:40 | |
| *** NayanaD has joined #openstack-infra | 08:40 | |
| *** sjing_ has quit IRC | 08:41 | |
| anteaya | ttx ^ | 08:43 |
| ttx | anteaya: no, I mean a IRC proxy so that they stay connected and can see questions when they get up. | 08:44 |
| ttx | (like znc or bip, or a permanent irssi setup) | 08:44 |
| *** jorisroovers has joined #openstack-infra | 08:45 | |
| *** davidhadas has joined #openstack-infra | 08:45 | |
| Krast | why not email？ | 08:46 |
| anteaya | ttx great thanks | 08:47 |
| anteaya | Krast: we use IRC, email is for larger conversations | 08:47 |
| anteaya | there is an expectation that if I pop into a channel and have a question, the PTL (at least) will read backscroll when they can and answer me | 08:48 |
| anteaya | saves waiting and makes for a faster process for working collaboratively with so many projects | 08:48 |
| ttx | anteaya: you might be interested to find solutions for https://bugs.launchpad.net/openstack-ci/+bug/1250758 | 08:49 |
| ttx | did we lose soren's bot again ? | 08:49 |
| * anteaya clicks | 08:49 | |
| ttx | bug 1250758 | 08:49 |
| anteaya | seems we did | 08:49 |
| anteaya | what time zone is he in? | 08:49 |
| ttx | he shall be up any time now | 08:50 |
| *** SergeyLukjanov has joined #openstack-infra | 08:50 | |
| anteaya | okay I am going to go to bed again soon | 08:50 |
| anteaya | yeah, I can look at 1250758 with fungi mordred and the gang once they are up | 08:51 |
| ttx | sounds like an interesting side-project | 08:51 |
| ttx | used to be rare to update calendar, now it's kike every 2 days | 08:51 |
| anteaya | I think mordred was hoping owncloud would solve the calendar issue, but I don't think it is possible to modify a group calendar | 08:51 |
| ttx | like* | 08:51 |
| anteaya | no kidding | 08:51 |
| anteaya | yeah, not sure how many sides I have what with deciding to join neutron and all | 08:52 |
| anteaya | but whining about lack of time doesn't make me unique so I will address it as best I can | 08:52 |
| *** NayanaD has quit IRC | 08:54 | |
| anteaya | Krast: hi I am on weechat and am just getting used to it | 08:55 |
| anteaya | I am not sure how to accept what you are trying to send me | 08:55 |
| anteaya | would a paste or an etherpad work instead? | 08:55 |
| *** nati_ueno has quit IRC | 08:55 | |
| Krast | - - | 08:55 |
| Krast | I use hexchat in windows | 08:56 |
| Krast | I don't know how to use.... bad... | 08:56 |
| anteaya | what were you trying to send? | 08:58 |
| Krast | <anteaya> sorry I just try the new irc client... | 08:58 |
| anteaya | ah okay | 08:59 |
| anteaya | I am off to try to get more sleep | 08:59 |
| anteaya | talk to you later | 08:59 |
| Krast | ok~~see you～ | 09:00 |
| *** sarob has joined #openstack-infra | 09:03 | |
| *** jpich has joined #openstack-infra | 09:06 | |
| *** sarob has quit IRC | 09:08 | |
| *** bingbu has quit IRC | 09:08 | |
| *** matsuhashi has quit IRC | 09:08 | |
| *** matsuhashi has joined #openstack-infra | 09:11 | |
| *** johnthetubaguy has joined #openstack-infra | 09:13 | |
| *** yassine has joined #openstack-infra | 09:14 | |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Document what it takes to be a check/gate test. https://review.openstack.org/55761 | 09:18 |
| *** yassine_ has joined #openstack-infra | 09:18 | |
| *** plomakin has quit IRC | 09:19 | |
| *** Ryan_Lane has quit IRC | 09:26 | |
| *** SergeyLukjanov is now known as _SergeyLukjanov | 09:26 | |
| *** boris-42 has quit IRC | 09:27 | |
| *** _SergeyLukjanov has quit IRC | 09:27 | |
| *** derekh has joined #openstack-infra | 09:28 | |
| *** boris-42 has joined #openstack-infra | 09:29 | |
| openstackgerrit | Mehdi Abaakouk proposed a change to openstack-infra/devstack-gate: Sqlalchemy pool settings for ceilometer https://review.openstack.org/56173 | 09:30 |
| *** SergeyLukjanov has joined #openstack-infra | 09:32 | |
| *** nosnos_ has joined #openstack-infra | 09:33 | |
| *** nosnos has quit IRC | 09:37 | |
| *** jorisroovers has quit IRC | 09:40 | |
| *** senk has joined #openstack-infra | 09:41 | |
| *** jhesketh has quit IRC | 09:43 | |
| *** matsuhashi has quit IRC | 09:45 | |
| *** matsuhashi has joined #openstack-infra | 09:46 | |
| *** matsuhashi has quit IRC | 09:51 | |
| *** matsuhashi has joined #openstack-infra | 09:51 | |
| *** che-arne has quit IRC | 09:53 | |
| *** evgeniyl has joined #openstack-infra | 09:58 | |
| *** yamahata has quit IRC | 10:00 | |
| *** rfolco has joined #openstack-infra | 10:01 | |
| *** boris-42 has quit IRC | 10:02 | |
| *** boris-42 has joined #openstack-infra | 10:02 | |
| *** sarob has joined #openstack-infra | 10:04 | |
| *** Krast has quit IRC | 10:07 | |
| *** davidhadas has quit IRC | 10:07 | |
| *** Krast has joined #openstack-infra | 10:07 | |
| *** Krast has quit IRC | 10:07 | |
| *** davidhadas has joined #openstack-infra | 10:07 | |
| *** sarob has quit IRC | 10:08 | |
| *** ArxCruz has quit IRC | 10:08 | |
| *** jhesketh has joined #openstack-infra | 10:12 | |
| *** SergeyLukjanov is now known as _SergeyLukjanov | 10:13 | |
| *** _SergeyLukjanov has quit IRC | 10:13 | |
| *** ljjjustin has quit IRC | 10:19 | |
| *** senk has quit IRC | 10:25 | |
| *** jhesketh__ has joined #openstack-infra | 10:29 | |
| openstackgerrit | Joshua Hesketh proposed a change to openstack-infra/config: Fix typo in sqlalchemy-migrate upstream path https://review.openstack.org/56178 | 10:31 |
| openstackgerrit | Joshua Hesketh proposed a change to openstack-infra/config: Reorder gerrit projects alphabetically https://review.openstack.org/56179 | 10:31 |
| openstackgerrit | Joshua Hesketh proposed a change to openstack-infra/config: Add turbo-hipster into stackforge https://review.openstack.org/56180 | 10:31 |
| openstackgerrit | Joshua Hesketh proposed a change to openstack-infra/config: Add turbo-hipster to zuul+jenkins https://review.openstack.org/56181 | 10:31 |
| openstackgerrit | Russell Bryant proposed a change to openstack-infra/reviewstats: Add received reviews to review stats https://review.openstack.org/55999 | 10:32 |
| *** marun has joined #openstack-infra | 10:34 | |
| *** SergeyLukjanov has joined #openstack-infra | 10:35 | |
| *** matsuhashi has quit IRC | 10:36 | |
| *** matsuhashi has joined #openstack-infra | 10:36 | |
| *** SergeyLukjanov has quit IRC | 10:37 | |
| *** matsuhashi has quit IRC | 10:41 | |
| *** yfried has joined #openstack-infra | 10:41 | |
| *** matsuhashi has joined #openstack-infra | 10:49 | |
| *** beagles has quit IRC | 10:52 | |
| *** ArxCruz has joined #openstack-infra | 10:54 | |
| *** pblaho has joined #openstack-infra | 10:54 | |
| openstackgerrit | Michael Still proposed a change to openstack-infra/jeepyb: Allow automatic subscription to DocImpact bugs https://review.openstack.org/56158 | 10:54 |
| *** boris-42 has quit IRC | 10:54 | |
| *** matsuhashi has quit IRC | 10:56 | |
| *** matsuhashi has joined #openstack-infra | 10:57 | |
| *** devvesa has joined #openstack-infra | 10:58 | |
| *** talluri has quit IRC | 11:00 | |
| *** matsuhas_ has joined #openstack-infra | 11:00 | |
| *** matsuhashi has quit IRC | 11:00 | |
| *** talluri has joined #openstack-infra | 11:03 | |
| *** sarob has joined #openstack-infra | 11:04 | |
| *** beagles has joined #openstack-infra | 11:07 | |
| *** sarob has quit IRC | 11:09 | |
| *** giulivo has joined #openstack-infra | 11:11 | |
| giulivo | hi guys, I'm exploring job-builder for non-openstack related tasks and it is great fun | 11:11 |
| *** SergeyLukjanov has joined #openstack-infra | 11:12 | |
| giulivo | but I don't seem to find any option capable of setting the "Restrict where this project can be run" properties | 11:12 |
| giulivo | can you suggest how to set that option? | 11:12 |
| ekarlso | uhm, is www.openstack.org down ? | 11:14 |
| * fifieldt can't get to the admin interface | 11:15 | |
| ogelbukh | http://www.downforeveryoneorjustme.com/www.openstack.org | 11:15 |
| ogelbukh | it suggests it's not just you | 11:15 |
| ekarlso | ok :p | 11:16 |
| *** SergeyLukjanov has quit IRC | 11:16 | |
| fifieldt | the server is up | 11:18 |
| fifieldt | but it looks like silverstripe is down | 11:18 |
| *** pcm_ has joined #openstack-infra | 11:18 | |
| *** SergeyLukjanov has joined #openstack-infra | 11:18 | |
| *** SergeyLukjanov has quit IRC | 11:19 | |
| *** guohliu has quit IRC | 11:20 | |
| *** senk has joined #openstack-infra | 11:21 | |
| *** pabelanger has quit IRC | 11:21 | |
| evgeniyl | Hi, I have a question regarding launchpad + gerrit intergration for stackforge projects, e.g. we had a bug https://bugs.launchpad.net/fuel/+bug/1248960 and then made a patch https://review.openstack.org/#/c/55552/ , patch was merged, but bot didn't update bug's status. | 11:24 |
| *** pcm_ has quit IRC | 11:25 | |
| *** pcm_ has joined #openstack-infra | 11:25 | |
| *** che-arne has joined #openstack-infra | 11:26 | |
| *** plomakin has joined #openstack-infra | 11:27 | |
| *** lcestari has joined #openstack-infra | 11:30 | |
| *** SergeyLukjanov has joined #openstack-infra | 11:35 | |
| *** adalbas has joined #openstack-infra | 11:38 | |
| openstackgerrit | Nikita Konovalov proposed a change to openstack-infra/storyboard: Pagination added to task lists https://review.openstack.org/55813 | 11:42 |
| pblaho | evgeniyl: bug is on Fix committed.... it will be closed when commit gets to release | 11:42 |
| pblaho | evgeniyl: it bothers me too... as I usually have to custom search and switch off Fix committed state to see what to work on | 11:43 |
| openstackgerrit | Joshua Hesketh proposed a change to openstack-infra/config: Add turbo-hipster to zuul+jenkins https://review.openstack.org/56181 | 11:44 |
| evgeniyl | pblaho, we moved it to "Fix committed" by hands, because bot didn't do it. | 11:44 |
| *** HenryG has joined #openstack-infra | 11:45 | |
| *** SergeyLukjanov has quit IRC | 11:45 | |
| *** nosnos_ has quit IRC | 11:52 | |
| pblaho | evgeniyl: aah... I have encountered some issues in the past too... check names of project in gerrit and in LP | 11:52 |
| pblaho | evgeniyl: and I remember some issue with project not being in proper group or something in LP | 11:53 |
| *** nosnos has joined #openstack-infra | 11:53 | |
| *** nosnos has quit IRC | 11:53 | |
| *** nosnos has joined #openstack-infra | 11:54 | |
| *** senk has quit IRC | 11:54 | |
| ogelbukh | evgeniyl: https://launchpad.net/fuel/+edit - set "Part of" to 'openstack' | 11:54 |
| evgeniyl | ogelbukh, it's already a part of 'openstack', you can see it on the main page https://launchpad.net/fuel "Part of: OpenStack" | 11:56 |
| *** matsuhas_ has quit IRC | 11:58 | |
| *** nosnos has quit IRC | 11:58 | |
| ogelbukh | evgeniyl: then add 'stackforge/fuel': 'fuel' in this map: https://github.com/openstack-infra/jeepyb/blob/master/jeepyb/projects.py#L134 | 11:58 |
| *** matsuhashi has joined #openstack-infra | 11:59 | |
| *** nosnos has joined #openstack-infra | 11:59 | |
| openstackgerrit | Russell Bryant proposed a change to openstack-infra/reviewstats: Add some per day stats https://review.openstack.org/56200 | 12:00 |
| ArxCruz | mordred: https://review.openstack.org/#/c/55827/ doesn't seems to be an error in my patch what should I do? | 12:00 |
| evgeniyl | ogelbukh, thanks, I'll try. | 12:00 |
| *** matsuhashi has quit IRC | 12:03 | |
| *** yaguang has quit IRC | 12:03 | |
| *** nosnos has quit IRC | 12:04 | |
| *** sarob has joined #openstack-infra | 12:05 | |
| *** sarob has quit IRC | 12:09 | |
| *** CaptTofu has quit IRC | 12:18 | |
| *** CaptTofu has joined #openstack-infra | 12:18 | |
| *** hashar has quit IRC | 12:20 | |
| *** sandywalsh has quit IRC | 12:23 | |
| *** fifieldt has quit IRC | 12:25 | |
| *** masayukig has quit IRC | 12:28 | |
| ttx | ha. ha. ha. http://www.openstack.org/ is down. | 12:29 |
| *** masayukig has joined #openstack-infra | 12:30 | |
| *** yassine_ has quit IRC | 12:33 | |
| *** yassine has quit IRC | 12:33 | |
| *** yassine has joined #openstack-infra | 12:33 | |
| *** masayukig has quit IRC | 12:34 | |
| anteaya | ttx yup | 12:35 |
| anteaya | I can hear mordred exploding now | 12:35 |
| anteaya | which is good | 12:35 |
| *** sandywalsh has joined #openstack-infra | 12:35 | |
| anteaya | I am in the process of trying to get -neutron to actually consume info that will help them | 12:36 |
| anteaya | o.o disappearing does not help me | 12:36 |
| *** masayukig_mob has joined #openstack-infra | 12:37 | |
| openstackgerrit | Evgeniy L proposed a change to openstack-infra/jeepyb: Map all fuel stackforge repos to fuel at launchpad https://review.openstack.org/56204 | 12:38 |
| sdague | ttx: that's still not under infra control, right? | 12:38 |
| ttx | sdague: no its' not | 12:38 |
| sdague | nice | 12:38 |
| *** dims has joined #openstack-infra | 12:39 | |
| ttx | sdague: mordred was complaining about it just yesterday. Now like anteaya said he must be ever more pissed | 12:39 |
| mordred | I have tweeted to collier again | 12:39 |
| mordred | which is the best way I know of doing anything | 12:39 |
| ttx | nice timing for an embarassing outage | 12:39 |
| ttx | mordred: did you DoS it to prove your point ? | 12:40 |
| *** masayukig_mob has quit IRC | 12:40 | |
| mordred | sigh | 12:41 |
| mordred | no | 12:41 |
| mordred | it appears capable of breaking all by itself | 12:41 |
| *** yaguang has joined #openstack-infra | 12:42 | |
| anteaya | well I may as well make some breakfast | 12:43 |
| anteaya | I'm sure a teachable moment for -neutron devs will present itself again in the future | 12:43 |
| *** yaguang has quit IRC | 12:44 | |
| *** yaguang has joined #openstack-infra | 12:44 | |
| *** masayukig has joined #openstack-infra | 12:47 | |
| *** masayukig has quit IRC | 12:47 | |
| *** jamesmcarthur has joined #openstack-infra | 12:47 | |
| *** nati_ueno has joined #openstack-infra | 12:50 | |
| ttx | not sure if fungi got the opportunity to raise the issue at the call yesterday | 12:50 |
| *** hashar has joined #openstack-infra | 12:51 | |
| sdague | stackforge question, if I wanted a new repo that's currently doesn't exist, can I get jeepyb to build it from whole cloth, or do I always need a starting point to import? | 12:52 |
