| *** thuc has joined #openstack-infra | 00:00 | |
| *** zns has quit IRC | 00:04 | |
| *** jcooley_ has quit IRC | 00:04 | |
| *** jcooley_ has joined #openstack-infra | 00:05 | |
| *** mrodden has quit IRC | 00:05 | |
| clarkb | pleia2: https://review.openstack.org/#/c/69510/ has been reviewed | 00:05 |
|---|---|---|
| pleia2 | clarkb: thanks! I knew the rebase was coming | 00:06 |
| *** mgagne has quit IRC | 00:07 | |
| openstackgerrit | Khai Do proposed a change to openstack-infra/config: Fix update java alternatives on redhat systems https://review.openstack.org/77683 | 00:08 |
| clarkb | zaro: ^ well thats interesting | 00:08 |
| zaro | ugh! that alternatives patch failed the linter. try again ^ | 00:08 |
| clarkb | zaro: 'true' is the correct value | 00:08 |
| clarkb | not true | 00:09 |
| clarkb | everything ever is bad | 00:09 |
| zaro | what??? | 00:09 |
| clarkb | zaro: true is a boolean value | 00:09 |
| clarkb | we want to run the true command which is a string | 00:09 |
| jeblair | clarkb: your philosophy is truly dark today | 00:09 |
| clarkb | 'true' is correct | 00:09 |
| *** antonym has quit IRC | 00:10 | |
| clarkb | jeblair: I am just trying to figure out how to work around puppet lint and it makes one unhappy | 00:10 |
| *** andre__ has quit IRC | 00:11 | |
| *** julim has quit IRC | 00:14 | |
| *** openstackstatus has joined #openstack-infra | 00:15 | |
| *** sc68cal has quit IRC | 00:15 | |
| jeblair | clarkb, fungi: i upgraded the python irc library on eavesdrop.o.o and restarted statusbot | 00:16 |
| jeblair | clarkb, fungi: i believe the new version fixes at least one common bug we see | 00:16 |
| jeblair | looking at doing the same for gerritbot | 00:16 |
| clarkb | jeblair: ok | 00:17 |
| fungi | jeblair: excellent! | 00:17 |
| clarkb | nibalizer: https://review.openstack.org/#/c/77683/7..8/modules/jenkins/manifests/slave.pp will that work? puppet lint wants the trues to be booleans but I am worried that exec may not coerce that properly | 00:17 |
| jeblair | (i'm manually testing these out in private channels to make sure it doesn't break the bots) | 00:17 |
| clarkb | jesusaurus: ^ you may know too | 00:17 |
| *** alexpilotti has quit IRC | 00:17 | |
| pleia2 | I can test it, sec | 00:17 |
| clarkb | pleia2: awesome thanks | 00:18 |
| fungi | clarkb: /bin/true maybe? | 00:18 |
| pleia2 | Error: Parameter command failed on Exec[update-java-alternatives]: Command must be a String, got value of class TrueClass at /root/config/modules/jenkins/manifests/slave.pp:110 | 00:18 |
| pleia2 | Wrapped exception: | 00:18 |
| *** sc68cal has joined #openstack-infra | 00:18 | |
| pleia2 | Command must be a String, got value of class TrueClass | 00:18 |
| pleia2 | :( | 00:18 |
| clarkb | pleia2: fungi: ya I think we need to '/bin/true' | 00:18 |
| clarkb | assuming /bin/true is the common location | 00:18 |
| notmyname | FYI reed has migrated answers to ask.o.o, and I've killed LP answers for swift. other projects will probably be doing the same shortly | 00:18 |
| jeblair | notmyname: great, thanks | 00:19 |
| fungi | notmyname: awesome--thanks for the heads up | 00:19 |
| notmyname | hmm...I should send an email... :-) | 00:19 |
| pleia2 | clarkb: '/bin/true/' works \o/ | 00:19 |
| clarkb | zaro: can you update that change to use '/bin/true' ? | 00:19 |
| clarkb | zaro: also :( this is annoying | 00:20 |
| jesusaurus | clarkb: what are you trying to do with that exec? | 00:20 |
| clarkb | jesusaurus: noop it | 00:20 |
| nibalizer | clarkb: why don't you just /bin/true | 00:21 |
| clarkb | jesusaurus: because there is a before => Exec[thatexec] elsewhere in the code that we need | 00:21 |
| clarkb | nibalizer: thats what we are doing | 00:21 |
| nibalizer | well but if you spell out /bin/true puppet lint will chill out | 00:21 |
| clarkb | nibalizer: right that is what we are doing | 00:21 |
| * pleia2 frowns at pip giving her a hard time on fedora again | 00:21 | |
| clarkb | just wondering if there is a puppet lint magic to make it stop | 00:21 |
| clarkb | or otherwise rely on the path arg properly | 00:21 |
| *** rossella_s has quit IRC | 00:21 | |
| nibalizer | is see unless => true not unless => '/bin/true' | 00:22 |
| clarkb | nibalizer: right, patchset hasn't been updated yet but see scrollback | 00:22 |
| clarkb | from about 3 minutes ago | 00:22 |
| nibalizer | okay | 00:22 |
| nibalizer | also what I do here if I have a dependency like this is i make an anchor called anchor['update-java-alternatives'] that i require elsewhere in the codebase | 00:23 |
| clarkb | nibalizer: isn't that puppet 3 only? | 00:23 |
| nibalizer | the i hang the exec we need off of that anchor, and don't have the exec in the other place | 00:23 |
| nibalizer | no anchor is in stdlib | 00:23 |
| clarkb | zaro: ^ we can do that instead if you want | 00:24 |
| nibalizer | you could also do the incredibly magical | 00:24 |
| *** antonym has joined #openstack-infra | 00:24 | |
| nibalizer | Otherthing['thing] -> Exec <| $name == 'update-java-alternatives' |> | 00:24 |
| nibalizer | but thats just horrible so I wouldn't reccomend it | 00:25 |
| zaro | would the 'anchor' replace the 'before'? | 00:25 |
| clarkb | zaro: no, you would change the before to be before => Anchor['update-java-alternatives'] | 00:26 |
| nibalizer | zaro: before => Anchor['update-java'] | 00:26 |
| clarkb | zaro: then in all of the execs you would require => Anchor['update-java-alternatives'] | 00:26 |
| nibalizer | we should also verify the version of stdlib we have has anchor | 00:26 |
| *** prad has joined #openstack-infra | 00:26 | |
| clarkb | nibalizer: 3.2.0 | 00:27 |
| *** rwsu has quit IRC | 00:27 | |
| nibalizer | ya you're good | 00:27 |
| zaro | ok. i'll give that a try | 00:28 |