| dims | sdague, i may have stabilized gate-tempest-devstack-vm-neutron-large-ops - 4 straight SUCCESS runs so far - https://review.openstack.org/#/c/56075/ | 12:54 |
| sdague | dims: nice | 12:54 |
| *** senk has joined #openstack-infra | 12:55 | |
| *** masayukig has joined #openstack-infra | 12:55 | |
| mordred | sdague: you can get jeepyb to build it whole cloth - but you will right now just get an empty repo with a gitreview in it | 12:57 |
| mordred | sdague: however, I have a tdl item to get jeepyb to run cookiecutter to make you a new repo | 12:57 |
| mordred | gimme a few minutes and I can hack up that patch and you can be a guinea pig | 12:57 |
| *** masayukig has quit IRC | 12:59 | |
| *** masayukig has joined #openstack-infra | 12:59 | |
| sdague | mordred: this isn't a python project, so empy is fine | 12:59 |
| mordred | sdague: ok. just don't put in an upstream and you'll be set | 13:00 |
| *** sc68cal has joined #openstack-infra | 13:00 | |
| sdague | cool | 13:00 |
| sdague | yeh, this is the idea I was pitching to jd__ this morning about an os-emacs project that had some common python config setup that emacs users might like to bootstrap with | 13:01 |
| dstufft | does it include installing vim? ;P | 13:03 |
| * dstufft hides | 13:03 | |
| dims | dstufft, lol | 13:04 |
| *** masayukig has quit IRC | 13:04 | |
| *** talluri has quit IRC | 13:04 | |
| *** masayukig has joined #openstack-infra | 13:04 | |
| jd__ | I think the purpose is to be efficient, so no dstufft :-) | 13:04 |
| *** masayukig has quit IRC | 13:05 | |
| *** masayukig has joined #openstack-infra | 13:05 | |
| *** sarob has joined #openstack-infra | 13:05 | |
| dstufft | jd__: :D | 13:06 |
| dstufft | jd__: I mostly joke, I don't use either ;P | 13:06 |
| jd__ | dstufft: what do you use then? ed? :) | 13:07 |
| dstufft | well I can use vim I suppose, I started emacs once, i'm pretty sure it's still running on a shell somewhere waiting for me to figure out how to exit it | 13:07 |
| anteaya | ttx so if this group just edited the wiki with their meeting time: https://wiki.openstack.org/wiki/Meetings#Neutron_IPv6_sub-team_Meeting | 13:07 |
| *** talluri has joined #openstack-infra | 13:07 | |
| anteaya | how does their meeting time get added to the ical feed? | 13:07 |
| anteaya | please teach me | 13:07 |
| sc68cal | +1 - since I did the edit to the wiki | 13:07 |
| anteaya | so I can do a better job fixing that bug | 13:07 |
| anteaya | :D | 13:07 |
| sc68cal | hate to consume people's time for things that i should know, but don't. | 13:07 |
| *** jhesketh has quit IRC | 13:07 | |
| *** jhesketh__ has quit IRC | 13:08 | |
| *** dprince has joined #openstack-infra | 13:08 | |
| anteaya | sc68cal: it is the only way to learn, so hopefully we will both learn it | 13:09 |
| anteaya | biggest part is to learn well and then share | 13:09 |
| anteaya | always chances to share | 13:10 |
| sc68cal | +1 | 13:10 |
| *** davidhadas has quit IRC | 13:10 | |
| *** sarob has quit IRC | 13:10 | |
| *** thomasem has joined #openstack-infra | 13:12 | |
| *** DenisEgorenko has joined #openstack-infra | 13:13 | |
| sdague | dstufft: that's where you made a mistake, you start emacs-server, and never need to stop it | 13:13 |
| dstufft | sdague: a text editor with a server inside it? o.O | 13:14 |
| * dstufft is lazy and uses ST3 | 13:14 | |
| sdague | dstufft: yeh, because you can send files to it | 13:14 |
| *** toddmorey has joined #openstack-infra | 13:15 | |
| ttx | anteaya: I'm subscribed to the page and I translate that into a gcal meeting | 13:16 |
| ttx | anteaya: manual, time-consuming and error-prone process | 13:16 |
| ttx | I just create a google calendar meeting, set to UTC time, on that common calendar | 13:17 |
| ttx | annegentle and myself have rights to it. I'm fine with adding anyone who wants to play, but would rather just automate the process | 13:18 |
| ttx | since if we have multiple people it becomes difficult to track who does the update | 13:19 |
| sdague | this is one of those times, I again, wish we used a CMS instead of a wiki. Then we could do something like this automatically - http://mhvlug.org/calendar/month/2013-11 | 13:19 |
| ogelbukh | just a reminder: diagnostics api project stackforge request is still in review here https://review.openstack.org/#/c/53858/4 | 13:20 |
| anteaya | ttx eww | 13:21 |
| anteaya | right | 13:22 |
| anteaya | can you add me for this one | 13:22 |
| anteaya | I will add it to the calendar and then see the process myself | 13:22 |
| anteaya | then remove my rights and I will have better feel what I am after when I am fixing that bug | 13:22 |
| *** boris-42 has joined #openstack-infra | 13:22 | |
| anteaya | can we do that? | 13:22 |
| *** dmsimard has joined #openstack-infra | 13:23 | |
| ttx | anteaya: I already added it | 13:23 |
| ttx | anteaya: next time I see one we can do that :) | 13:23 |
| anteaya | ttx okay thanks | 13:24 |
| *** senk has quit IRC | 13:24 | |
| *** sgran has joined #openstack-infra | 13:26 | |
| *** dkliban has joined #openstack-infra | 13:27 | |
| *** sgran has left #openstack-infra | 13:27 | |
| *** CaptTofu has quit IRC | 13:28 | |
| DenisEgorenko | hi guys, i have some questions about Cacti. | 13:28 |
| DenisEgorenko | 1. do you use thresholds in your monitoring? | 13:28 |
| DenisEgorenko | 2. I have some problem: i create some tresholds and specify two e-mail to receive warnings and alerts. And when alert or warning was triggered it delivered only to one of email. But when some host is down i have delivery notifications to both of email. What's maybe wrong? | 13:28 |
| *** CaptTofu has joined #openstack-infra | 13:29 | |
| mordred | DenisEgorenko: we do not have thresholds set | 13:29 |
| *** mrmartin has joined #openstack-infra | 13:29 | |
| mordred | DenisEgorenko: we use our lovely developers as an alert system :) | 13:30 |
| mrmartin | re | 13:30 |
| DenisEgorenko | mordred: thank you) maybe, do you know what i should to do with my problem?) | 13:31 |
| *** dims has quit IRC | 13:32 | |
| mrmartin | guys, am I the only one who is experiencing that openstack.org is down currently? | 13:32 |
| ogelbukh | no, it's actually down atm ( | 13:32 |
| *** dims has joined #openstack-infra | 13:33 | |
| mrmartin | it returns an 500 Internal server error http status code | 13:33 |
| *** masayukig has quit IRC | 13:38 | |
| *** masayukig has joined #openstack-infra | 13:38 | |
| *** dkranz has joined #openstack-infra | 13:40 | |
| mrmartin | I opened a ticket for the website issue https://bugs.launchpad.net/openstack-ci/+bug/1250854 | 13:42 |
| *** jorisroovers has joined #openstack-infra | 13:47 | |
| sdague | mrmartin: well, it's not actually under infra control | 13:47 |
| sdague | so that's not all that useful | 13:48 |
| sdague | hmmm... I wonder if beating people up on the foundation mailing list is appropriate for this :) | 13:49 |
| mrmartin | sdague: ok, we need to find somebody who can restart the site, a bit unpleasant that we can't promote Openstack after the summit | 13:49 |
| sdague | mrmartin: everyone agrees | 13:49 |
| *** nati_ueno has quit IRC | 13:49 | |
| sdague | this is actually a sore spot that openstack.org isn't run by the infra team, because it's not fixable by the technical team | 13:49 |
| *** nati_ueno has joined #openstack-infra | 13:49 | |
| mrmartin | I wrote a mail to reed. | 13:51 |
| *** ericw has joined #openstack-infra | 13:52 | |
| sdague | yeh, I think we need to be more public than tath | 13:52 |
| *** mfer has joined #openstack-infra | 13:52 | |
| *** nati_ueno has quit IRC | 13:54 | |
| dstufft | fwiw openstack.org loads for me | 13:54 |
| mordred | yup. it's back up | 13:55 |
| *** talluri has quit IRC | 13:57 | |
| *** jorisroovers has quit IRC | 13:58 | |
| mrmartin | works for me too | 13:58 |
| sdague | well, I trolled the foundation list anyway | 13:59 |
| sdague | it feels like we need a public conversation about this, not just emails / tweets to reed and sparky | 13:59 |
| mrmartin | maybe some failover deployment model could be nice | 14:00 |
| *** dmsimard1 has joined #openstack-infra | 14:01 | |
| *** dmsimard has quit IRC | 14:04 | |
| *** ben_duyujie has joined #openstack-infra | 14:05 | |
| *** dmsimard has joined #openstack-infra | 14:05 | |
| *** osanchez has quit IRC | 14:05 | |
| *** osanchez has joined #openstack-infra | 14:05 | |
| dstufft | trolling lists is always a good time | 14:06 |
| *** dmsimard1 has quit IRC | 14:08 | |
| *** DenisEgorenko has left #openstack-infra | 14:08 | |
| *** jamesmcarthur has quit IRC | 14:11 | |
| *** julim has joined #openstack-infra | 14:11 | |
| *** jamesmcarthur has joined #openstack-infra | 14:12 | |
| fungi | ESCROLBK | 14:13 |
| *** rfolco has quit IRC | 14:13 | |
| *** xchu has joined #openstack-infra | 14:14 | |
| *** mriedem has joined #openstack-infra | 14:15 | |
| *** alcabrera has joined #openstack-infra | 14:19 | |
| *** jcoufal-mob has joined #openstack-infra | 14:22 | |
| *** jamesmcarthur has quit IRC | 14:24 | |
| *** senk has joined #openstack-infra | 14:24 | |
| *** jcoufal-mob has quit IRC | 14:25 | |
| *** jcoufal-mob has joined #openstack-infra | 14:25 | |
| *** rfolco has joined #openstack-infra | 14:26 | |
| *** mrmartin has quit IRC | 14:28 | |
| fungi | ttx: mordred: there was nobody to complain to on the call. tom and margie were the only other two people to call in for it | 14:29 |
| mordred | ttx: awesome | 14:29 |
| *** jcoufal-mob has quit IRC | 14:30 | |
| *** mrmartin has joined #openstack-infra | 14:30 | |
| *** weshay has joined #openstack-infra | 14:32 | |
| fungi | i think having more than just the usual complainy handful of us expressing concern about this would be more likely to help build inertia anyway, so maybe a few more outages are a good thing | 14:33 |
| fungi | anteaya: ttx: on the calendar front, i'm keen on clarkb's earlier suggestion as well because it could 1. get us away from (proprietary) google calendars and 2. provide the user groups with a potential solution which isn't (proprietary) meetup | 14:35 |
| *** sarob has joined #openstack-infra | 14:36 | |
| *** oubiwann has joined #openstack-infra | 14:36 | |
| *** jamesmcarthur has joined #openstack-infra | 14:37 | |
| *** jamesmcarthur has quit IRC | 14:37 | |
| ogelbukh | fungi: hello | 14:38 |
| *** toddmorey has quit IRC | 14:38 | |
| ogelbukh | fungi: any chance you could look at https://review.openstack.org/#/c/53858/4 ? | 14:38 |
| mordred | fungi: I missed the calendar discussion | 14:39 |
| *** sarob has quit IRC | 14:40 | |
| mrmartin | fungi: we are working on a community portal that can replace meetup | 14:41 |
| mordred | ah - I see the scrollback now | 14:44 |
| fungi | mrmartin: oh, right, forgot that was what that did. maybe it could also eventually handle general calendaring for the project? is there enough overlap there with its current mission and design? | 14:46 |
| mrmartin | it will contain a calendaring for events | 14:48 |
| sdague | fungi: what is the calendaring proposal? | 14:48 |
| *** alexpilotti has joined #openstack-infra | 14:48 | |
| mrmartin | and we try to integrate meetup.com in the first round | 14:48 |
| sdague | I've spent a lot of time on ical hacking over the years, so am very keen on the discussion | 14:48 |
| sdague | mrmartin: what are you building it with? | 14:48 |
| fungi | sdague: just wondering about whether the new groups system would be a good place for ical feeds of other things to eventually get managed, like project calendars | 14:49 |
| fungi | in a perfect world, something should also supplant our use of eventbrite. i'm not keen on the fact that to attend a free software conference i need to sign up with a proprietary service which then continues to spam me about other people's events they're also marketing | 14:49 |
| mrmartin | fungi, but first we want to integrate with membership database through OpenID | 14:49 |
| mrmartin | and we are planning a speakers list also | 14:49 |
| fungi | mrmartin: sounds great | 14:50 |
| *** che-arne has quit IRC | 14:52 | |
| *** dcramer_ has joined #openstack-infra | 14:52 | |
| *** davidhadas has joined #openstack-infra | 14:55 | |
| *** senk has quit IRC | 14:55 | |
| *** CaptTofu has quit IRC | 14:57 | |
| *** CaptTofu has joined #openstack-infra | 14:57 | |
| *** zehicle_at_dell has quit IRC | 15:01 | |
| anteaya | sdague mrmartin https://bugs.launchpad.net/openstack-ci/+bug/1250758 | 15:03 |
| anteaya | fungi has already commented on it | 15:03 |
| mrmartin | anteaya: oh I saw that | 15:04 |
| chmouel | perhaps something nice to have in the future on the infra https://github.com/mozilla/togetherjs http://lwn.net/SubscriberLink/572905/86f242efffe0930c/ | 15:04 |
| openstackgerrit | Fei Long Wang proposed a change to openstack-infra/reviewstats: Add new core member for Glance https://review.openstack.org/56230 | 15:05 |
| *** alexpilotti has quit IRC | 15:05 | |
| *** UtahDave has joined #openstack-infra | 15:09 | |
| *** mrodden has joined #openstack-infra | 15:09 | |
| *** zehicle_at_dell has joined #openstack-infra | 15:12 | |
| fungi | chmouel: i wonder whether that's a viable alternative to disqus? i know docs was wanting to replace it | 15:12 |
| chmouel | fungi: is that trying to replace disqus ? I have understood that the alternative is to etherpads | 15:13 |
| annegentle | fungi: chmouel: we want to put ask.o.o threads in instead of disqus threads for docs | 15:13 |
| annegentle | chmouel: the alternative is threaded discussions that aren't through a proprietary tool | 15:14 |
| fungi | chmouel: ahh, okay. it was unclear to me what togetherjs did from its readme | 15:14 |
| chmouel | annegentle, fungi: https://togetherjs.com/ | 15:14 |
| fungi | got it | 15:14 |
| chmouel | see the video, this is more a collobaration kind of wsgi editor | 15:14 |
| openstackgerrit | Monty Taylor proposed a change to openstack-dev/cookiecutter: Add post hook to do git initialization https://review.openstack.org/56233 | 15:14 |
| annegentle | chmouel: yeah I see togetherjs as an etherpad sub | 15:14 |
| fungi | annegentle: right, i couldn't remember whether you had arrived at a solution. forgot it was going to be integration with ask.o.o | 15:14 |
| annegentle | fungi: yah no worries, I don't know how far along we are really :) | 15:15 |
| chmouel | perhaps this is something that can be add to blueprints/storyboard for collobarotive work | 15:15 |
| *** guohliu has joined #openstack-infra | 15:16 | |
| fungi | perhaps. looks like it might be good for collaborating on less text-y web content (forms, diagrams, et cetera) | 15:16 |
| sdague | mrmartin: so what base infrastructure are you using for the community site you are spinning up | 15:17 |
| *** alexpilotti has joined #openstack-infra | 15:18 | |
| mrmartin | sdague: it is Drupal / PHP | 15:18 |
| mordred | that's pretty neat | 15:19 |
| sdague | mrmartin: so you know I wrote this - https://drupal.org/project/meetup_events ? | 15:19 |
| mrmartin | sdague: not yet, but sounds great | 15:20 |
| sdague | and would be super happy to fix anything needed for your use case | 15:20 |
| mrmartin | sdague: we are extending a Drupal Commons distribution, and I always welcome new contributors | 15:21 |
| sdague | sure, that plugin should mostly be drop in, it's kind of configurable about how you handle events | 15:21 |
| fungi | sdague: http://git.openstack.org/cgit/openstack-infra/groups | 15:22 |
| sdague | though it currently only hits a single meetup, but I should be able to extend not too bad to multiple | 15:22 |
| mrmartin | we definietly need that | 15:22 |
| sdague | yep | 15:22 |
| sdague | I previously had a request for it from another group, just didn't get around to it. But supporting the openstack use case means I could justify work time on it :) | 15:23 |
| fungi | sdague: and http://groups-dev.openstack.org/ | 15:23 |
| sdague | ok, cool. I've got another flight on sunday, I'll pull this before the flight to look | 15:24 |
| fungi | currently the groups-dev server is running on continuous deployment from the openstack-infra/groups project | 15:24 |
| sdague | nice | 15:25 |
| *** xchu has quit IRC | 15:25 | |
| mrmartin | sdague: I try to recruit some new Drupal developers on the Drupal Weekend event at Budapest, this weekend | 15:25 |
| sdague | mrmartin: is there already an event content type baked in there | 15:25 |
| mrmartin | sdague: yep, but this patch sits on my dev machine, but we can enlist envents currently | 15:25 |
| *** jergerber has joined #openstack-infra | 15:26 | |
| mrmartin | sdague: this is the staging site, we need to do a lot, gating, prod deployment, etc. | 15:26 |
| sdague | mrmartin: sure, but as soon as the event content type is up there I can make sure the meetup_events groks the event definition | 15:27 |
| sdague | and the many to many meetups support | 15:28 |
| *** markmc has joined #openstack-infra | 15:28 | |
| mrmartin | sdague: great, I'll ping you when we are there, first of all, we need to do OpenID integration with membership database | 15:29 |
| anteaya | mrmartin sdague I'll keep ttx's bug assigned to me, let me know if you feel your work will cover that use case as well as a time line for doing so | 15:29 |
| sdague | mrmartin: sounds great. Also, for when you get to it, I've done *way* too much hacking on drupal's ical support. So might be helpful there | 15:30 |
| *** mdenny has joined #openstack-infra | 15:30 | |
| *** mgagne has joined #openstack-infra | 15:31 | |
| *** rnirmal has joined #openstack-infra | 15:33 | |
| *** masayukig has quit IRC | 15:35 | |
| *** sarob has joined #openstack-infra | 15:36 | |
| *** alexpilotti has quit IRC | 15:37 | |
| *** jamesmcarthur has joined #openstack-infra | 15:39 | |
| *** sarob has quit IRC | 15:41 | |
| *** che-arne has joined #openstack-infra | 15:41 | |
| openstackgerrit | Monty Taylor proposed a change to openstack-infra/jeepyb: Create new repos using cookiecutter https://review.openstack.org/56236 | 15:42 |
| mordred | fungi, clarkb ^^ there's a fun one for ya. | 15:43 |
| *** sdake has quit IRC | 15:43 | |
| mordred | sdague: you might find that interesting too | 15:43 |
| *** sdake has joined #openstack-infra | 15:43 | |
| *** datsun180b has joined #openstack-infra | 15:46 | |
| sdague | mordred: cool | 15:49 |
| *** yaguang has quit IRC | 15:53 | |
| openstackgerrit | A change was merged to openstack-infra/config: Add swiftsync. https://review.openstack.org/56061 | 15:55 |
| *** che-arne has quit IRC | 15:55 | |
| *** rcleere has joined #openstack-infra | 15:56 | |
| *** senk has joined #openstack-infra | 15:56 | |
| mordred | fungi: https://review.openstack.org/#/c/56179/1 | 15:57 |
| mordred | fungi: I'd like to land that, but I'm not 100% sure how to review it | 15:57 |
| openstackgerrit | A change was merged to openstack-infra/config: Fix typo in sqlalchemy-migrate upstream path https://review.openstack.org/56178 | 15:57 |
| mordred | it SEEMS like a solid patch | 15:57 |
| *** jergerber has quit IRC | 15:58 | |
| fungi | probably we should try to load the original and the new yaml files in a parser and compare them programmatically? | 15:58 |
| mgagne | mordred: load both files and diff the 2 contents? | 15:58 |
| fungi | heh | 15:59 |
| openstackgerrit | Julien Danjou proposed a change to openstack/requirements: Update stevedore https://review.openstack.org/56240 | 15:59 |
| mordred | fungi, mgagne: hrm. not a bad idea | 15:59 |
| fungi | but yes, i would very, very much like to see that land | 15:59 |
| mordred | I'm rebasing it on the swiftsync change right now | 15:59 |
| openstackgerrit | Monty Taylor proposed a change to openstack-infra/config: Add turbo-hipster into stackforge https://review.openstack.org/56180 | 15:59 |
| openstackgerrit | Monty Taylor proposed a change to openstack-infra/config: Add turbo-hipster to zuul+jenkins https://review.openstack.org/56181 | 15:59 |
| openstackgerrit | Monty Taylor proposed a change to openstack-infra/config: Reorder gerrit projects alphabetically https://review.openstack.org/56179 | 15:59 |
| *** senk1 has joined #openstack-infra | 16:00 | |
| fungi | also nice would be a gate-config-projects-lint job which as a start can confirm it's alpha order | 16:00 |
| fungi | then later we can expand it to do some more thorough validation as well | 16:00 |
| *** wenlock has joined #openstack-infra | 16:01 | |
| *** senk has quit IRC | 16:01 | |
| *** alcabrera is now known as alcabrera|afk | 16:03 | |
| *** jergerber has joined #openstack-infra | 16:03 | |
| openstackgerrit | A change was merged to openstack-dev/pbr: Add the semver documentation https://review.openstack.org/48855 | 16:05 |
| mordred | fungi: ok. I believe that the change is good | 16:06 |
| mordred | fungi: I have verified it by loading each yaml file into pytohn | 16:06 |
| mordred | and then sorting the list of dicts in python | 16:07 |
| fungi | awesome. let's try to cram these through then, or else they'll be hell to keep rebasing | 16:07 |
| mordred | nsort = sorted(ns[1], key=lambda k: k['project']) | 16:07 |
| mordred | and then verifying that the resulting lists of dicts are equal | 16:07 |
| mordred | which they are | 16:07 |
| *** sarob has joined #openstack-infra | 16:07 | |
| *** hogepodge has joined #openstack-infra | 16:08 | |
| mordred | done | 16:09 |
| openstackgerrit | A change was merged to openstack-infra/config: Reorder gerrit projects alphabetically https://review.openstack.org/56179 | 16:10 |
| openstackgerrit | A change was merged to openstack-infra/config: Add turbo-hipster into stackforge https://review.openstack.org/56180 | 16:10 |
| openstackgerrit | A change was merged to openstack-infra/config: Add turbo-hipster to zuul+jenkins https://review.openstack.org/56181 | 16:11 |
| openstackgerrit | Monty Taylor proposed a change to openstack-infra/config: Add rubick project to stackforge https://review.openstack.org/53858 | 16:11 |
| *** pblaho has quit IRC | 16:11 | |
| *** rcleere has quit IRC | 16:12 | |
| *** sarob has quit IRC | 16:12 | |
| openstackgerrit | Monty Taylor proposed a change to openstack-infra/config: Add the tomograph project to stackforge. https://review.openstack.org/56115 | 16:13 |
| mordred | fungi: ok. I rebased the two outstanding changes that were touching projects.yaml | 16:13 |
| fungi | great--thanks! | 16:14 |
| *** zehicle_at_dell has quit IRC | 16:16 | |
| *** danger_fo_away is now known as danger_fo | 16:17 | |
| openstackgerrit | Monty Taylor proposed a change to openstack-infra/config: Add the tomograph project to stackforge. https://review.openstack.org/56115 | 16:17 |
| *** zehicle_at_dell has joined #openstack-infra | 16:17 | |
| *** hashar has quit IRC | 16:17 | |
| *** hashar has joined #openstack-infra | 16:18 | |
| *** rcleere has joined #openstack-infra | 16:19 | |
| dims | some good news, looks like when https://review.openstack.org/#/c/56075/ goes through, we can switch gate-tempest-devstack-vm-neutron-large-ops to a voting job on all the projects where it is non-voting now. got 7 good back to back runs (logstash query - http://bit.ly/1cUUh7L) | 16:19 |
| mgagne | mordred: according to my tests, both lists are identical | 16:19 |
| *** ryanpetrello has joined #openstack-infra | 16:19 | |
| mgagne | mordred: http://paste.openstack.org/show/52444/ | 16:20 |
| *** che-arne has joined #openstack-infra | 16:20 | |
| *** mrmartin has quit IRC | 16:20 | |
| openstackgerrit | Monty Taylor proposed a change to openstack-infra/config: Add rubick project to stackforge https://review.openstack.org/53858 | 16:21 |
| mordred | fungi: ok. NOW I have rebased | 16:21 |
| mordred | dims: AWESOME | 16:21 |
| fungi | heh | 16:21 |
| fungi | mgagne: thanks for double-checking! | 16:21 |
| mgagne | fungi: I did some basic checks and introduced errors to test the script. | 16:22 |
| *** afazekas has quit IRC | 16:23 | |
| *** luhrs1 has joined #openstack-infra | 16:24 | |
| *** luhrs1 has quit IRC | 16:24 | |
| jlk | mordred: were you able to assign rackspace-auth-openstack to my account (j2solutions) ? | 16:24 |
| *** jorisroovers has joined #openstack-infra | 16:25 | |
| fungi | jlk: right now https://pypi.python.org/pypi/rackspace-auth-openstack says Package Index Owner: sdirector (and doesn't mention any other accounts) so i'm guessing not yet | 16:26 |
| jlk | k | 16:27 |
| fungi | normally the package index owners and package index maintainers will be displayed on the project's pypi page | 16:28 |
| *** yfried has quit IRC | 16:31 | |
| *** xeyed4good has joined #openstack-infra | 16:31 | |
| *** xeyed4good has left #openstack-infra | 16:31 | |
| annegentle | ttx: do you happen to remember the etherpad link for the Summit 101 talk? | 16:32 |
| *** senk1 has quit IRC | 16:32 | |
| ttx | annegentle: no, but my firefox awesomebar knows it | 16:37 |
| ttx | https://etherpad.openstack.org/p/icehouse-101 | 16:37 |
| *** romcheg has joined #openstack-infra | 16:41 | |
| *** hogepodge has quit IRC | 16:42 | |
| annegentle | ttx: yeehaw! thanks | 16:43 |
| romcheg | Morning folks | 16:43 |
| romcheg | I need some help with devstack-gate. | 16:43 |
| *** nikhil__|afk is now known as nik | 16:45 | |
| romcheg | I found an invocation of the configure_tempest.sh script: http://paste.openstack.org/show/52459/ | 16:45 |
| anteaya | hey romcheg welcome | 16:46 |
| romcheg | However, I cannot find it in devstack. Where else should I look for it? | 16:46 |
| *** nik is now known as Guest68799 | 16:46 | |
| romcheg | Morning anteaya | 16:46 |
| anteaya | romcheg: I think the -qa channel would have more access to folks that would know | 16:46 |
| anteaya | are you in -qa yet? | 16:46 |
| romcheg | anteaya: no. I'm going to join it now | 16:47 |
| anteaya | k | 16:47 |
| anteaya | see you there | 16:47 |
| *** alcabrera|afk is now known as alcabrera | 16:48 | |
| *** davidhadas has quit IRC | 16:48 | |
| *** UtahDave has quit IRC | 16:50 | |
| anteaya | how do I run the test coverage script/tool against neutron to evaluate the level of coverage for unit tests? | 16:50 |
| anteaya | I have folks wanting to help close the gap and I need to know where to point them | 16:50 |
| *** yassine has quit IRC | 16:52 | |
| *** gyee has joined #openstack-infra | 16:53 | |
| *** hogepodge has joined #openstack-infra | 16:53 | |
| openstackgerrit | Johannes Erdfelt proposed a change to openstack-infra/jenkins-job-builder: Make -o option required if using 'test' command https://review.openstack.org/56250 | 16:54 |
| *** nicedice_ has joined #openstack-infra | 16:55 | |
| *** hogepodge has quit IRC | 16:56 | |
| *** ^d has joined #openstack-infra | 16:56 | |
| *** hogepodge has joined #openstack-infra | 16:58 | |
| *** jorisroovers has quit IRC | 17:00 | |
| *** UtahDave has joined #openstack-infra | 17:02 | |
| *** mrmartin has joined #openstack-infra | 17:04 | |
| *** blamar has quit IRC | 17:06 | |
| *** yolanda has quit IRC | 17:06 | |
| fungi | anteaya: i think 'tox -ecover' | 17:08 |
| *** yolanda has joined #openstack-infra | 17:09 | |
| *** UtahDave has quit IRC | 17:13 | |
| *** pblaho has joined #openstack-infra | 17:13 | |
| *** markmc has quit IRC | 17:13 | |
| *** jergerber has quit IRC | 17:14 | |
| *** zehicle_at_dell has quit IRC | 17:15 | |
| anteaya | fungi how easy is that | 17:16 |
| clarkb | morning | 17:16 |
| * anteaya heads for -ecover | 17:16 | |
| clarkb | fungi: anteaya calagator is the calendar thing | 17:16 |
| anteaya | morning clarkb | 17:16 |
| *** zehicle_at_dell has joined #openstack-infra | 17:16 | |
| anteaya | calagator | 17:16 |
| clarkb | sdague: fungi: I am going to approve d-g changes now if they haven't been approved yet | 17:16 |
| anteaya | I kept thinking alligator | 17:16 |
| anteaya | so I got the end correct | 17:16 |
| fungi | clarkb: i think they have not | 17:16 |
| fungi | still trying to get un-buried this morning | 17:16 |
| *** hashar has quit IRC | 17:17 | |
| *** Guest68799 is now known as kin | 17:17 | |
| mordred | morning clarkb | 17:17 |
| mordred | jlk: I was not - but I will do it right now | 17:17 |
| mordred | jlk: you are now the owner | 17:18 |