| nibalizer | zaro: feel free to ping me if it gets weird on ya | 00:28 |
| nibalizer | also you'll want to namespace your anchor as ['classname::update-java-alternatives'] for sanity | 00:28 |
| fungi | it's already puppet, how much weirder can it get? ;) | 00:28 |
| clarkb | it could be chef | 00:28 |
| fungi | zing! | 00:28 |
| nibalizer | fungi: see the requirement on the virtual resource collector above | 00:28 |
| fungi | nibalizer: yeah, i should have known better than to even ask | 00:29 |
| *** sahumada_ has joined #openstack-infra | 00:29 | |
| *** jcoufal has quit IRC | 00:29 | |
| *** sahumada has quit IRC | 00:29 | |
| *** amotoki has quit IRC | 00:30 | |
| *** rwsu has joined #openstack-infra | 00:30 | |
| *** andreaf has quit IRC | 00:31 | |
| *** ryanpetrello has quit IRC | 00:31 | |
| *** ryanpetrello has joined #openstack-infra | 00:32 | |
| *** mrodden has joined #openstack-infra | 00:34 | |
| openstackgerrit | Khai Do proposed a change to openstack-infra/config: Fix update java alternatives on redhat systems https://review.openstack.org/77683 | 00:35 |
| zaro | nibalizer, pleia2, clarkb : ^ | 00:35 |
| clarkb | fungi: jeblair mordred is there a reason we are running old gearman plugin on our jenkins servers? | 00:35 |
| *** mrodden1 has joined #openstack-infra | 00:35 | |
| nibalizer | zaro: uh classname won't expand for you. i meant for you to fill it out | 00:36 |
| openstackgerrit | Joshua Hesketh proposed a change to openstack-infra/zuul: Add configurable footer-message reports https://review.openstack.org/77740 | 00:36 |
| fungi | clarkb: i think we just didn't get around to upgrading them. i didn't notice myself until i had done the most recent upgrades/downgrades or i would have taken care of it at the same time | 00:36 |
| jhesketh__ | clarkb, jeblair, anteaya, mikal: ^^\ | 00:36 |
| clarkb | fungi: roger | 00:37 |
| *** bhuvan_ has quit IRC | 00:37 | |
| *** bhuvan has quit IRC | 00:37 | |
| zaro | nibalizer: ohh. i'm puppet novice. will fix that up. | 00:37 |
| nibalizer | zaro: :) | 00:37 |
| *** mrodden has quit IRC | 00:38 | |
| nibalizer | jeblair: sorry :( | 00:38 |
| nibalizer | zaro: also you'll need to create the anchor, its a type in puppet so something like: | 00:39 |
| nibalizer | anchor { 'jenkins::slave::update-java-alternatives': } at the top of the file | 00:39 |
| nibalizer | but still inside the class | 00:39 |
| *** sabari has quit IRC | 00:40 | |
| *** sabari_ has quit IRC | 00:40 | |
| clarkb | fungi: I ask about the plugin version due to https://review.openstack.org/#/c/73214/2 puppet seems to want the latest version. I am going to assume we copy pasta'd old verion out of a tarball for server builds | 00:40 |
| clarkb | jeblair: ^ is that something we should try to fix immediately after feature freeze? | 00:41 |
| lyxus | Hello folks, would you know why by any change my logs are doubled http://208.113.169.228/nuage-ci/00_74200_8/ | 00:41 |
| jeblair | clarkb: sounds good | 00:42 |
| clarkb | lyxus: what do you mean by doubled? and whihc log file? | 00:42 |
| jeblair | lyxus: devstack symlinks stack.sh.log.DATE -> stack.sh.log | 00:42 |
| clarkb | oh that doubling | 00:42 |
| jeblair | lyxus: so if you just copy everything, you'll get one copy for the real file, and one for the symlink | 00:42 |
| jeblair | lyxus: i think we do something in devstack-gate-wrap.sh (in devstack-gate) to deal with that | 00:43 |
| zaro | nibalizer: would that mean i would need 3 different anchors? because update-java-alternatives is different for fedora, debian and redhat | 00:43 |
| clarkb | we do, we only copy the non dated files into the jenkins workspace iirc | 00:43 |
| *** dhellmann is now known as dhellmann_ | 00:43 | |
| clarkb | or maybe we delete the symlinks post job | 00:43 |
| *** openstackgerrit has quit IRC | 00:44 | |
| lyxus | clarkb, jeblair , yes you are correct. I didn't even saw that its was sym link | 00:44 |
| clarkb | zaro: just one. And remember we don't need it in fedora anymore because we can delete the exec. | 00:44 |
| jeblair | restarting gerritbot to pick up the new irclib version | 00:44 |
| *** openstackgerrit has joined #openstack-infra | 00:44 | |
| *** ryanpetrello has quit IRC | 00:44 | |
| anteaya | jhesketh__: so on success they get the footer message, on on failure they get the failure-message along with the footer message? | 00:45 |
| jhesketh__ | anteaya: correct, but they get the success-message on success too | 00:46 |
| *** atiwari has quit IRC | 00:46 | |
| jhesketh__ | so it goes: {success|failure}-message, - list of results, footer-message | 00:46 |
| * anteaya looks again for sucess-message | 00:46 | |
| jeblair | clarkb: see the logstash graph, it's dropped. | 00:46 |
| anteaya | ah, already in the code | 00:46 |
| *** nati_ueno has joined #openstack-infra | 00:46 | |
| clarkb | jeblair: yup, I think we are close to the event horizon where we can keep up vs where we can't | 00:47 |
| jeblair | clarkb: queueing systems are fun, eh? :) | 00:47 |
| jhesketh__ | anteaya: yeah, in my test it just uses the default success-message | 00:47 |
| clarkb | and so being on either side of that line for long periods results in the triangle | 00:47 |
| clarkb | jeblair: tons | 00:47 |
| clarkb | jeblair: I was brainstorming about using nodepool to burst workers. I think the major piece of pain there will be firewall rule updates | 00:48 |
| anteaya | jhesketh__: I like it, thank you | 00:48 |
| clarkb | jeblair: maybe we can switch the workers to using internal addresses for ES and make all of that wide open? then nodepool could probably listen to the gearman for logstash and spin up workers for us. Hard part then is telling nodepool when to spin down | 00:48 |