| *** dmsimard has left #openstack-infra | 17:18 | |
| mordred | jlk: feel free to remove me any time (but I'll let you do that so that we verify you do have ownership) | 17:18 |
| *** hogepodge has quit IRC | 17:23 | |
| notmyname | lifeless: your country is cool. they aren't making me get a visa as I fly through on my way to LCA :-) | 17:24 |
| *** hogepodge has joined #openstack-infra | 17:24 | |
| *** pblaho has quit IRC | 17:24 | |
| *** jergerber has joined #openstack-infra | 17:25 | |
| fungi | and they also film hobbits there, in their natural habitat | 17:25 |
| clarkb | fungi: sdague: d-g changes approved I will keep an eye on them | 17:25 |
| clarkb | notmyname: you do need a visa for LCA though | 17:26 |
| clarkb | (assuming you are a US citizen) | 17:26 |
| anteaya | notmyname: where are you coming from that NZ is on the way to Perth? | 17:29 |
| *** ben_duyujie has quit IRC | 17:30 | |
| *** mkerrin has joined #openstack-infra | 17:33 | |
| *** reed has joined #openstack-infra | 17:34 | |
| reed | and I'm back | 17:35 |
| reed | good day to all | 17:36 |
| *** markwash has quit IRC | 17:39 | |
| openstackgerrit | Monty Taylor proposed a change to openstack-infra/nodepool: Add availability zone support https://review.openstack.org/56259 | 17:40 |
| openstackgerrit | Monty Taylor proposed a change to openstack-infra/nodepool: Add ability to filter on flavor name https://review.openstack.org/55844 | 17:40 |
| mordred | clarkb: ^^ new beta of hpcloud uses actual az support in nova | 17:40 |
| *** hogepodge has quit IRC | 17:40 | |
| mordred | as a follow on to the patch above, we may need to actually add new logic into nodepool | 17:40 |
| mordred | because AIUI, quota is region-wide, so our quota calculations will be off if we treat azs as a whole other provider | 17:41 |
| *** toddmorey has joined #openstack-infra | 17:41 | |
| clarkb | mordred: can we treat an entire region as being one az? | 17:42 |
| mordred | we can treat an entire region as being a provider in nodepol.yaml like we do now | 17:43 |
| mordred | but - to get placement of a node in different azs, you have to provide an argument to the nova api | 17:43 |
| *** hogepodge has joined #openstack-infra | 17:43 | |
| mordred | which means we need to be able to give a provider a list of azs, so that nodepool can spread the love | 17:43 |
| fungi | we may care less about which azs we spin them up in, if the quotas are no longer per-az | 17:43 |
| mordred | I think we want to spread across azs | 17:44 |
| mordred | so that if an az goes down | 17:44 |
| mordred | we don't lose that as a provider | 17:44 |
| fungi | yeah, good point. less manual fiddling during outages | 17:44 |
| mordred | yup. it's purely about managing how we calculate quota | 17:44 |
| fungi | speaking of outages, the resolution on bug 1250854 is mildly frightening | 17:44 |
| clarkb | mordred: so we won't get automagically load balanced (I suppose being explicit is probably better in any case) | 17:45 |
| * fungi is trying to imagine what manner of catastrophe befell the site, that a restore from backup was required | 17:45 | |
| mordred | clarkb: that's correct | 17:45 |
| *** jpich has quit IRC | 17:45 | |
| mordred | fungi: well, it's tough to know, because it's running in magical cloud sites | 17:45 |
| fungi | magicland | 17:45 |
| clarkb | fungi: I try not to think about it | 17:46 |
| fungi | where flying unicorns crap proprietary rainbows | 17:46 |
| *** osanchez has quit IRC | 17:46 | |
| openstackgerrit | Monty Taylor proposed a change to openstack-infra/config: Add new HP Cloud regions https://review.openstack.org/56260 | 17:47 |
| mordred | clarkb, fungi ^^ but that will require the change above to nodepool to be operational | 17:47 |
| mordred | I do not think we should put significant load on the new hp cloud stuff until we rework quota calc to understand azs better | 17:48 |
| mordred | but I think I'd like jeblair to be around before we try to do that | 17:48 |
| clarkb | mordred: ++ so review now but don't approve | 17:48 |
| mordred | no - I think we can appove | 17:48 |
| mordred | we just need to land nodepool before config | 17:48 |
| clarkb | but not until jeblair is back? | 17:49 |
| mordred | it will be non-optimal in exteme quota-pushing cases or outages | 17:49 |
| mordred | we can totally land now | 17:49 |
| clarkb | oh I misread your jeblair comment | 17:49 |
| mordred | I think there is an additional optimization we can do to grok azs later | 17:49 |
| clarkb | that was in referral to he quota stuff | 17:49 |
| * clarkb properly reviews then | 17:49 | |
| *** mfer has quit IRC | 17:49 | |
| *** blamar has joined #openstack-infra | 17:51 | |
| *** denis_makogon has joined #openstack-infra | 17:51 | |
| clarkb | mordred: is the image id the same in new hpcloud? | 17:53 |
| mordred | yes | 17:53 |
| *** romcheg has quit IRC | 17:54 | |
| reed | fungi, not sure how Marton knows that the resolution was 'restored from backup' | 17:54 |
| mrmartin | hi reed | 17:55 |
| notmyname | clarkb: ya I got that | 17:55 |
| mrmartin | somebody wrote that here | 17:55 |
| reed | today's kernel upgrade must have screwed up display refresh :( | 17:55 |
| reed | hi mrmartin, did you talk to toddmorey? | 17:55 |
| reed | oh, ok | 17:55 |
| anteaya | mordred can you take a peek at this? http://paste.openstack.org/show/52472/ | 17:56 |
| notmyname | anteaya: SFO. miles count for united and it's relatively cheap (ie half the price of SFO->HKG->PER) | 17:56 |
| reed | fungi, since I adopted your gpg.conf hell opened its gates and demons started walking around my systems | 17:56 |
| anteaya | I tried running tox -ecover inside neutron, in the hopes of getting some feedback on unit test coverage | 17:56 |
| anteaya | notmyname: really? | 17:57 |
| anteaya | so SFO > NZ > PER? | 17:57 |
| notmyname | anteaya: ya. SFO->AKL->MEL->PER | 17:57 |
| anteaya | wow | 17:57 |
| anteaya | going to be a fun first leg | 17:57 |
| notmyname | yup | 17:57 |
| notmyname | actually, it's shorter (or similar) to hong kong | 17:58 |
| anteaya | cool | 17:58 |
| *** senk has joined #openstack-infra | 17:58 | |
| jlk | mordred: thanks, I've published. | 17:59 |
| *** senk has quit IRC | 17:59 | |
| *** senk has joined #openstack-infra | 18:00 | |
| *** derekh has quit IRC | 18:00 | |
| openstackgerrit | A change was merged to openstack-infra/devstack-gate: Extract setup_project as a single function https://review.openstack.org/55028 | 18:01 |
| fungi | reed: how so? | 18:01 |
| anteaya | mordred it looks like pbr and requirements to me, yes? | 18:01 |
| reed | fungi, how can I say it..... nothing works :) | 18:01 |
| reed | fungi, search-keys fails, recv-keys fails | 18:02 |
| reed | makes me sad | 18:02 |
| fungi | reed: did you download and install the ca key mentioned in that section right after the sample config? | 18:02 |
| clarkb | 1/3 d-g changes merged. e-r nicely gave me a bug to reverify the other two against so trying again | 18:02 |
| reed | fungi, yes | 18:02 |
| fungi | reed: what error message are you getting? | 18:03 |
| reed | $ gpg2 --search-keys me@not.mn | 18:03 |
| reed | gpg: searching for "me@not.mn" from hkps server hkps.pool.sks-keyservers.net | 18:03 |
| reed | gpgkeys: HTTP search error 77: | 18:03 |
| reed | gpg: key "me@not.mn" not found on keyserver | 18:03 |
| reed | gpg: keyserver internal error | 18:03 |
| reed | gpg: keyserver search failed: Keyserver error | 18:04 |
| reed | while http://hkps.pool.sks-keyservers.net/pks/lookup?op=vindex&search=me%40not.mn&fingerprint=on | 18:04 |
| clarkb | that reminds me I should do key signing stuff tonight | 18:04 |
| *** yfried has joined #openstack-infra | 18:04 | |
| anteaya | looks like it failed on greenlet: UnicodeDecodeError: 'ascii' codec can't decode byte 0xe2 in position 66: ordinal not in range(128) | 18:04 |
| *** krotscheck has joined #openstack-infra | 18:05 | |
| fungi | reed: weird. for me it seems to work fine. does this... http://paste.openstack.org/show/52478/ | 18:06 |
| mordred | anteaya: looking | 18:06 |
| *** MarkAtwood has quit IRC | 18:07 | |
| reed | fungi, i'll tweak gpg.conf more and see if hell's gates close | 18:08 |
| fungi | reed: what gpg version does gpg2 --version report? | 18:08 |
| fungi | i'm testing with 2.0.20 at the moment | 18:08 |
| reed | gpg2 --version gpg (GnuPG) 2.0.19 libgcrypt 1.5.0 | 18:08 |
| anteaya | thanks | 18:09 |
| mordred | anteaya: | 18:09 |
| mordred | greenlet.h:8:20: fatal error: Python.h: No such file or directory | 18:09 |
| mordred | anteaya: you need to apt-get install python-dev | 18:09 |
| anteaya | can do | 18:09 |
| reed | fungi, http://paste.openstack.org/show/52480/ | 18:11 |
| reed | pgp is a huge source of frustration | 18:12 |
| anteaya | ah, who prepares invitation letters for the foundation for people who need a visa to summits? | 18:14 |
| anteaya | I just got asked to prepare an invitation letter for someone who needs a visa to Canada and I have never done that before. | 18:15 |
| clarkb | anteaya: there is an email address on the summit site iirc | 18:15 |
| *** mrmartin has quit IRC | 18:15 | |
| *** senk has quit IRC | 18:15 | |
| *** melwitt has joined #openstack-infra | 18:16 | |
| anteaya | clarkb: thanks | 18:16 |
| fungi | reed: the few hits i get for error 77 suggest it may be failing x.509 verification on the hkps server cert | 18:17 |
| *** markwash has joined #openstack-infra | 18:17 | |
| *** rhsu has joined #openstack-infra | 18:18 | |
| clarkb | MITM! | 18:18 |
| *** jorisroovers has joined #openstack-infra | 18:19 | |
| fungi | reed: can you confirm that 'sha256sum ~/.gnupg/sks-keyservers.netCA.pem' returns 0666ee848e03a48f3ea7bb008dbe9d63dfde280af82fb4412a04bf4e24cab36b | 18:19 |
| *** jroovers has joined #openstack-infra | 18:19 | |
| fungi | i'll also note that i have that file owned by my account with 0644 perms, so maybe if you have it group-writable gnupg gets unhappy? | 18:20 |
| clarkb | argh d-g failed again with new and exciting errors | 18:21 |
| *** wenlock has quit IRC | 18:21 | |
| *** UtahDave has joined #openstack-infra | 18:23 | |
| *** talluri has joined #openstack-infra | 18:23 | |
| *** jorisroovers has quit IRC | 18:23 | |
| *** mfer has joined #openstack-infra | 18:26 | |
| mordred | clarkb: excellent | 18:27 |
| clarkb | I think the swift thing is biting me, so I am rechecking jog0's devstack change to build up more confidence in that change | 18:27 |
| sdague | have we seen dtroyer back yet since HK? | 18:30 |
| dtroyer | sdague: no, he's still sleeping | 18:30 |
| sdague | heh | 18:31 |
| sdague | ok, I just +Aed jog0's patch | 18:31 |
| fungi | reed: it's also possible, depending on the system where you're running, that you may also need the gnupg-curl package installed | 18:31 |
| *** wenlock has joined #openstack-infra | 18:32 | |
| koolhead17 | hello world!! | 18:34 |
| anteaya | hello koolhead17 | 18:34 |
| reed | fungi, is this normal? http://paste.openstack.org/show/52491/ | 18:34 |
| anteaya | we said we would talk and we never did | 18:34 |
| koolhead17 | anteaya: how have you been? :) | 18:34 |
| anteaya | so much for good intentions | 18:34 |
| anteaya | koolhead17: good | 18:34 |
| anteaya | over committed like everyone else, but happy | 18:35 |
| koolhead17 | anteaya: we are on cloud ^^ you know | 18:35 |
| * koolhead17 pokes reed | 18:35 | |
| anteaya | koolhead17: sorry I missed the reference you made | 18:35 |
| anteaya | my brain is somewhere else | 18:35 |
| * reed smiles at koolhead17 and keeps on going as if pokes did not exist | 18:35 | |
| fungi | reed: yep, that's normal. it's being more verbose about trust levels because of the show-uid-validity option in your config | 18:35 |
| koolhead17 | anteaya: we can soon do google hangout | 18:36 |
| koolhead17 | reed: :) | 18:36 |
| anteaya | koolhead17: ah okay, except I hate them | 18:36 |
| anteaya | what is the status of the asterisk server? | 18:36 |
| anteaya | can koolhead and I talk on that? | 18:36 |
| jog0 | sdague: thanks | 18:36 |
| koolhead17 | anteaya: no idea. | 18:36 |
| jog0 | clarkb: I ran out of time last night to finish my babysitter.py script | 18:36 |
| anteaya | sorry was asking fungi clarkb | 18:36 |
| reed | anteaya, I have high hopes with https://togetherjs.com/ | 18:37 |
| reed | s/with/on | 18:37 |
| reed | jog0, what's that? | 18:38 |
| jog0 | reed: in order to test fixes for transient bugs, you have to hit recheck no bug a few dozen times to have any confidence | 18:38 |
| jog0 | aka babysit zuul | 18:38 |
| anteaya | reed: how are you envisioning togetherjs working? | 18:39 |
| reed | anteaya, dreaming to have it working together with groups.openstack.org one day | 18:39 |
| *** johnthetubaguy has quit IRC | 18:39 | |
| anteaya | okay | 18:39 |
| reed | anteaya, as soon as it adds support to video and screen sharing it will immediately become #1 option for online meetings, IMHO | 18:40 |
| *** krtaylor has quit IRC | 18:43 | |
| reed | fungi, resolved! I forgot to specify the default key in gpg.conf | 18:43 |
| reed | geez, that gpg thing is nasty | 18:43 |
| *** alcabrera is now known as alcabrera|afk | 18:45 | |
| *** jkyle has joined #openstack-infra | 18:49 | |
| *** ogelbukh1 has quit IRC | 18:50 | |
| *** mrmartin has joined #openstack-infra | 18:51 | |
| *** esker has joined #openstack-infra | 18:52 | |
| *** jerryz has joined #openstack-infra | 18:52 | |
| *** zehicle_at_dell has quit IRC | 18:53 | |
| lifeless | notmyname: indeed :) | 18:58 |
| anteaya | fungi: I'm guessing this is not the hoped for output of running tox -ecover: http://paste.openstack.org/show/52500/ | 18:58 |
| clarkb | anteaya: fungi: seems ok it says it passed | 19:00 |
| clarkb | you now need to open the .cover/index.html to look at the coverage | 19:00 |
| *** talluri has quit IRC | 19:01 | |
| clarkb | jog0: sdague: the devstack change failed the postgres gate test | 19:02 |
| clarkb | with a rest client 503 which I think is related to the swift thing :/ | 19:02 |
| anteaya | clarkb: ah | 19:02 |
| anteaya | thanks | 19:02 |
| *** alcabrera|afk is now known as alcabrera | 19:03 | |