| jeblair | clarkb, fungi: i wonder if we should start using neutron for our servers to get a private network for this sort of thing? | 00:49 |
| clarkb | jeblair: ya, I think we get private network by default that should work | 00:49 |
| clarkb | but we can be fancier if we like | 00:49 |
| clarkb | and put everything on a virtual L2 network | 00:49 |
| jhesketh__ | anteaya: no worries.. the other way I was toying with is that you only have success/failure messages and you put your footer in there and use a variable for the results.. eg: success-message: "Success! {results} details at this URL" | 00:49 |
| *** nati_uen_ has quit IRC | 00:49 | |
| jeblair | clarkb: oh, is the internal network customer-private? i thought it was accessible to anyone in the dc... | 00:50 |
| fungi | does neutron set up gre tunnels between all nodes by default? | 00:50 |
| clarkb | jeblair: still havne't thought through all of what would be required to nodepool logstash workers though, so open to ideas :) | 00:50 |
| jogo | so https://review.openstack.org/#/c/44255/ has been approved but hasn't been merged and its been a while | 00:50 |
| clarkb | jeblair: yes | 00:50 |
| clarkb | at least that was how the trove stuff worked | 00:50 |
| jeblair | clarkb: heat might be the better choice for this... | 00:50 |
| clarkb | hrm good point | 00:50 |
| *** oubiwan__ has quit IRC | 00:51 | |
| fungi | jogo: it depends on another change which has not been approved | 00:51 |
| jogo | fungi: ahaha | 00:51 |
| jogo | woops | 00:51 |
| jeblair | clarkb: oh? you can't access another customer's trove instance? | 00:51 |
| clarkb | jeblair: correct | 00:51 |
| jogo | my bad | 00:51 |
| anteaya | jhesketh__: would it be unfair of me to ask for a quick draft of your other option? | 00:51 |
| *** UtahDave has quit IRC | 00:51 | |
| clarkb | jeblair: if you need that you need to proxy the trove instance | 00:51 |
| *** nati_ueno has quit IRC | 00:51 | |
| jhesketh__ | anteaya: perhaps we could discuss the preferred way before I do? | 00:52 |
| jhesketh__ | I'm happy to build it the other way, but don't really want to waste time on it | 00:52 |
| clarkb | jeblair: in any case we should be able to add worker nodes until the logstash queue derivative goes negative | 00:52 |
| clarkb | jeblair: and below some derivative value remove nodes | 00:52 |
| jhesketh__ | one downside is that it'll be a config change to layout.yaml as existing zuul's will need to include their results in their messages | 00:52 |
| clarkb | I should probably think about this more post feature freeze | 00:52 |
| jeblair | clarkb: regardless, we might want to tighten it down even more than "our account" to just things that need to be on certain networks | 00:53 |
| *** ryanpetrello has joined #openstack-infra | 00:53 | |
| clarkb | jeblair: ya, and we should be able to do that with neutron | 00:53 |
| *** Ryan_Lane has quit IRC | 00:53 | |
| jeblair | jhesketh__, anteaya: from what i've heard (without reviewing), the footer approach sounds good | 00:53 |
| clarkb | carve a /24 for workers and a /24 for es nodes and allow them to tlak to each other | 00:53 |
| anteaya | jhesketh__: yeah let's not try to reconfigure existing zuuls | 00:54 |
| anteaya | that makes many users unhappy | 00:54 |
| anteaya | then they show up in here, unhappy | 00:54 |
| *** oubiwan__ has joined #openstack-infra | 00:54 | |
| *** dims has joined #openstack-infra | 00:54 | |
| jhesketh__ | well you could use black magic to see if the results were inserted into the message and if not, append them | 00:54 |
| jeblair | and clarkb gets them even more depressed with 'everything ever is bad' | 00:54 |
| jhesketh__ | but that's a little magical for me | 00:54 |
| *** ryanpetrello has quit IRC | 00:54 | |
| anteaya | jhesketh__: yeah, I dont' like the mysterious in the code | 00:55 |
| anteaya | let's stay with what you have | 00:55 |
| jhesketh__ | sure, sounds good to me | 00:55 |
| clarkb | jeblair: sorry :( | 00:55 |
| jeblair | clarkb: no it's awesome, you are a coremudgeon. | 00:56 |
| jeblair | i think it should be our new motto | 00:56 |
| *** Ryan_Lane has joined #openstack-infra | 00:56 | |
| fungi | bad things bad. good things bad too. all things bad | 00:56 |
| *** nati_ueno has joined #openstack-infra | 00:57 | |
| pleia2 | zaro, clarkb: heading out of the office, will be around later if you need any testing of the patch-that-wouldn't-die :) | 00:57 |
| *** thuc_ has joined #openstack-infra | 00:58 | |
| *** thuc_ has quit IRC | 00:59 | |
| *** thuc_ has joined #openstack-infra | 00:59 | |
| *** thuc_ has quit IRC | 00:59 | |
| *** oubiwan__ has quit IRC | 00:59 | |
| *** thuc_ has joined #openstack-infra | 01:00 | |
| *** thuc_ has quit IRC | 01:00 | |
| *** thuc has quit IRC | 01:00 | |
| *** thuc has joined #openstack-infra | 01:01 | |
| *** thuc has quit IRC | 01:01 | |
| jeblair | i'm hearing 'all along the watchtower' for the second time today. i think you all probably know what that means. | 01:01 |
| *** cody-somerville has quit IRC | 01:02 | |
| *** thuc has joined #openstack-infra | 01:02 | |
| fungi | your hendrix collection is on repeat? | 01:02 |
| fungi | (i know mine is) | 01:02 |
| zaro | clarkb: i don't understand how you can set java alternatives 2 different ways with only 1 anchor. | 01:04 |
| fungi | (second guess, s/hendrix/dylan/) | 01:04 |
| zaro | clarkb: we still need to set for debian and centos. so how can you define one anchor to do both? | 01:05 |
| clarkb | zaro: the anchor is just enforcing order | 01:05 |
| clarkb | zaro: both execs require the same anchor | 01:05 |
| clarkb | zaro: then the package install does before => Anchor... | 01:06 |
| clarkb | zaro: that enforces the order of the exec | 01:06 |
| *** oubiwan__ has joined #openstack-infra | 01:07 | |
| *** oubiwan__ has joined #openstack-infra | 01:08 | |
| david-lyle | if one more person with openstack/requirements core has a minute to take a look at https://review.openstack.org/#/c/77015/ we can finalize support for Django 1.6 in Horizon | 01:09 |
| *** nati_ueno has quit IRC | 01:09 | |
| *** nati_ueno has joined #openstack-infra | 01:10 | |
| *** melwitt1 has quit IRC | 01:10 | |
| *** nati_ueno has quit IRC | 01:10 | |
| *** nati_ueno has joined #openstack-infra | 01:11 | |
| anteaya | david-lyle: https://review.openstack.org/#/admin/groups/131,members | 01:11 |
| anteaya | doug or sean might be your best bets unless you want to wait until tomorrow | 01:12 |
| *** nati_ueno has quit IRC | 01:12 | |
| *** wenlock has quit IRC | 01:12 | |
| *** CaptTofu has quit IRC | 01:12 | |
| *** nati_ueno has joined #openstack-infra | 01:12 | |
| openstackgerrit | Khai Do proposed a change to openstack-infra/config: Fix update java alternatives on redhat systems https://review.openstack.org/77683 | 01:13 |
| zaro | clarb : as i understand it ^ | 01:13 |
| zaro | clarkb: ^ | 01:13 |
| clarkb | zaro: really close | 01:14 |
| *** jcoufal has joined #openstack-infra | 01:14 | |
| *** nati_ueno has quit IRC | 01:14 | |
| *** Ryan_Lane has quit IRC | 01:15 | |
| *** nati_ueno has joined #openstack-infra | 01:15 | |
| openstackgerrit | Jeremy Stanley proposed a change to openstack-infra/config: Move git03.openstack.org to a new server https://review.openstack.org/77743 | 01:15 |
| *** jcoufal has quit IRC | 01:15 | |
| *** nati_ueno has quit IRC | 01:16 | |
| *** nati_ueno has joined #openstack-infra | 01:16 | |
| *** nati_ueno has quit IRC | 01:17 | |
| *** rwsu has quit IRC | 01:17 | |
| fungi | git03 has been offlined in the haproxy in preparation for ^ | 01:17 |
| *** nati_ueno has joined #openstack-infra | 01:17 | |
| fungi | i'll give it a few minutes before i press forward with the corresponding dns changes | 01:18 |
| *** nati_ueno has quit IRC | 01:18 | |
| *** nati_ueno has joined #openstack-infra | 01:19 | |
| *** markmc has quit IRC | 01:19 | |
| fungi | still waiting out the old ttl on those rrs anyway | 01:19 |
| *** oubiwan__ has quit IRC | 01:21 | |
| clarkb | ok | 01:21 |
| *** oubiwan__ has joined #openstack-infra | 01:21 | |
| *** nati_ueno has quit IRC | 01:22 | |
| *** nati_ueno has joined #openstack-infra | 01:22 | |
| *** nati_ueno has quit IRC | 01:23 | |
| *** nati_ueno has joined #openstack-infra | 01:23 | |
| *** nati_ueno has quit IRC | 01:24 | |
| david-lyle | anteaya: thanks! that list seems a little stale at first glance ;) | 01:24 |
| *** nati_ueno has joined #openstack-infra | 01:25 | |
| *** dprince has quit IRC | 01:25 | |
| *** nati_ueno has quit IRC | 01:26 | |
| *** nati_ueno has joined #openstack-infra | 01:26 | |
| *** ArxCruz_ has quit IRC | 01:26 | |
| *** hogepodge has quit IRC | 01:26 | |
| *** nati_uen_ has joined #openstack-infra | 01:27 | |
| *** nati_ueno has quit IRC | 01:27 | |
| anteaya | david-lyle: yes it is | 01:28 |
| anteaya | I am not sure who can clean it up | 01:28 |
| anteaya | I tried to offer my suggestions for its culling a week or so back | 01:28 |
| *** nati_uen_ has quit IRC | 01:28 | |
| anteaya | if you do better than myself, that would be great | 01:28 |
| *** nosnos has joined #openstack-infra | 01:29 | |
| *** nati_ueno has joined #openstack-infra | 01:29 | |
| *** nati_ueno has quit IRC | 01:30 | |
| *** derekh has quit IRC | 01:30 | |
| *** nati_ueno has joined #openstack-infra | 01:30 | |
| *** nati_ueno has quit IRC | 01:31 | |
| *** nati_ueno has joined #openstack-infra | 01:31 | |
| *** nati_ueno has quit IRC | 01:32 | |
| *** oubiwan__ has quit IRC | 01:34 | |
| *** CaptTofu has joined #openstack-infra | 01:34 | |
| clarkb | zaro: gah apparently I didn't publish my comments | 01:36 |
| clarkb | zaro: totally thought I did, they are posted now | 01:36 |
| *** TheWinnerFjk has joined #openstack-infra | 01:41 | |
| *** TheWinnerFjk has left #openstack-infra | 01:41 | |
| *** zehicle_at_dell has quit IRC | 01:42 | |
| anteaya | I can't believe there is only 4 in the gate | 01:44 |
| *** oubiwan__ has joined #openstack-infra | 01:48 | |
| *** jcooley_ has quit IRC | 01:49 | |
| *** jcooley_ has joined #openstack-infra | 01:49 | |
| clarkb | anteaya: we have merged >100 things today | 01:50 |
| fungi | clarkb: was https://launchpad.net/bugs/1287348 a duplicate? | 01:52 |
| clarkb | fungi: it came in right after I finished. anteaya was tracking down if it was necessary | 01:53 |
| *** jcooley_ has quit IRC | 01:53 | |
| clarkb | shivharris was looking into it | 01:53 |
| fungi | clarkb: oh, right, that one | 01:54 |
| *** amcrn has quit IRC | 01:55 | |
| *** jcooley_ has joined #openstack-infra | 01:56 | |
| openstackgerrit | A change was merged to openstack-infra/devstack-gate: fix incorrect use of local https://review.openstack.org/77726 | 01:59 |
| *** jcooley_ has quit IRC | 02:02 | |
| *** mgagne has joined #openstack-infra | 02:05 | |
| *** jcooley_ has joined #openstack-infra | 02:07 | |
| *** jcooley_ has quit IRC | 02:08 | |
| *** markmc has joined #openstack-infra | 02:08 | |
| *** mgagne has quit IRC | 02:08 | |
| *** mgagne has joined #openstack-infra | 02:09 | |
| *** thuc has quit IRC | 02:12 | |
| *** thuc has joined #openstack-infra | 02:12 | |
| *** jcooley_ has joined #openstack-infra | 02:13 | |
| *** jcooley_ has quit IRC | 02:13 | |
| *** banix has joined #openstack-infra | 02:14 | |
| *** ryanpetrello has joined #openstack-infra | 02:14 | |
| *** mgagne has quit IRC | 02:15 | |
| *** thuc has quit IRC | 02:17 | |
| *** jcooley_ has joined #openstack-infra | 02:26 | |
| *** jcooley_ has quit IRC | 02:26 | |
| *** gokrokve has quit IRC | 02:30 | |
| *** khyati has quit IRC | 02:31 | |