| jog0 | clarkb: :( | 19:03 |
| clarkb | I think we may need to work on recreating the fails locally | 19:04 |
| fungi | reed: oh, i guess you had more than one | 19:04 |
| fungi | very odd behavior though | 19:04 |
| jog0 | clarkb: yeah, and monitor how much data we are pushing into swift | 19:04 |
| reed | fungi, one valid key and one retired | 19:05 |
| *** zehicle_at_dell has joined #openstack-infra | 19:06 | |
| jog0 | clarkb: did you just volunteer ? | 19:07 |
| *** zehicle_at_dell has quit IRC | 19:07 | |
| sdague | so... yeh, not good enough | 19:07 |
| clarkb | jog0: no | 19:07 |
| clarkb | :) | 19:07 |
| jog0 | we could always bump that number up to 15 GB or something silly | 19:07 |
| *** zehicle_at_dell has joined #openstack-infra | 19:08 | |
| jog0 | and babysit the patch to see if that works or not | 19:08 |
| *** ogelbukh1 has joined #openstack-infra | 19:08 | |
| clarkb | I have other things I really need to do today and falling into a debug cycle won't help that | 19:08 |
| jog0 | clarkb: heh same, bumping patch to 15GB and babysit it is | 19:08 |
| *** Ryan_Lane has joined #openstack-infra | 19:09 | |
| fungi | reed: yeah, i have a couple of valid keys (old one is deprecated but not revoked). it's probably been long enough i can set a reasonable expiration on it and let it age out | 19:11 |
| *** SergeyLukjanov has joined #openstack-infra | 19:12 | |
| *** jamesmcarthur has quit IRC | 19:17 | |
| jog0 | clarkb sdague: my patch failed for a different reason :) | 19:17 |
| jog0 | http://paste.openstack.org/show/52506/ | 19:17 |
| jog0 | notmyname: turns out swift wasn't designed to run in a single VM :) | 19:17 |
| clarkb | I never thought I could be so happy about a test failure :) | 19:17 |
| *** hashar has joined #openstack-infra | 19:17 | |
| jog0 | its database is locked error :/ | 19:18 |
| mordred | woot | 19:18 |
| mordred | clarkb: http://pypi.openstack.org/openstack/Ubuntu-12.04/ | 19:18 |
| *** davidhadas has joined #openstack-infra | 19:18 | |
| mordred | we have SOME wheels at least | 19:18 |
| mordred | I'm guessing the fact that we don't have more means there was a problem with one of the jobs | 19:18 |
| mordred | http://pypi.openstack.org/openstack/CentOS%206.4/ | 19:18 |
| mgagne | Chrome thinks this page is in Slovenian | 19:18 |
| mordred | Alex_Gaynor, dstufft ^^ | 19:18 |
| dstufft | what'd I do | 19:19 |
| clarkb | jog0: I wonder if the increased concurrency is affecting both disk use and db lock stuff | 19:20 |
| mordred | dstufft: just showing that we've got some wheels in our mirror now | 19:20 |
| dstufft | mordred: noice! | 19:20 |
| mordred | https://jenkins.openstack.org/job/post-mirror-python27/285/consoleFull | 19:20 |
| notmyname | jog0: I'm shocked! | 19:20 |
| dstufft | mordred: I see openstack stuff in red http://pythonwheels.com/ | 19:21 |
| clarkb | mordred: lol max recursion depth | 19:21 |
| *** hashar has quit IRC | 19:21 | |
| clarkb | mysql-python not a happy camper | 19:22 |
| mordred | dstufft: yah. I haven't landed the 'publish wheels to pypi' code yet | 19:22 |
| jog0 | clarkb: yeah, well in swift a db is a sqllite file | 19:22 |
| clarkb | jog0: oh! can we point it at mysql/postgres instead? | 19:22 |
| jog0 | clarkb: it doesn't make sense in swift land | 19:23 |
| clarkb | it does if swift can't handle concurrent HEAD requests | 19:23 |
| jog0 | they put sqllite files into swift and use replicas etc | 19:23 |
| mordred | clarkb: you are right. mysql-python is NOT a happy camper | 19:23 |
| jog0 | clarkb: it can't because we have one replica (I think) | 19:23 |
| mordred | clarkb: if I were to, say, fork mysql-python and remove its explicit distribute depend, could we use that instead? | 19:23 |
| clarkb | mordred: is upstream not receptive to patches like that? | 19:24 |
| jog0 | notmyname: ^ how can we make swift more performant in the devstack based gate | 19:24 |
| clarkb | (I am not familiar with the upstream for mysql-python) | 19:24 |
| mordred | clarkb: upstream is violently against removing the distribute depend | 19:24 |
| mordred | but, honestly, I do not understand why that is breaking in that manner | 19:24 |
| clarkb | wow | 19:24 |
| clarkb | mordred: because distribute is broken and is getting into an infinite loop | 19:25 |
| mordred | but I don't understand why it's getting broken distribute | 19:25 |
| mordred | since we should be updating that in sequence properly in the venv | 19:26 |
| clarkb | mysql-python is running _setup-distribute | 19:26 |
| clarkb | is their setup.py interacting with wheelstuff poorly | 19:27 |
| mordred | ah. well | 19:27 |
| *** ^d is now known as ^demon|away | 19:27 | |
| mordred | pypi-mirror does NOT update distribute in the venv | 19:27 |
| mordred | because who cares | 19:27 |
| mordred | except, it seems, perhaps we should | 19:27 |
| mordred | so that mysql-pyhton will not try to download one | 19:28 |
| *** rockstar has quit IRC | 19:28 | |
| Alex_Gaynor | mordred: are builds using wheels to install stuff? | 19:28 |
| mordred | Alex_Gaynor: we're getting closer to that | 19:29 |
| mordred | Alex_Gaynor: right now we're just working on getting wheels uploaded properly to our mirror | 19:29 |
| Alex_Gaynor | :) | 19:29 |
| mordred | next I'll do the things to make sure that our release jobs upload wheels to pypi as well | 19:29 |
| mordred | and then we'll land the change to start consuming them | 19:30 |
| dstufft | mordred: why in the world is upstream against removing the dependency for a dead project | 19:30 |
| mordred | dstufft: I have long since stopped trying to understand what's going on there | 19:30 |
| dstufft | he does releaze the latest distribute release is an empty tarball with nothing but a setup.py that points to the latest setuptools right? | 19:30 |
| dstufft | like distribute is not a thing | 19:30 |
| openstackgerrit | A change was merged to openstack-infra/reviewstats: Add new core member for Glance https://review.openstack.org/56230 | 19:31 |
| *** sarob has joined #openstack-infra | 19:31 | |
| *** sarob has quit IRC | 19:31 | |
| clarkb | d-g changes failed again :( | 19:32 |
| *** sarob has joined #openstack-infra | 19:32 | |
| clarkb | I will try pushing them through again after lunch | 19:32 |
| *** hashar has joined #openstack-infra | 19:33 | |
| openstackgerrit | Monty Taylor proposed a change to openstack-infra/pypi-mirror: Add distribute to the pre-install list https://review.openstack.org/48888 | 19:33 |
| mordred | clarkb: ^^ | 19:33 |
| clarkb | looking | 19:33 |
| clarkb | mordred: what is with running those commands twice? | 19:36 |
| clarkb | they seem to use the same venv_dir | 19:36 |
| mordred | clarkb: the first time installs them. the second time builds wheels of them | 19:37 |
| clarkb | that is why there is a pair of for loops, but you have two pairs of for loops | 19:37 |
| clarkb | oh are we creating a new venv with teh venv format command? | 19:37 |
| mordred | yes | 19:37 |
| mordred | have I mentioned we need to refactor this script? | 19:38 |
| mordred | the second for loop in the second set can probably die | 19:38 |
| mordred | but it also is not breaking much | 19:38 |
| clarkb | we should figure out how to use functions in those scripts :) | 19:39 |
| mordred | yah | 19:39 |
| mordred | I dunno though | 19:39 |
| mordred | functions are hard | 19:39 |
| Shrews | the problem is... | 19:40 |
| Shrews | you aren't doing it in java | 19:40 |
| * Shrews ducks | 19:40 | |
| * mordred throws a Shrews at Shrews | 19:41 | |
| mordred | Shrews: btw - sandy and I will be coming through durham after thanksgiving. you gonna be around for beer? | 19:41 |
| Shrews | mordred: likely will be | 19:42 |
| clarkb | Shrews: we should probably use scala or haskell if the lack of functions is a problem. those languages are functional | 19:43 |
| *** sarob has quit IRC | 19:44 | |
| clarkb | fungi: sdague: I am going to stop trying to push those d-g changes through until later today. Feel free to kick them if you want to babysit | 19:44 |
| sdague | clarkb: dammit you missed the obvious go joke | 19:44 |
| *** sarob has joined #openstack-infra | 19:44 | |
| clarkb | sdague: what was the go joke? | 19:45 |
| fungi | go | 19:45 |
| fungi | it is its own joke | 19:45 |
| sdague | any time we bring up another language, it must be go :) | 19:45 |
| clarkb | sdague: oh ya fail (ganeti is actually partially writtin in haskell so clearly Go isn't the right choice for us, haskell would be better) | 19:47 |
| clarkb | sdague: jog0: do the swift tests in tempest run in different classes? any chance we can consolidate them so they run in a single process? | 19:47 |
| clarkb | I hate that idea but if it gets things moving again it may not be the absolute worst idea I have ever had | 19:47 |
| sdague | so I'm not sure why that would help | 19:48 |
| clarkb | sdague: the problem seems to be around concurrent requests (DB locked) | 19:48 |
| sdague | well, we could always just lock all the swift tests | 19:48 |
| sdague | so only one runs at a time | 19:48 |
| *** sarob has quit IRC | 19:49 | |
| sdague | honestly, it feels like we need swift team engagement here instead of just shooting in the dark | 19:49 |
| *** adalbas has quit IRC | 19:49 | |
| clarkb | ++ | 19:49 |
| clarkb | they are in their meeting right now | 19:49 |
| sdague | ryanpetrello was hoping this could land before he cut a release, to ensure all was good - https://review.openstack.org/#/c/54333/ | 19:51 |
| jog0 | mordred: https://review.openstack.org/#/c/53372/7 | 19:52 |
| jog0 | where are the test-req cahnges | 19:52 |
| jog0 | changes* | 19:53 |
| jog0 | clarkb: what sdague said | 19:53 |
| clarkb | jog0: mordred for 53372, I think we need to look at that machinery again. there is at least one bug where stable branch changes are applied to master branch | 19:53 |
| clarkb | which derps things pretty hard (I think that is why keystoneclient changes in that patchset) | 19:54 |
| jog0 | clarkb: lolz | 19:54 |
| *** hogepodge has quit IRC | 19:54 | |
| *** DennyZhang has joined #openstack-infra | 19:55 | |
| *** jamesmcarthur has joined #openstack-infra | 19:55 | |
| *** nibalizer has joined #openstack-infra | 19:56 | |
| jog0 | bad jenkins | 19:57 |
| sdague | ianw: hey, it would be super awesome if you could actually view the logs that RedHat-CI is posting back up - http://people.redhat.com/~iwienand/55992/ | 19:58 |
| sdague | when I try to in the browser it says they are corrupted | 19:59 |
| *** sandywalsh has quit IRC | 19:59 | |
| notmyname | sdague: not sure I followed, but if you want swift team engagement, ask in #openstack-swift | 19:59 |
| *** devvesa has quit IRC | 20:00 | |
| hub_cap | hey guys, -alt's default meeting topic is goofy fyi | 20:00 |
| hub_cap | not killing anyone but its funny | 20:00 |
| *** alcabrera is now known as alcabrera|afk | 20:02 | |
| *** ArxCruz has quit IRC | 20:03 | |
| *** alcabrera has joined #openstack-infra | 20:06 | |
| *** mrmartin has quit IRC | 20:07 | |
| *** lcestari has quit IRC | 20:07 | |
| *** alcabrera|afk has quit IRC | 20:09 | |
| *** SergeyLukjanov has quit IRC | 20:10 | |
| *** sandywalsh has joined #openstack-infra | 20:11 | |
| *** DennyZhang has quit IRC | 20:16 | |
| *** marun has quit IRC | 20:16 | |
| *** sandywalsh has quit IRC | 20:17 | |
| *** toddmorey has quit IRC | 20:18 | |
| *** krtaylor has joined #openstack-infra | 20:25 | |
| *** sarob has joined #openstack-infra | 20:26 | |
| *** Ryan_Lane has quit IRC | 20:27 | |
| *** Ryan_Lane has joined #openstack-infra | 20:28 | |
| *** Ryan_Lane has quit IRC | 20:28 | |
| *** krotscheck has quit IRC | 20:30 | |
| *** sarob has quit IRC | 20:31 | |
| *** sandywalsh has joined #openstack-infra | 20:31 | |
| *** Ryan_Lane has joined #openstack-infra | 20:36 | |
| *** toddmorey has joined #openstack-infra | 20:37 | |
| *** dprince has quit IRC | 20:42 | |
| openstackgerrit | Jeremy Stanley proposed a change to openstack-infra/config: Stop running stable/folsom jobs https://review.openstack.org/56277 | 20:42 |
| fungi | clarkb: ^ first stab at the config changes | 20:42 |
| fungi | i'll work up the d-g and devstack changes next and -2 them until that clears | 20:43 |
| fungi | we also need to remember to adjust the all-projects acl since that's not stored in git | 20:43 |
| *** sarob has joined #openstack-infra | 20:44 | |
| *** jamesmcarthur has quit IRC | 20:46 | |
| jog0 | sdague: did you see this http://julien.danjou.info/blog/2013/openstack-summit-icehouse-ceilometer | 20:47 |
| jog0 | 'Ultimately, I would also want Ceilometer moving towards providing its own set of Tempest tests as part of the code base. That way, it'd be as easy for core reviewers to refuse a patch if it doesn't provide functional tests as it is to refuse it if it doesn't provide unit tests. As we'll do for the documentation. | 20:48 |
| *** acabrera has joined #openstack-infra | 20:48 | |
| *** alcabrera has quit IRC | 20:49 | |
| *** boris-42 has quit IRC | 20:49 | |
| *** jamesmcarthur has joined #openstack-infra | 20:54 | |
| *** herndon_ has joined #openstack-infra | 20:57 | |
| *** acabrera is now known as alcabrera | 20:57 | |
| sdague | yeh, I think we disagree there. But we can work that one out later | 20:58 |
| sdague | I realized at summit how little folks understand the integrated gate | 20:58 |
| sdague | which was one of the reasons for that state of the world part of my incubation / integration threshold proposal | 20:59 |
| *** Ryan_Lane has quit IRC | 20:59 | |
| *** Ryan_Lane1 has joined #openstack-infra | 20:59 | |
| *** hogepodge has joined #openstack-infra | 21:00 | |
| ianw | sdague: hmm, sorry about that, it appears they're not really zipped. let me dig into the script posting them | 21:02 |
| zaro | mgagne: around? | 21:03 |
| mgagne | zaro: sup | 21:04 |
| zaro | mgagne: i can't get this to work. https://review.openstack.org/#/c/54003 | 21:04 |
| mgagne | zaro: checking (still suffering from jet lag) | 21:04 |
| *** saper has quit IRC | 21:05 | |
| zaro | mgagne: i setup the yaml conf but when i run it the code for inject-password never gets executed. i don't see what is the problem. | 21:06 |