| *** yamahata has quit IRC | 02:36 | |
| *** sparkycollier has joined #openstack-infra | 02:42 | |
| *** dcramer__ has joined #openstack-infra | 02:44 | |
| *** rfolco has quit IRC | 02:50 | |
| *** wenlock has joined #openstack-infra | 02:52 | |
| openstackgerrit | Michael Krotscheck proposed a change to openstack-infra/storyboard: [WIP] Added DELETE method for projects, stories, and tasks. https://review.openstack.org/77763 | 02:52 |
| openstackgerrit | Michael Krotscheck proposed a change to openstack-infra/storyboard-webclient: Enable story and project 'deletion' https://review.openstack.org/77764 | 02:54 |
| openstackgerrit | Michael Krotscheck proposed a change to openstack-infra/storyboard-webclient: Task, story, and project fixes https://review.openstack.org/77013 | 02:54 |
| *** ryanpetrello has quit IRC | 02:55 | |
| *** krotscheck has quit IRC | 02:57 | |
| *** amotoki has joined #openstack-infra | 02:58 | |
| *** nati_ueno has joined #openstack-infra | 02:59 | |
| *** nati_ueno has quit IRC | 02:59 | |
| *** michchap has quit IRC | 03:01 | |
| *** michchap has joined #openstack-infra | 03:01 | |
| *** sparkycollier has quit IRC | 03:03 | |
| *** sparkycollier has joined #openstack-infra | 03:04 | |
| *** amcrn has joined #openstack-infra | 03:08 | |
| openstackgerrit | A change was merged to openstack-infra/config: Move git03.openstack.org to a new server https://review.openstack.org/77743 | 03:10 |
| *** sparkycollier has quit IRC | 03:10 | |
| *** sparkycollier has joined #openstack-infra | 03:12 | |
| *** yamahata has joined #openstack-infra | 03:22 | |
| *** thuc has joined #openstack-infra | 03:23 | |
| *** dstanek_afk has joined #openstack-infra | 03:24 | |
| *** dstanek has quit IRC | 03:24 | |
| fungi | the replacement git03 is in the pool now and handling requests successfully as far as i can tell | 03:24 |
| clarkb | fungi: I notice that git04 is doing less work too | 03:24 |
| fungi | i'll keep an ear to the ground for an hour or so before i raise the ttl on its dns rrs and delete the old server | 03:24 |
| fungi | clarkb: yep, on the whole the newer servers do seem slightly happier | 03:25 |
| *** gokrokve has joined #openstack-infra | 03:27 | |
| *** dstanek_afk is now known as dstanek | 03:27 | |
| *** thuc has quit IRC | 03:28 | |
| *** khyati has joined #openstack-infra | 03:29 | |
| zaro | clarkb: remove '$' only or '$::' | 03:31 |
| zaro | ? | 03:31 |
| clarkb | zaro: it should match the first anchor | 03:31 |
| *** sparkycollier has quit IRC | 03:31 | |
| clarkb | zaro: since they need to match each other | 03:31 |
| *** khyati has quit IRC | 03:32 | |
| *** SumitNaiksatam has joined #openstack-infra | 03:32 | |
| *** CaptTofu has quit IRC | 03:32 | |
| *** vkozhukalov has joined #openstack-infra | 03:32 | |
| dstufft | y'all leaving for OFTC? | 03:34 |
| fungi | some of us can't stand the heat (constant ddos'ing) and would prefer to get out of the kitchen (freenode) | 03:35 |
| dstufft | you're gonna lose people fwiw (not me, because I collect IRCs like they are pokemon) | 03:36 |
| mordred | dstufft: mmm. pokemon | 03:36 |
| fungi | many (most?) are already in channels on both networks anyway | 03:36 |
| clarkb | I have a feeling that if freenode were to fall over tomorrow the migration would complete itself organically in about a day | 03:37 |
| *** chandan_kumar has joined #openstack-infra | 03:37 | |
| fungi | clarkb: you mean fall over again? | 03:37 |
| clarkb | ya | 03:38 |
| clarkb | part of that is with feature freeze people absolutely need to communicate | 03:38 |
| dstufft | *Shrug* just saying :) I know more than one person who considers anything not on freenode as dead to them :) That's not really a comment on if you should or shouldn't make the move, just stating that you're gonna lose people in the transition :) | 03:38 |
| fungi | the last big one where i couldn't find a connected freenode server for most of a day happened on a weekend, so it was probably less noticeable to a lot of people | 03:39 |
| *** dcramer__ has quit IRC | 03:39 | |
| clarkb | dstufft: I feel sorry for those people | 03:39 |
| clarkb | what do they do when freenode is gone | 03:39 |
| clarkb | take the day off? | 03:39 |
| fungi | pray to the ghost of rob levin to stop siphoning away all teh ircs | 03:40 |
| dstufft | I must have an awesome freenode server b/c I don't hardly ever notice problems :) | 03:41 |
| clarkb | dstufft: also aiui its still a proposal. If folks feel that strongly they should make their objections known | 03:41 |
| *** annegentle has quit IRC | 03:42 | |
| * fungi wonders which server you were on saturday before last. i must have tried every one in the dns round-robin and the only one which eventually had anyone in the channels was dixon, which spent the better part of the day telling me it was full and to try again later | 03:42 | |
| fungi | i will also readily acknowledge that there's nothing stopping oftc from being beset by skr1pt kiddies, but having spent many years in channels on both networks it seems like it's been far less of a target overall | 03:44 |
| clarkb | maybe we should use hipchat | 03:44 |
| * clarkb needs more beer apparently | 03:44 | |
| fungi | i hear it's hip with all the hippest hipster hippies | 03:44 |
| fungi | says so right there in the name | 03:45 |
| openstackgerrit | Khai Do proposed a change to openstack-infra/config: Fix update java alternatives on redhat systems https://review.openstack.org/77683 | 03:46 |
| dstufft | what you need | 03:46 |
| dstufft | is a new openstack project | 03:46 |
| dstufft | IRC as a Service | 03:46 |
| dstufft | then just make irc.openstack.org and split it across multiple clouds | 03:46 |
| clarkb | dstufft: please no | 03:47 |