| *** hogepodge has quit IRC | 21:07 | |
| mgagne | zaro: http://paste.openstack.org/show/52539/ | 21:09 |
| mgagne | zaro: Jenkins ver. 1.535 | 21:10 |
| zaro | mgagne: let me try with your config. i'm already using latest jenkins ver 1.539 | 21:10 |
| mgagne | zaro: I can try against latest | 21:11 |
| zaro | mgagne: first try was on 1.502 , but that didn't work either. | 21:11 |
| openstackgerrit | Jeremy Stanley proposed a change to openstack-infra/devstack-gate: Remove folsom-specific logic https://review.openstack.org/56284 | 21:11 |
| *** mrmartin has joined #openstack-infra | 21:13 | |
| *** hogepodge has joined #openstack-infra | 21:14 | |
| zaro | mgagne: here's my config http://paste.openstack.org/show/52540/ | 21:15 |
| fungi | clarkb: ^ there's the other one. looks like devstack itself doesn't have anything in need of cleaning up | 21:15 |
| mgagne | zaro: Jenkins 1.539 + Environment Injector Plugin 1.89: works fine | 21:15 |
| * fungi disappears for a bit to start cooking dinner | 21:15 | |
| zaro | mgagne: the inject-password does not write out anything, but the envfile does work just fine. | 21:15 |
| mgagne | zaro: I see where this is going... | 21:15 |
| zaro | mgagne: ohh i should check the plugin version. | 21:16 |
| openstackgerrit | Michael Still proposed a change to openstack-infra/jeepyb: Allow automatic subscription to DocImpact bugs https://review.openstack.org/56158 | 21:16 |
| zaro | mgagne: i had envinject ver 1.88, will try with 1.89 to see | 21:17 |
| mgagne | zaro: I can't get the properties file to be set | 21:17 |
| mgagne | zaro: I think both configs are conflicting | 21:17 |
| zaro | mgagne: haha!! | 21:17 |
| mgagne | zaro: each overriding each others | 21:17 |
| *** mrmartin has quit IRC | 21:18 | |
| mgagne | zaro: or not | 21:18 |
| zaro | mgagne: actuualy it shouldn't matter because i'm just using jjb in test mode. | 21:19 |
| clarkb | fungi: will look shortly | 21:19 |
| clarkb | sdague: what sorts of misunderstanding of the gate are you seeing? is it the general how the gate works and why symetric vs asymmetric matters? | 21:20 |
| mgagne | zaro: can you post the XML? | 21:20 |
| mgagne | zaro: with your config: http://paste.openstack.org/show/52545/ | 21:20 |
| zaro | mgagne: here's mine http://paste.openstack.org/show/52546/ | 21:21 |
| mgagne | zaro: where do I complain about paste.o.o being slow? =) | 21:21 |
| *** julim has quit IRC | 21:21 | |
| mgagne | zaro: did you forget to checkout the change/review ? | 21:22 |
| *** jhesketh__ has joined #openstack-infra | 21:22 | |
| zaro | mgagne: you complain about it here. we host it. | 21:22 |
| *** hogepodge has quit IRC | 21:23 | |
| zaro | mgagne: complain to clarkb or fungi | 21:23 |
| *** jamesmcarthur has quit IRC | 21:23 | |
| *** jhesketh has joined #openstack-infra | 21:23 | |
| mgagne | zaro: well, does it feel slow to you? | 21:23 |
| zaro | mgagne: hey, looks like it all works for you. i do have it checked out, code is there. | 21:23 |
| zaro | mgagne: not really | 21:24 |
| zaro | mgagne: moved my jenkins envinject plugin to 1.89 but still not working. argg! | 21:26 |
| *** Ryan_Lane1 is now known as Ryan_Lane | 21:26 | |
| *** Ryan_Lane has joined #openstack-infra | 21:26 | |
| mgagne | zaro: can't get env-file to work | 21:27 |
| zaro | mgagne: what? it looks like paste 52545 shows envfile just fine. | 21:28 |
| mgagne | zaro: not showing up in UI | 21:28 |
| zaro | mgagne: ohh it's probably because you don't have that plugin installed. | 21:28 |
| clarkb | fungi: https://review.openstack.org/#/c/56277/1/modules/openstack_project/files/zuul/layout-dev.yaml does that make the postgres job non voting | 21:28 |
| mgagne | zaro: "Inject environment variables to the build process" ? | 21:28 |
| clarkb | fungi: it seems like the job isn't currently voting but it is | 21:28 |
| clarkb | fungi: any idea of what is going on there? | 21:29 |
| zaro | mgagne: no, that's a different plugin. it's the envfile plugin. | 21:29 |
| mgagne | zaro: damn | 21:29 |
| zaro | mgagne: your generated xml looks good. that's what i want to get but i can't seem to get it. | 21:29 |
| mgagne | zaro: how about you clean jjb from python lib? | 21:30 |
| zaro | mgagne: ok. i see. maybe it's my python env. i'll try to setup a clean virtualenv and see what happens. | 21:31 |
| mikal | Is a -1 sufficient to kill a pending merge? | 21:31 |
| mikal | Or does it need to be a -2? | 21:31 |
| mgagne | zaro: wait =) | 21:32 |
| clarkb | fungi: nevermind that is the dev file | 21:32 |
| clarkb | mikal: needs -2 | 21:32 |
| clarkb | mikal: assuming that there is still at least one + | 21:32 |
| mgagne | zaro: now that I installed env-file, it broke inject-passwords | 21:32 |
| clarkb | *at least one +2 | 21:32 |
| mikal | Ugh | 21:33 |
| mikal | I can't -2 the patchset currently merging | 21:33 |
| zaro | mgagne: did you restart jenkins? | 21:33 |
| mgagne | zaro: yes, testing something | 21:33 |
| mikal | As it is no longer the most recent patchset | 21:33 |
| mikal | Will a -2 on the most recent patchset work? | 21:33 |
| clarkb | mikal: if there is a more recent patchset that will prevent merging as well | 21:34 |
| clarkb | mikal: you can only merge the latest patchset | 21:34 |
| mikal | clarkb: oh, cool | 21:34 |
| clarkb | -2'ing the latest patchset will do it too | 21:34 |
| clarkb | mikal: can I see the change? | 21:34 |
| mikal | https://review.openstack.org/#/c/55117/ | 21:35 |
| mikal | Its not in the gate any more it seems | 21:35 |
| mikal | But not message in the review saying the merge terminated early | 21:35 |
| mikal | s/not/no/ | 21:35 |
| clarkb | mikal: ok, I think zuul silently drops it when the new patchset shows up | 21:35 |
| mikal | Ahhh, ok | 21:36 |
| clarkb | yup you are good | 21:36 |
| mikal | Excellent, thanks | 21:36 |
| mikal | For my next trick I shall email openstack-dev and ask for reviews of my jeepyb patch without indicating what project the change is for | 21:37 |
| mgagne | zaro: ok, both plugins are working. I forgot I reverted to master meanwhile. | 21:37 |
| *** yolanda has quit IRC | 21:38 | |
| zaro | mgagne: are you in a virt env? | 21:39 |
| mgagne | zaro: no | 21:39 |
| mgagne | zaro: testing with Ubuntu 12.04 | 21:39 |
| zaro | mgagne: maybe wondering if one of the dependent python libs are causing my problem. | 21:40 |
| *** Ryan_Lane has quit IRC | 21:40 | |
| *** Ryan_Lane has joined #openstack-infra | 21:40 | |
| *** krtaylor has quit IRC | 21:40 | |
| zaro | mgagne: actually that probably doesn't make sense because the envfile one works. | 21:40 |
| mgagne | zaro: how about you add a big print and see if it gets executed | 21:41 |
| zaro | mgagne: tried that. it doesn't get executed at all. | 21:41 |
| mgagne | root@jenkins:/opt/jenkins-job-builder# md5sum jenkins_jobs/modules/wrappers.py | 21:42 |
| mgagne | ae7c9bff0981ae7c3f790e49d31f88b0 jenkins_jobs/modules/wrappers.py | 21:42 |
| zaro | mgagne: not sure why the other wrappers get executed, but not the inject password one. | 21:42 |
| mgagne | lets see if we have the same file | 21:42 |
| devananda | question about sample config files -- do most projects regenerate their sample config manually, or is there an infra hook to do it on each commit? | 21:43 |
| zaro | ae7c9bff0981ae7c3f790e49d31f88b0 jenkins_jobs/modules/wrappers.py | 21:44 |
| zaro | mgagne: ^ | 21:44 |
| mgagne | zaro: :-/ is inject-passwords in setup.py ? | 21:44 |
| zaro | mgagne: yes, it's there. | 21:44 |
| mgagne | zaro: I think your env is cursed | 21:44 |
| zaro | mgagne: tools/jenkins-jobs.py --conf ./jenkins_jobs.ini test ./config -o ./out | 21:44 |
| zaro | mgagne: that's how i execute | 21:45 |
| *** rockstar has joined #openstack-infra | 21:45 | |
| *** rockstar has joined #openstack-infra | 21:45 | |
| mgagne | zaro: jenkins-jobs --flush-cache test -o /tmp/jjb /etc/jenkins_jobs/config | 21:46 |
| mgagne | zaro: output here: http://paste.openstack.org/show/52555/ | 21:46 |
| mgagne | zaro: yaml here: http://paste.openstack.org/show/52559/ | 21:46 |
| zaro | mgagne: yeah. everything looks like it should work, but doesn't. i'll get the MAN to take a look. | 21:49 |
| zaro | mgagne: thanks for helping. | 21:49 |
| mgagne | zaro: only thing left to try: creating a new venv | 21:50 |
| *** hashar has quit IRC | 21:50 | |
| *** sdake_ has quit IRC | 21:50 | |
| *** sdake has quit IRC | 21:51 | |
| zaro | mgagne: tried that. i've being trying your yaml within a new venv, but still not working. | 21:51 |
| *** zehicle_at_dell has quit IRC | 21:51 | |
| *** zehicle_at_dell has joined #openstack-infra | 21:51 | |
| *** ArxCruz has joined #openstack-infra | 21:52 | |
| mgagne | zaro: and have you tried to add the big print to a well known working plugin instead? just to see if you are hitting the right version of the file | 21:52 |
| zaro | mgagne: you mean jjb plugin? i put the print in the inject-passwords function in jjb | 21:53 |
| mgagne | zaro: how about env-file ? | 21:53 |
| *** sdake has joined #openstack-infra | 21:54 | |
| zaro | mgagne: yes, put in both. only one in env-file executes. | 21:54 |
| *** sdake_ has joined #openstack-infra | 21:55 | |
| *** sdake_ has quit IRC | 21:55 | |
| *** sdake_ has joined #openstack-infra | 21:55 | |
| mgagne | zaro: wth... | 21:55 |
| *** krtaylor has joined #openstack-infra | 21:56 | |
| *** sarob has quit IRC | 21:59 | |
| *** sarob has joined #openstack-infra | 22:00 | |
| *** oubiwann has quit IRC | 22:01 | |
| *** sarob has quit IRC | 22:04 | |
| notmyname | lifeless: around? | 22:05 |
| *** hogepodge has joined #openstack-infra | 22:08 | |
| *** hogepodge has quit IRC | 22:09 | |
| zaro | mgagne: clarkb figured it out. | 22:10 |
| mgagne | zaro: tell me! | 22:10 |
| *** dkranz has quit IRC | 22:11 | |
| mgagne | zaro: you were typing on the wrong keyboard? :D | 22:11 |
| zaro | mgagne: it was that i installed jjb from pipy just to get all of it's dependencies. then i uninstalled jjb so i can run my version locally. | 22:12 |
| lifeless | notmyname: hi | 22:13 |
| mgagne | zaro: why was your print working then? | 22:13 |
| clarkb | mgagne: zaro: I think python found some in between version somewhere | 22:13 |
| *** pcrews has quit IRC | 22:14 | |
| zaro | when i installed jjb from local, i.e. pip install . | 22:14 |
| zaro | mgagne: it worked. | 22:14 |
| zaro | mgagne: i think it was still referencing the version i installed from pypi somehow. | 22:15 |
| zaro | mgagne: that's why the print for inject-password didn't execute, probably because it wasn't even there in the old version. | 22:16 |
| *** mfer has quit IRC | 22:16 | |
| notmyname | lifeless: I wanted to ask you about hardware stuff again | 22:17 |
| notmyname | lifeless: testing clusters and all that | 22:17 |
| melwitt | hi, is it possible to see coverage reports in jenkins? for example, I go to https://jenkins.openstack.org/job/neutron-coverage/165/ but I don't see a way to see the html report | 22:17 |
| notmyname | lifeless: should we discuss in here, or somewhere else? (I'm not sure if it would distract -infra or -swift) | 22:18 |
| clarkb | melwitt: couple things to know. We have three jenkins masters so the latest job may not have been on that particular jenkins server and second we don't let jenkins render the coverage itself since doing so is slow and when you have three masters that doesn't work so well | 22:19 |
| *** weshay has quit IRC | 22:20 | |
| clarkb | melwitt: instead all build logs are copied to http://logs.openstack.org/. You will find coverage under http://logs.openstack.org/$FIRST_2_CHARS_OF_GIT_SHA/GIT_SHA for the change you want coverage on | 22:20 |
| lifeless | notmyname: here is fine I think. | 22:20 |
| clarkb | fungi: sdague: I have rekicked d-g changes | 22:21 |
| lifeless | notmyname: for starters - wiki.o.o/wiki/TripleO/TripleOCloud | 22:21 |
| notmyname | lifeless: ok. so what are you planning? | 22:21 |
| notmyname | ok | 22:21 |
| lifeless | and .../Regions | 22:21 |
| melwitt | clarkb: perfect, thank you! | 22:21 |
| *** pcm_ has quit IRC | 22:22 | |
| clarkb | melwitt: np let me know if you have more questions | 22:22 |
| melwitt | much appreciated | 22:23 |
| *** dkliban has quit IRC | 22:23 | |
| *** sarob has joined #openstack-infra | 22:24 | |
| *** pcrews has joined #openstack-infra | 22:29 | |
| reed | may all the known and unknown gods deliver pain for a very prolonged time to anybody who thinks software patents are a good idea | 22:31 |
| reed | may the channel repeat:" amen" | 22:31 |
| clarkb | amen | 22:31 |
| *** sarob_ has joined #openstack-infra | 22:31 | |
| mordred | amen | 22:32 |
| reed | we may want to start with this prayer all our meetings :) | 22:33 |
| *** alcabrera has quit IRC | 22:33 | |
| notmyname | lifeless: how is code deployed to the tripleo cloud? ie does it have to go through a certain orchestration framework? | 22:33 |
| reed | do you know if we ship modified versions of software licensed with BSD or other permissive license in openstack? | 22:34 |
| lifeless | notmyname: heat + nova baremetal | 22:34 |
| lifeless | notmyname: e.g. 'openstack deployment' :) | 22:34 |
| *** sarob has quit IRC | 22:35 | |
| *** loq_mac has joined #openstack-infra | 22:35 | |
| *** sarob_ has quit IRC | 22:36 | |
| notmyname | lifeless: well, (heat + nova) deployment ;-) | 22:36 |
| clarkb | reed: can you define ship? | 22:36 |
| *** loq_mac has quit IRC | 22:36 | |
| mordred | clarkb: I was just about to ask the same question | 22:36 |
| reed | clarkb, code that we host on our git repositories or code that we package in the official tarballs | 22:37 |
| *** danger_fo is now known as danger_fo_away | 22:37 | |
| *** yfried has quit IRC | 22:37 | |
| notmyname | lifeless: what do you imagine the steps to be to deploy swift across some subset of the machines? | 22:38 |
| reed | clarkb, if I git clone an openstack repository, is there a chance I get also BSD licensed, *derivative* code for which the foundation has no *copyright*? | 22:38 |
| *** jpeeler has quit IRC | 22:38 | |
| *** mriedem has quit IRC | 22:39 | |