| clarkb | dstufft: however, you could probably use the messaging as a service project as the backend | 03:47 |
| fungi | there are enough of us here who have (or continue to) run irc servers to recognize what a time sink that turns into | 03:47 |
| clarkb | zaro: almost | 03:48 |
| clarkb | zaro: just a one character change then I think its ready | 03:48 |
| fungi | clarkb: is this the change which asymptotically approaches merging? | 03:48 |
| clarkb | ya | 03:48 |
| fungi | eventually we're going to see it get to sub-character changes (i suppose with composing codepoints in unicode that's possible too) | 03:49 |
| clarkb | the limit of change 77683 as patchsets approach infinity is merged | 03:49 |
| dstufft | fungi: I've done that :D I mostly like suggesting it to see who I can convince to waste their time >:] | 03:50 |
| * fungi has plenty of ways to waste time already | 03:50 | |
| fungi | speaking of wasting time, i've had a scripted loop running all day trying to get good image updates in hpcloud-az2... clarkb: have you heard anything at all about the situation we've got there? | 03:52 |
| clarkb | fungi: no new responses to the ticket | 03:52 |
| clarkb | I can followup tomorrow | 03:52 |
| fungi | guess i'll just leave this running overnight and dream of working clouds | 03:52 |
| fungi | we're about 50% of the way through the rsync of docs-draft onto its new filesystem too, so hopefully i can wrap that up and expand it and the logs volume tomorrow | 03:53 |
| fungi | i'll roughly double docs-draft and throw a couple more terabytes at logs | 03:54 |
| clarkb | fungi: speaking of, for the ES servers you used the documented static.o.o process? | 03:54 |
| fungi | clarkb: to the letter, yes | 03:54 |
| clarkb | I think you said you did | 03:54 |
| clarkb | great, I think we may want to grow those too | 03:55 |
| clarkb | but later | 03:55 |
| clarkb | now that I figured out indexes weren't deleting we have a lot more headroom | 03:55 |
| fungi | sure, we'll need another quota bump to do so though, or wait until we can get our static logs into swift | 03:55 |
| clarkb | ya I don't think its a big rush | 03:55 |
| clarkb | but the other day I thought we were running out of disk fast | 03:56 |
| clarkb | we were, but only beacuse we weren't deleting old data | 03:56 |
| openstackgerrit | A change was merged to openstack-dev/pbr: Remove copyright from empty files https://review.openstack.org/75876 | 03:57 |
| fungi | i'm switching into lower-bandwidth mode to rest up for the ever increasing onslaught tomorrow | 03:57 |
| clarkb | I'm going to jinx it but this ff was much better than havana's so far | 03:57 |
| fungi | i'm just assuming it hasn't *really* ramped up yet like it's going to | 03:58 |
| *** khyati has joined #openstack-infra | 04:02 | |
| *** zehicle_at_dell has joined #openstack-infra | 04:08 | |
| *** hdd_ has joined #openstack-infra | 04:09 | |
| *** sc68cal has quit IRC | 04:17 | |
| *** sc68cal has joined #openstack-infra | 04:20 | |
| *** wchrisj has quit IRC | 04:27 | |
| morganfainberg | clarkb, so if i were to propose a periodic job that works like the transifex stuff (but for sample config generation) would it make more sense to use a tox command (get all the deps) or do it all ... more shell like? I'm inclined to say the former | 04:28 |
| morganfainberg | fungi, ^ | 04:28 |
| morganfainberg | tox -egenerate_sample_config perhaps | 04:29 |
| *** pcrews has quit IRC | 04:29 | |
| *** harlowja is now known as harlowja_away | 04:30 | |
| *** thuc has joined #openstack-infra | 04:30 | |
| *** thuc has quit IRC | 04:33 | |
| *** dcramer__ has joined #openstack-infra | 04:33 | |
| *** thuc has joined #openstack-infra | 04:33 | |
| *** harlowja_away is now known as harlowja | 04:35 | |
| *** wchrisj has joined #openstack-infra | 04:36 | |
| tchaypo | :qa | 04:36 |
| clarkb | if you need a virtualenv use tox. or even tox -evenv -- stuff | 04:36 |
| *** jcooley_ has joined #openstack-infra | 04:37 | |
| *** thuc has quit IRC | 04:37 | |
| openstackgerrit | Russell Bryant proposed a change to openstack-infra/config: Trim down gantt check/gate jobs https://review.openstack.org/71648 | 04:43 |
| *** nati_ueno has joined #openstack-infra | 04:45 | |
| morganfainberg | clarkb, yep. actually going to call the run-tox script. | 04:51 |
| *** banix has quit IRC | 04:52 | |
| *** chandan_kumar has quit IRC | 04:53 | |
| morganfainberg | clarkb, so i would like this to run periodically if it's added to the periodic pipeline how does that work? or should this be just a job that gets fired up post merge? - we want to remove the sample config generation from pep8 in keystone | 04:54 |
| morganfainberg | trying to think of the "best" way to do this w/o overloading things if we merge a bunch of things at once. | 04:54 |
| clarkb | periodic jobs go in a periodic pipeline | 04:54 |
| morganfainberg | clarkb, like the propose translation ones? | 04:55 |
| morganfainberg | clarkb or are those triggered? | 04:55 |
| openstackgerrit | A change was merged to openstack-dev/pbr: Declare support for Python versions in setup.cfg https://review.openstack.org/75868 | 04:56 |
| openstackgerrit | Khai Do proposed a change to openstack-infra/config: Fix update java alternatives on redhat systems https://review.openstack.org/77683 | 04:56 |
| *** thuc_ has joined #openstack-infra | 05:01 | |
| *** jcooley_ has quit IRC | 05:02 | |
| *** talluri has joined #openstack-infra | 05:04 | |
| *** mrda is now known as mrda_away | 05:09 | |
| openstackgerrit | Morgan Fainberg proposed a change to openstack-infra/config: Add in scaffolding for sample_config update periodic task https://review.openstack.org/77789 | 05:10 |
| *** vkozhukalov has quit IRC | 05:10 | |