| lifeless | notmyname: no specifically - 'OpenStack Deployment' - thats the program name; we're deploying OpenStack in the test cluster as a functional test of OpenStack Deployment | 22:39 |
| *** blamar has quit IRC | 22:39 | |
| clarkb | reed: there absolutely is a 100% chance you will get code that the foundation has no copyright on (because the foundation/cla does not require copyright assignment) | 22:39 |
| lifeless | notmyname: to deploy swift on some of the machines - there are several scenarios: | 22:40 |
| clarkb | reed: and depending on the project yes, there is some bundling of very permissive source code | 22:40 |
| lifeless | - part of the deployed TripleOCloud - the trunk CD deployed cloud | 22:40 |
| clarkb | eg BSD | 22:40 |
| lifeless | - as gating tests | 22:40 |
| lifeless | - as check tests | 22:40 |
| mordred | reed: I agree with everything clarkb just said | 22:40 |
| lifeless | for the first case, we're about to start doing that - the patches are going into tripleo right now | 22:40 |
| lifeless | for the second case, we need two regions, and we need a multinode test definition for it | 22:41 |
| openstackgerrit | A change was merged to openstack-infra/jenkins-job-builder: Inject passwords to the build as environment variables support added https://review.openstack.org/54003 | 22:41 |
| clarkb | mordred: pbr for example bundles distutils2 or whatever it was called right? | 22:41 |
| lifeless | for the third case, we're bringing up the emulated bm environment for it now - https://etherpad.openstack.org/p/tripleo-test-cluster | 22:41 |
| mordred | clarkb: that's right, pbr includes a forked copy of code that used to be d2to1 | 22:41 |
| *** jamesmcarthur has joined #openstack-infra | 22:42 | |
| notmyname | lifeless: is the first one involved in any testing? | 22:43 |
| *** Hunner is now known as Hunner_irssi | 22:43 | |
| lifeless | notmyname: it deploys in a while loop (literally - while True: deploy() :)) | 22:44 |
| *** Hunner_tail is now known as Hunner | 22:44 | |
| lifeless | notmyname: from trunk, so it gets signal fairly quickly, but it doesn't stop patches landing. | 22:44 |
| lifeless | notmyname: we're working up the path for testing - we have experimental, heading towards silent now. | 22:44 |
| notmyname | lifeless: ah ok. so a separate process could be testing the cluster? | 22:45 |
| *** CaptTofu has quit IRC | 22:45 | |
| lifeless | notmyname: we plan to run tempest against the deployed cluster as well | 22:45 |
| lifeless | notmyname: thats also in progress ;) | 22:45 |
| fungi | clarkb: i think you probably already figured it out, but going through scrollback now... removing the branch negation pattern on that job only changes when it runs, not whether it votes | 22:45 |
| *** CaptTofu has joined #openstack-infra | 22:46 | |
| lifeless | notmyname: so yes, separate process, for the first case | 22:46 |
| reed | clarkb, by copyright I meant to say [I|C]CLA | 22:46 |
| notmyname | lifeless: but that's only black-box tests (ie functional tests from an external perspective). anything else would need to be part of your gate or check tests, right? | 22:46 |
| lifeless | we're pretty flexible | 22:46 |
| lifeless | what sort of thing do you want? | 22:46 |
| clarkb | fungi: ya, I couldn't figure out why it was set to voting: false, but I think that is because it is the dev layout | 22:46 |
| clarkb | prod layout doesn't do that | 22:47 |
| mordred | reed: well, it is not possible for code to enter our systems without someone submitting it via CLA | 22:47 |
| *** thomasem has quit IRC | 22:47 | |
| reed | clarkb, the bundling that you mentioned, is it just an import 'verbatim' from a remote repo or is it code we have forked and maintain separately in our repo? | 22:47 |
| mordred | reed: so I, for instance, submitted the BSD code in pbr to the project under ther terms of the CLA | 22:47 |
| *** flaper87 is now known as flaper87|afk | 22:47 | |
| notmyname | lifeless: for swift, "kill a particular server/service; PUT and object; GET the object; restore the server/service; GET the object" etc | 22:47 |
| mordred | reed: forked and maintain | 22:47 |
| reed | mordred, interesting | 22:47 |
| clarkb | all code must be submitted by people who have signed a CLA, but that code may come from permissively licensed locations | 22:47 |
| clarkb | reed: apache foundation has a doc on this | 22:47 |
| mordred | that's correct | 22:47 |
| reed | clarkb, do you have a pointer? | 22:48 |
| notmyname | lifeless: we have those now in our probetests, but they are geared specifically for a a dev environment, not a cluster | 22:48 |
| clarkb | reed: ya looking it up now | 22:48 |
| lifeless | notmyname: so a busy loop doing PUT/GET during a deployment would have that happen ;) | 22:48 |
| clarkb | reed: http://www.apache.org/legal/3party.html category A | 22:48 |
| lifeless | notmyname: since a deployment is a rolling wave of downtime on individual machines/processes | 22:48 |
| notmyname | lifeless: no. I mean I understand that, but that's not what I want. it's good to have, but not what I want most | 22:49 |
| *** whoops has joined #openstack-infra | 22:49 | |
| lifeless | ok | 22:49 |
| lifeless | so, you want a cluster wide test where the driver can poke at specific services to force them up/down ? | 22:49 |
| notmyname | lifeless: what I want most to to explicitly trigger and test certain failure conditions and maybe config parameters | 22:49 |
| notmyname | lifeless: correct | 22:49 |
| lifeless | how big a cluster do you need? | 22:49 |
| notmyname | lifeless: well, I'm currently deciding between your cluster and going through the red tape and setup time of getting donated hardware | 22:50 |
| notmyname | lifeless: 10 machines would be good | 22:51 |
| lifeless | notmyname: we're getting donated hardware too... so I would worry a little about partitioning the donors | 22:51 |
| lifeless | notmyname: do you want this running continually, or a once-a-day job kind of thing? | 22:51 |
| notmyname | lifeless: like you, I have some verbal commitments already :-) | 22:51 |
| lifeless | notmyname: cool | 22:52 |
| notmyname | lifeless: depends on the scale. with a small cluster, it would need to be once-a-day because it wouldn't be able to keep up with the patch queue | 22:52 |
| notmyname | lifeless: ideally, I'd like to see it as a gate for every patch | 22:52 |
| lifeless | ok, so that has specific infra requirements | 22:52 |
| notmyname | sure, but I'm also wanting to start small so it might actually get done :-) | 22:52 |
| lifeless | right, just its good to know so you can set expectations | 22:53 |
| lifeless | https://review.openstack.org/#/c/55761/ | 22:53 |
| lifeless | is my attempt to document what was in the infra teams heads about this | 22:53 |
| notmyname | great | 22:53 |
| jog0 | anyone else getting some grenade failures | 22:54 |
| openstackgerrit | lifeless proposed a change to openstack-infra/config: Document what it takes to be a check/gate test. https://review.openstack.org/55761 | 22:54 |
| lifeless | notmyname: so - 10 machines once a day we can do right now I think | 22:54 |
| jog0 | http://logs.openstack.org/35/51735/7/check/check-grenade-devstack-vm/a39edbe/console.html#_2013-11-13_20_52_57_791 | 22:54 |
| notmyname | lifeless: FWIW, I'll shoot for the third-party tests if I use my own hardware | 22:54 |
| notmyname | lifeless: ah ok | 22:55 |
| lifeless | notmyname: to move to a gate we'll need two clusters scaled for running 40 test runs per hour | 22:55 |
| notmyname | lifeless: and we could deploy swift on bare metal there? any chance we could have static IPs etc to make config easier? | 22:55 |
| lifeless | notmyname: which if you need 10 machines in a cluster may be a considerable number of machines | 22:55 |
| fungi | devananda: projects regenerate their samples manually from what i've seen. reviewers seem to request that from a contributor when they see changes which would impact configuration options | 22:55 |
| clarkb | jog0: looks like the rackspace ubuntu mirror failed us | 22:55 |
| notmyname | lifeless: I rounded up to get to 10 machines ;-) | 22:56 |
| lifeless | notmyname: IP allocation is all up to nova/neutron : hard coding would be a deep mistake. | 22:56 |
| devananda | fungi: that's what i thought. thanks | 22:56 |
| notmyname | I can do more with more servers :-) | 22:56 |
| lifeless | notmyname: right, we all can :) | 22:56 |
| notmyname | lifeless: hmm..ok. that means we'd have to have some sort of template for generating configs and rings | 22:56 |
| lifeless | notmyname: so I want to see the hardware donated to the TripleOCloud put to good use for all of OpenStack | 22:56 |
| fungi | reed: amen and hallelujah brother | 22:56 |
| lifeless | notmyname: yes, we're doing that via heat. | 22:56 |
| fungi | reed: also we have usgcla and content coming in via usa public domain | 22:57 |
| lifeless | notmyname: it's a problem we have to solve, or we can't deploy swift via ehat. | 22:57 |
| jog0 | clarkb: http://graphite.openstack.org/graphlot/?width=586&from=00%3A00_20131112&_salt=1384383455.631&height=308&until=23%3A59_20131113&target=stats.zuul.pipeline.check.job.check-grenade-devstack-vm.FAILURE | 22:57 |
| jog0 | wil just use recheck thanks clarkb | 22:58 |
| lifeless | notmyname: so I'm very happy to do anything I can to help swift get solid functional gate tests. | 22:58 |
| lifeless | notmyname: I think we should be discussing the form of them with sdague | 22:58 |
| lifeless | notmyname: since being able to poke at cloud internals in multinode tests is something we'll want for all projects | 22:58 |
| lifeless | notmyname: there's no obvious reason to me why we can't express your testing needs in Tempest | 22:59 |
| reed | fungi, that is less of an issue since it's code that comes in via the Individual CLAs, too | 22:59 |
| notmyname | lifeless: the issue has simply been one of prioritization. we already have lots of tests (not in tempest) and other orgs have their QA tests (not in tempest) | 22:59 |
| notmyname | lifeless: we have 3 kinds of tests in swift: unit, functional, and probe | 23:00 |
| lifeless | notmyname: sure, but we're gearing up to run tempest against the CD deployed cloud: if we do this right you get your tests running automatically with all the goodness glued in. | 23:00 |
| lifeless | notmyname: I think thats a lot better than having to run up N different environments - unless there is something unique about the test requirements? | 23:01 |
| notmyname | lifeless: unfortunately, the only cluster-style tests that are done for swift are done by orgs in their interal QA. all openstack tests are done against simulated clusters | 23:01 |
| lifeless | Right, but that is changing. | 23:01 |
| notmyname | well, that's what we're talking about right? :-) | 23:01 |
| lifeless | Indeed :) | 23:01 |
| lifeless | russellb: do you want 2x+2 for reviewstats now, or just +2 from not-author ? | 23:02 |
| lifeless | russellb: and have you put any thought into having a gate job for it ? | 23:03 |
| notmyname | lifeless: do you have a team either at HP or within tripleo that is working on building the tests and deployment tools necessary to do all of this? | 23:03 |
| sdague | so I guess the question is about the gate though, right now swift is bouncing the gate pretty often | 23:03 |
| lifeless | sdague: it's about all the things :) | 23:03 |
| sdague | lifeless: agreed :) | 23:03 |
| sdague | but I'd rather not have swift bouncing out everyone else's code from the gate preventing them from merging in the near term | 23:04 |
| fungi | reed: actually, the usgcla group members do not have any icla enforcement. unlike the ccla, which still requires the icla to be agreed | 23:04 |
| sdague | I think it caused more resets than neutron today :) | 23:04 |
| lifeless | notmyname: the tripleo team is working on deployment right now; we're delegating test content to tempest (as we should :)) but will get involved there as needed to get to completely robust deployments with early detection of faults | 23:04 |
| notmyname | sdague: aside from the brief conversations you and I have had, I'm not aware of this. you should mention it in -swift if you want swift team assistance | 23:04 |
| sdague | notmyname: you know, we do talk about other projects, not just in their project specific channels :) like nova and neutron show up in -infra and -qa as well, we're all one happy family | 23:05 |
| lifeless | everyone should be in all the channels :P | 23:05 |
| lifeless | more seriously, I think project specific folk should be in their project + qa + infra + -dev | 23:06 |
| notmyname | sdague: indeed, and I'm always in here. as are a few other swift devs. but we all started with "everyone in one channel" originally, and it's not a scalable model. no need to recreate -dev from 2 years ago here in -infra | 23:06 |
| sdague | honestly, jog0 and clarkb have been the ones running at this one, as unfortunately I don't have the time to do the debug this one | 23:06 |
| fungi | sdague: still the thing where we've bumped the loopback to 4g but swift is being asked to create more than that (presumably by glance during some parallel overlap)? | 23:06 |
| sdague | fungi: yes | 23:06 |
| *** oubiwann has joined #openstack-infra | 23:06 | |
| sdague | notmyname: typically gate resets are done here, because you need lots of infra support to handle the situation | 23:07 |
| russellb | lifeless: 1 non-author is probably fine for now. and as for gate job, i've thought about it :) | 23:07 |
| russellb | haven't done anything | 23:07 |
| notmyname | sdague: where is the alerting for this. have I missed it in emails? | 23:07 |
| openstackgerrit | A change was merged to openstack-infra/reviewstats: Show how many patch sets were created https://review.openstack.org/55996 | 23:07 |
| jog0 | notmyname: lets talk in openstack-swift about the issues | 23:07 |
| notmyname | sdague: ie how do we know that swift is breaking gates? | 23:07 |
| russellb | lifeless: i've been very slack on tasks in the category of making it prettier than a quick hack | 23:07 |