| openstackgerrit | Morgan Fainberg proposed a change to openstack-infra/config: Enable periodic sample config job for Keystone https://review.openstack.org/77790 | 05:10 |
| *** Ryan_Lane has joined #openstack-infra | 05:13 | |
| openstackgerrit | Morgan Fainberg proposed a change to openstack-infra/config: Add in scaffolding for sample_config update periodic task https://review.openstack.org/77789 | 05:14 |
| openstackgerrit | Morgan Fainberg proposed a change to openstack-infra/config: Enable periodic sample config job for Keystone https://review.openstack.org/77790 | 05:16 |
| *** wchrisj has quit IRC | 05:17 | |
| openstackgerrit | Morgan Fainberg proposed a change to openstack-infra/config: Add in scaffolding for sample_config update periodic task https://review.openstack.org/77789 | 05:19 |
| *** vkozhukalov has joined #openstack-infra | 05:21 | |
| openstackgerrit | Morgan Fainberg proposed a change to openstack-infra/config: Add in scaffolding for sample_config update periodic task https://review.openstack.org/77789 | 05:21 |
| * morganfainberg grumbles about trying to type yaml | 05:21 | |
| openstackgerrit | Morgan Fainberg proposed a change to openstack-infra/config: Enable periodic sample config job for Keystone https://review.openstack.org/77790 | 05:24 |
| *** banix has joined #openstack-infra | 05:29 | |
| *** CaptTofu has joined #openstack-infra | 05:33 | |
| *** vkozhukalov has quit IRC | 05:35 | |
| openstackgerrit | Sergey Lukjanov proposed a change to openstack-infra/config: Savanna: add job to build swift fs for hadoop https://review.openstack.org/76923 | 05:36 |
| *** jcooley_ has joined #openstack-infra | 05:36 | |
| *** CaptTofu has quit IRC | 05:38 | |
| *** gokrokve has quit IRC | 05:38 | |
| *** gokrokve has joined #openstack-infra | 05:39 | |
| *** nicedice has quit IRC | 05:41 | |
| *** banix has quit IRC | 05:41 | |
| *** gokrokve has quit IRC | 05:43 | |
| *** zhiyan_ is now known as zhiyan | 05:44 | |
| *** VinayMahuli has joined #openstack-infra | 05:50 | |
| *** banix has joined #openstack-infra | 06:06 | |
| *** oubiwan__ has quit IRC | 06:06 | |
| *** oubiwan__ has joined #openstack-infra | 06:07 | |
| *** amcrn has quit IRC | 06:13 | |
| *** harlowja is now known as harlowja_away | 06:16 | |
| *** yongli has quit IRC | 06:21 | |
| *** gokrokve has joined #openstack-infra | 06:27 | |
| *** gokrokve_ has joined #openstack-infra | 06:29 | |
| *** ociuhandu has joined #openstack-infra | 06:30 | |
| *** gokrokve has quit IRC | 06:31 | |
| *** gyee has quit IRC | 06:34 | |
| *** hdd_ has quit IRC | 06:39 | |
| *** denis_makogon has joined #openstack-infra | 06:46 | |
| *** khyati has quit IRC | 06:48 | |
| *** thuc_ has quit IRC | 06:49 | |
| *** thuc has joined #openstack-infra | 06:50 | |
| *** thuc has quit IRC | 06:54 | |
| *** odyssey4me has quit IRC | 06:57 | |
| *** CaptTofu has joined #openstack-infra | 07:00 | |
| *** bogdando has joined #openstack-infra | 07:01 | |
| *** CaptTofu has quit IRC | 07:06 | |
| *** weshay has quit IRC | 07:08 | |
| *** saju_m has joined #openstack-infra | 07:12 | |
| *** mayu has joined #openstack-infra | 07:12 | |
| *** odyssey4me has joined #openstack-infra | 07:15 | |
| mayu | Following jaypipes's blog to construct third-party ci, who encounter this error http://paste.openstack.org/show/71560/ | 07:17 |
| *** yolanda_ has joined #openstack-infra | 07:28 | |
| *** nosnos has quit IRC | 07:31 | |
| *** nosnos_ has joined #openstack-infra | 07:31 | |
| *** banix has quit IRC | 07:35 | |
| *** mkoderer has joined #openstack-infra | 07:37 | |
| *** andreaf has joined #openstack-infra | 07:43 | |
| *** amotoki has quit IRC | 07:47 | |
| *** thuc has joined #openstack-infra | 07:50 | |
| *** oubiwan__ has quit IRC | 07:56 | |
| *** flaper87|afk is now known as flaper87 | 07:58 | |
| *** thuc has quit IRC | 07:58 | |
| *** wenlock has quit IRC | 08:00 | |
| *** morganfainberg is now known as morganfainberg_Z | 08:02 | |
| *** nosnos_ has quit IRC | 08:04 | |
| *** nosnos has joined #openstack-infra | 08:04 | |
| *** ildikov_ has joined #openstack-infra | 08:14 | |
| *** pblaho has joined #openstack-infra | 08:18 | |
| *** mrmartin has joined #openstack-infra | 08:18 | |
| *** jp_at_hp has joined #openstack-infra | 08:22 | |
| *** saju_m has quit IRC | 08:26 | |
| *** rlandy has joined #openstack-infra | 08:29 | |
| *** flaper87|afk has joined #openstack-infra | 08:35 | |
| *** hashar has joined #openstack-infra | 08:36 | |
| *** alexpilotti has joined #openstack-infra | 08:37 | |
| *** jgallard has joined #openstack-infra | 08:41 | |
| *** morganfainberg_Z is now known as morganfainberg | 08:44 | |
| *** morganfainberg is now known as morganfainberg_Z | 08:46 | |
| *** denis_makogon has quit IRC | 08:47 | |
| *** dizquierdo has joined #openstack-infra | 08:50 | |
| *** flaper87|afk has quit IRC | 08:50 | |
| *** rossella_s has joined #openstack-infra | 08:51 | |
| openstackgerrit | Nikita Konovalov proposed a change to openstack-infra/storyboard: Auth Token Middleware https://review.openstack.org/74735 | 08:52 |
| *** Ryan_Lane has quit IRC | 08:54 | |
| openstackgerrit | Sergey Kolekonov proposed a change to openstack-infra/jenkins-job-builder: Ruby metrics plugin support added https://review.openstack.org/73528 | 09:01 |
| *** CaptTofu has joined #openstack-infra | 09:03 | |
| *** arborism has joined #openstack-infra | 09:04 | |
| *** CaptTofu has quit IRC | 09:07 | |
| *** arborism is now known as amcrn | 09:07 | |
| *** jpich has joined #openstack-infra | 09:08 | |
| *** yassine has joined #openstack-infra | 09:09 | |
| *** e0ne has joined #openstack-infra | 09:10 | |
| *** ihrachys|sick is now known as ihrachys|wfh | 09:11 | |
| *** e0ne has quit IRC | 09:15 | |
| *** e0ne_ has joined #openstack-infra | 09:15 | |
| *** e0ne_ has quit IRC | 09:16 | |