| lifeless | russellb: so, I may land broken code :> | 23:07 |
| lifeless | we shall see | 23:07 |
| russellb | it happens | 23:08 |
| russellb | not critical | 23:08 |
| jog0 | clarkb sdague: lets chat in -swift | 23:08 |
| sdague | jog0: I actually have to call it a day, but carry on :) | 23:08 |
| russellb | lifeless: i really need to get it running in infra ... | 23:08 |
| openstackgerrit | A change was merged to openstack-infra/reviewstats: Add received reviews to review stats https://review.openstack.org/55999 | 23:08 |
| russellb | ooh, yay | 23:09 |
| openstackgerrit | A change was merged to openstack-infra/reviewstats: Add some per day stats https://review.openstack.org/56200 | 23:09 |
| lifeless | russellb: gating before rehoming IMNSHO | 23:10 |
| *** nati_ueno has joined #openstack-infra | 23:10 | |
| russellb | lifeless: fair enough | 23:10 |
| lifeless | in terms of priority | 23:10 |
| russellb | right | 23:10 |
| lifeless | more people to debug issues is nice, less issues is better :) | 23:10 |
| notmyname | lifeless: what would be your timeframe on the changes needed to get swift deployed on the tripleo cluster? | 23:12 |
| lifeless | is there any word when we'll get the requirements job sorted out? Is someone tackling it? [https://review.openstack.org/#/c/55857/1/requirements.txt] | 23:13 |
| lifeless | notmyname: looking | 23:14 |
| lifeless | notmyname: so we need to land: | 23:14 |
| clarkb | lifeless: no one is tackling it as far as I know | 23:14 |
| lifeless | https://review.openstack.org/#/c/56012/ | 23:14 |
| lifeless | https://review.openstack.org/#/c/56006/ | 23:14 |
| *** dcramer_ has quit IRC | 23:14 | |
| openstackgerrit | A change was merged to openstack-infra/devstack-gate: Extract git_fetch_at_ref https://review.openstack.org/55029 | 23:14 |
| lifeless | https://review.openstack.org/#/c/56010/ | 23:14 |
| clarkb | I think we should abandon all of the existing reviews and possibly disable the job until it works properly | 23:14 |
| lifeless | and I think that will get -a- swift in the overcloud | 23:14 |
| clarkb | sdague: 55029 merged, just one more to go | 23:15 |
| lifeless | after that we need to decide on scale | 23:15 |
| lifeless | clarkb: Agreed | 23:15 |
| clarkb | mordred: does that sound good to you? or would you rather fix the job? | 23:15 |
| *** datsun180b has quit IRC | 23:15 | |
| notmyname | lifeless: ok, I'll look at those patches | 23:15 |
| *** denis_makogon has quit IRC | 23:16 | |
| reed | any names to suggest for people that were in HK and willing to join a user group meeting in SFO next week to talk about technical news in OpenStack or case studies? | 23:16 |
| clarkb | pleia2: ^ | 23:16 |
| openstackgerrit | A change was merged to openstack-infra/devstack-gate: trigger devstack-gate test as early as possible https://review.openstack.org/55030 | 23:16 |
| clarkb | reed: maybe jeblair but I am not sure he can be contacted this week | 23:16 |
| *** pabelanger has joined #openstack-infra | 23:16 | |
| fungi | clarkb: mordred: the job needs to figure out what branch the triggering change was merged to, so it can set the appropriate branch on projects being proposed, right? and the challenge is that zuul doesn't have that information in the calling environment for jobs in the post pipeline? | 23:17 |
| clarkb | woot d-g changes in /me starts watching the gate like a hawk | 23:17 |
| clarkb | fungi: correct | 23:17 |
| clarkb | fungi: I think it was mostly fixed | 23:17 |
| reed | clarkb, that's the first name that came to mind :) he'll be jet lagged and with lots of stuff to do before thanksgiving :) | 23:17 |
| clarkb | fungi: it may just need to have the old jobs abandoned so that we ditch the old bad patches rather than continuing to update them | 23:17 |
| fungi | reed: Ryan_Lane was in hk and is somewhere near sfo, yah? | 23:18 |
| *** jergerber has quit IRC | 23:18 | |
| clarkb | fungi: yup | 23:18 |
| reed | fungi, that's a good suggestion, thanks | 23:18 |
| lifeless | so what is the current story on copyright claims in new files? | 23:19 |
| lifeless | Needed? Optional? Discouraged? Forbidden? | 23:19 |
| clarkb | I have been going with optional | 23:19 |
| reed | lifeless, IIRC it is still needed | 23:19 |
| fungi | lifeless: encouraged if accurately represented | 23:19 |
| russellb | fungi: +1 | 23:19 |
| russellb | license bit, needed of course | 23:20 |
| lifeless | right | 23:20 |
| russellb | copyright bit, good when accurate, in practice, treated as optional | 23:20 |
| fungi | bad/misleading/incorrect copyright assertions in files are the real concern | 23:20 |
| lifeless | just wondering in solum reviews whether to ask for it. | 23:20 |
| clarkb | lifeless: solum is free to do whatever they like >_> | 23:20 |
| reed | russellb, anybody from RH that we could invite to talk in SFO about things that happened in HK? | 23:20 |
| Ryan_Lane | ? | 23:20 |
| *** zehicle_at_dell has quit IRC | 23:20 | |
| Ryan_Lane | ah | 23:20 |
| Ryan_Lane | I won't be in SF next week | 23:20 |
| clarkb | Ryan_Lane: reed is recruiting people to do a summit debrief in SFO | 23:20 |
| lifeless | clarkb: they want to aim at openstack/; they need to play by openstack/ rules from the start | 23:20 |
| Ryan_Lane | well, I may | 23:20 |
| Ryan_Lane | what day? | 23:21 |
| Ryan_Lane | I'm leaving town on the 20th | 23:21 |
| clarkb | lifeless: I thought they were explicitly not aiming at openstack? | 23:21 |
| russellb | reed: summit debrief for what audience? | 23:21 |
| russellb | hm? solum? | 23:21 |
| russellb | definitely openstack | 23:21 |
| russellb | it's an openstack thing, built on top of openstack apis | 23:21 |
| reed | Ryan_Lane, it's cloudscaling that is organizing next week's meetup, on Thursday evening | 23:21 |
| clarkb | russellb: integrating. I swear the introduction email was all "We like openstack but won't be integrated" | 23:22 |
| russellb | i'm drawing a blank on who we have out there | 23:22 |
| Ryan_Lane | yeah, won't be around, sorry | 23:22 |
| clarkb | but maybe that is because currently they are not integrated and I read it wrong | 23:22 |
| reed | russellb, typically the audience is operators, developers, end users... different and varied :) | 23:22 |
| russellb | clarkb: well, regardless, they're sticking to openstack processes and such as much as possible | 23:22 |
| russellb | such that they could be integrated someday, maybe, but if not, at least making it easy for openstack devs to join in | 23:22 |
| *** jamesmcarthur has quit IRC | 23:24 | |
| lifeless | clarkb: if they aren't aiming at openstack, not all that much point being in stackforge :> | 23:24 |
| russellb | heh, true | 23:24 |
| clarkb | lifeless: sure there is | 23:24 |
| russellb | but they are aiming at openstack | 23:24 |
| russellb | very much so | 23:24 |
| clarkb | lifeless: lots of stackforge projects seem happy to be stackforge | 23:24 |
| clarkb | russellb: gotcha | 23:24 |
| *** mrodden has quit IRC | 23:25 | |
| *** sarob has joined #openstack-infra | 23:26 | |
| fungi | clarkb: lifeless: yeah for example i don't think the puppet modules are aiming for any sort of official integration, but rather are using stackforge because it eases/increases contribution from the openstack devs | 23:26 |
| openstackgerrit | Clark Boylan proposed a change to openstack-infra/devstack-gate: Test change do not merge. https://review.openstack.org/56296 | 23:27 |
| clarkb | 56296 will double check that the recent d-g changes didn't expode anything | 23:27 |
| *** toddmorey has quit IRC | 23:28 | |
| *** sarob_ has joined #openstack-infra | 23:29 | |
| *** masayukig has joined #openstack-infra | 23:30 | |
| clarkb | aahahahahahah http://logs.openstack.org/96/56296/1/check/check-tempest-devstack-vm-full/f49ed68/console.html | 23:30 |
| clarkb | fungi: ^ FYI I am looking for a quick fix, will propose a revert if I can't sort it out | 23:30 |
| *** sarob has quit IRC | 23:30 | |
| *** dkranz has joined #openstack-infra | 23:31 | |
| clarkb | sdague: if you are still around you may find this interesting | 23:32 |
| clarkb | I think I see the problem | 23:34 |
| *** whoops has quit IRC | 23:35 | |
| *** esker has quit IRC | 23:37 | |
| lifeless | notmyname: would you recommend injecting a seed via Heat and regenerating the ring locally, or regenerating the ring externally and injecting the full file to each node? | 23:38 |
| *** sarob has joined #openstack-infra | 23:38 | |
| *** sarob_ has quit IRC | 23:38 | |
| openstackgerrit | Clark Boylan proposed a change to openstack-infra/devstack-gate: Actually create $BASE/new dir and set perms. https://review.openstack.org/56297 | 23:39 |
| sdague | clarkb: yeh, barely around, what's up? | 23:39 |
| notmyname | lifeless: it doesn't matter to me. both should work. ie it doesn't matter how you make the file. just that the same resulting ring file is everywhere | 23:39 |
| clarkb | sdague: 56297 should fix it | 23:39 |
| lifeless | notmyname: ok. No aesthetic preference? | 23:39 |
| notmyname | lifeless: depends if you have an admin server around somewhere | 23:39 |
| clarkb | sdague: basically need extra privs to create dirs in /opt then we need to chown the new dirs so that the jenkins user can do stuff with them | 23:39 |
| notmyname | lifeless: swift is designed to be flexible to your operational needs :-) | 23:39 |
| *** beagles has quit IRC | 23:39 | |
| lifeless | notmyname: I'm seeking an opinion :) | 23:40 |
| clarkb | sdague: that change may need to be forced in can you take a look at it? | 23:40 |
| lifeless | notmyname: or a decision tree | 23:40 |
| clarkb | if it looks good to you I will force approve before I revert stuff | 23:40 |
| clarkb | fungi: ^ | 23:40 |
| sdague | clarkb: gotcha, I think it looks good | 23:40 |
| * clarkb pushes the big merge button | 23:41 | |
| notmyname | lifeless: if it is easier to build a ring one place and then push it out everywhere, do it that way. otherwise, if it's easier to command and control a bunch of servers as individuals, do it that way | 23:41 |
| *** sarob has quit IRC | 23:41 | |
| *** sarob has joined #openstack-infra | 23:41 | |
| openstackgerrit | A change was merged to openstack-infra/devstack-gate: Actually create $BASE/new dir and set perms. https://review.openstack.org/56297 | 23:42 |
| lifeless | notmyname: thanks | 23:42 |
| jog0 | clarkb: can we add a df to the end of devstack-gate? | 23:42 |
| jog0 | after tempest runs | 23:42 |
| clarkb | jog0: sure | 23:42 |
| clarkb | I don't think that will expose and sensitive info | 23:42 |
| lifeless | mordred: you need to update https://review.openstack.org/#/c/54628/ | 23:42 |
| lifeless | mordred: and either say 'yea' or 'nah, just land' | 23:43 |
| clarkb | sdague: I force merged and my test change is being tested now | 23:43 |
| lifeless | clarkb: https://review.openstack.org/#/c/54793/ | 23:43 |
| sdague | clarkb: cool | 23:43 |
| lifeless | and https://review.openstack.org/#/c/54792/ | 23:43 |
| lifeless | though I think they are on a high stack that will stop them landing | 23:43 |
| sdague | so I guess it didn't quite test itself | 23:43 |
| clarkb | sdague: not quite, but now it should | 23:43 |
| sdague | I should have thought to put a dummy on the end | 23:43 |
| lifeless | clarkb: is | 23:44 |
| lifeless | 2013-11-12 18:26:12.287 | Downloading/unpacking markupsafe (from Jinja2->nova==2014.1.dev726.g30fa37e) | 23:44 |
| lifeless | 2013-11-12 18:26:12.405 | Could not find any downloads that satisfy the requirement markupsafe (from Jinja2->nova==2014.1.dev726.g30fa37e) | 23:44 |
| lifeless | a known issue? | 23:44 |
| jog0 | clarkb: what file would I put the df in? | 23:44 |
| clarkb | lifeless: not known to me | 23:44 |
| *** sarob_ has joined #openstack-infra | 23:44 | |
| clarkb | jog0: somewhere we run an iptables -L, I would find that and do the df there | 23:44 |
| lifeless | clarkb: https://review.openstack.org/#/c/54791/ | 23:44 |
| clarkb | jog0: and capture the result in the same way as the iptables listing | 23:45 |
| *** rcleere has quit IRC | 23:45 | |
| clarkb | lifeless: I think that job bypasses the mirror | 23:45 |
| clarkb | because it is used to test the mirror and may have just hit a pypi blip | 23:45 |
| jog0 | clarkb: thanks | 23:46 |
| *** sarob has quit IRC | 23:47 | |
| openstackgerrit | Joe Gordon proposed a change to openstack-infra/devstack-gate: Run df after gate https://review.openstack.org/56298 | 23:47 |
| jog0 | clarkb: ^ | 23:47 |
| clarkb | lifeless: gah I left a note on that change saying reverify is fine but it kept my approval so it is being retested anyways :) | 23:48 |
| *** sarob_ has quit IRC | 23:49 | |
| herndon_ | hi infra - I just got a timeout during vm boot in a jenkins review job. I'm making a ceilometer change, so I am pretty sure my code did not cause a problem. Also, all of the other review jobs passed. I don't see any existing bugs about this (the scenario test that failed is tempest.scenario.test_large_ops.TestLargeOpsScenario.test_large_ops_scenario). Should I file a new bug about this? If so, which project? | 23:51 |
| jog0 | herndon_: link? | 23:51 |
| clarkb | herndon_: yes, file a new bug. Figuring outwhere to file it is often the trickiest part. jog0 probably has ideas | 23:51 |
| clarkb | possibly tempest and/or nova? | 23:52 |
| herndon_ | dang, you guys respond quick! | 23:52 |
| *** krtaylor has quit IRC | 23:53 | |
| jog0 | herndon_: if you ahve a link to the reeview I can take a quick look | 23:53 |
| herndon_ | yeah, digging it up | 23:53 |
| herndon_ | http://logs.openstack.org/63/56063/3/check/gate-tempest-devstack-vm-neutron-large-ops/bdf8676/ | 23:53 |
| jog0 | ohh neutron is broken | 23:54 |
| jog0 | thats non-gating | 23:54 |
| jog0 | there is abug somewhere for that too | 23:54 |
| *** beagles has joined #openstack-infra | 23:54 | |
| *** gyee has quit IRC | 23:55 | |
| herndon_ | so... I can just ignore it? | 23:56 |
| jog0 | yup | 23:57 |
| herndon_ | thanks! | 23:57 |
| *** matsuhashi has joined #openstack-infra | 23:57 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!