| *** jlibosva has joined #openstack-infra | 09:16 | |
| *** e0ne has joined #openstack-infra | 09:16 | |
| *** dizquierdo has quit IRC | 09:19 | |
| *** e0ne_ has joined #openstack-infra | 09:20 | |
| *** e0ne has quit IRC | 09:20 | |
| *** Ryan_Lane has joined #openstack-infra | 09:25 | |
| *** johnthetubaguy has joined #openstack-infra | 09:26 | |
| *** Ryan_Lane1 has joined #openstack-infra | 09:27 | |
| *** Ryan_Lane has quit IRC | 09:27 | |
| *** johnthetubaguy1 has joined #openstack-infra | 09:28 | |
| *** e0ne has joined #openstack-infra | 09:28 | |
| *** johnthetubaguy has quit IRC | 09:28 | |
| *** e0ne_ has quit IRC | 09:28 | |
| *** e0ne has quit IRC | 09:29 | |
| *** e0ne has joined #openstack-infra | 09:29 | |
| *** Ryan_Lane1 has quit IRC | 09:32 | |
| *** Ryan_Lane has joined #openstack-infra | 09:32 | |
| *** e0ne has quit IRC | 09:33 | |
| *** e0ne has joined #openstack-infra | 09:34 | |
| *** Ryan_Lane has quit IRC | 09:36 | |
| *** e0ne_ has joined #openstack-infra | 09:38 | |
| *** mflobo has joined #openstack-infra | 09:41 | |
| *** saju_m has joined #openstack-infra | 09:41 | |
| *** e0ne has quit IRC | 09:42 | |
| *** mflobo has quit IRC | 09:42 | |
| *** mflobo has joined #openstack-infra | 09:42 | |
| *** jhesketh__ has quit IRC | 09:42 | |
| *** viktors_away is now known as viktors | 09:43 | |
| *** krtaylor has quit IRC | 09:44 | |
| *** fbo_away is now known as fbo | 09:45 | |
| *** gilliard has joined #openstack-infra | 09:51 | |
| gilliard | Hi. Nova PEP8 job seems to be failing: https://bugs.launchpad.net/openstack-ci/+bug/1287629 | 09:51 |
| gilliard | http://logstash.openstack.org/#eyJzZWFyY2giOiJcIlJ1bnRpbWVFcnJvcjogVW5hYmxlIHRvIGltcG9ydCBtb2R1bGUgbm92YS5vcGVuc3RhY2suY29tbW9uLnNzbHV0aWxzXCIiLCJmaWVsZHMiOltdLCJvZmZzZXQiOjAsInRpbWVmcmFtZSI6Ijg2NDAwIiwiZ3JhcGhtb2RlIjoiY291bnQiLCJ0aW1lIjp7InVzZXJfaW50ZXJ2YWwiOjB9LCJzdGFtcCI6MTM5MzkyNjU2MTA4Mn0= | 09:52 |
| *** salv-orlando_ has joined #openstack-infra | 09:52 | |
| gilliard | I've been having this problem locally for about a week, so not sure if it's something in the code that's changed, or something in the environment. | 09:53 |
| *** salv-orlando has quit IRC | 09:55 | |
| *** salv-orlando_ is now known as salv-orlando | 09:55 | |
| *** mflobo has quit IRC | 09:56 | |
| *** afazekas has joined #openstack-infra | 10:05 | |
| *** nati_ueno has quit IRC | 10:12 | |
| openstackgerrit | Antoine Musso proposed a change to openstack-infra/jenkins-job-builder: Document the Zuul triggers https://review.openstack.org/77838 | 10:13 |
| *** yamahata has quit IRC | 10:18 | |
| *** jhesketh_ has quit IRC | 10:18 | |
| *** proffalken has joined #openstack-infra | 10:24 | |
| *** andreaf has quit IRC | 10:24 | |
| *** jd__ has joined #openstack-infra | 10:32 | |
| openstackgerrit | Antoine Musso proposed a change to openstack-infra/zuul: Document the Zuul triggers https://review.openstack.org/77843 | 10:32 |
| *** e0ne_ has quit IRC | 10:33 | |
| *** proffalken has quit IRC | 10:33 | |
| openstackgerrit | Michael Still proposed a change to openstack/requirements: Don't use oslo.messaging 1.3.0a8 https://review.openstack.org/77844 | 10:34 |
| *** e0ne has joined #openstack-infra | 10:34 | |
| *** e0ne_ has joined #openstack-infra | 10:35 | |
| *** ociuhandu has quit IRC | 10:35 | |
| *** ndipanov has joined #openstack-infra | 10:36 | |
| *** mrmartin has quit IRC | 10:38 | |
| *** e0ne has quit IRC | 10:38 | |
| *** morganfainberg_Z is now known as morganfainberg | 10:39 | |
| *** proffalken has joined #openstack-infra | 10:49 | |
| openstackgerrit | Antoine Musso proposed a change to openstack-infra/jenkins-job-builder: Content-Type can now be set for email-ext publisher https://review.openstack.org/75919 | 10:50 |
| *** thuc has joined #openstack-infra | 10:50 | |
| openstackgerrit | Antoine Musso proposed a change to openstack-infra/jenkins-job-builder: Content-Type can now be set for email-ext publisher https://review.openstack.org/75919 | 10:50 |
| sdague | gilliard: please read the bottom of the log you just linked to | 10:50 |
| sdague | "check_uptodate.sh: Please run /home/jenkins/workspace/gate-nova-pep8/tools/config/generate_sample.sh." | 10:50 |
| *** thuc has quit IRC | 10:55 | |
| *** salv-orlando has quit IRC | 10:56 | |
| *** salv-orlando has joined #openstack-infra | 10:56 | |
| gilliard | sdague: The problem was that jenkins couldn't run that script - a new version of oslo.messaging had broken it. | 10:57 |
| gilliard | mikal has a patch for it. | 10:57 |
| gilliard | Oh - I see you picked it up in #openstack-nova | 10:58 |
| *** CaptTofu has joined #openstack-infra | 11:03 | |
| *** amcrn has quit IRC | 11:04 | |
| *** jgallard has quit IRC | 11:05 | |
| ttx | sdague: maybe we should push in https://review.openstack.org/77844 | 11:08 |
| *** CaptTofu has quit IRC | 11:08 | |
| sdague | I am in no way convinced that's the real issue | 11:08 |
| ttx | hah :) | 11:09 |
| ttx | I suspected if you did not push it already there was a reason | 11:09 |
| *** ociuhandu has joined #openstack-infra | 11:10 | |
| sdague | well, also, only been awake for 30 minutes | 11:10 |
| *** saschpe has quit IRC | 11:12 | |
| *** yamahata has joined #openstack-infra | 11:12 | |
| *** amcrn has joined #openstack-infra | 11:13 | |
| *** saschpe has joined #openstack-infra | 11:16 | |
| *** morganfainberg is now known as morganfainberg_Z | 11:16 | |
| *** yamahata has quit IRC | 11:16 | |
| openstackgerrit | Nikita Konovalov proposed a change to openstack-infra/storyboard: Auth Token Middleware https://review.openstack.org/74735 | 11:17 |
| SergeyLukjanov | ttx, re https://review.openstack.org/77844 I think that it's better to !1.3.0a8 to make next fixed release just work | 11:21 |
| *** CaptTofu has joined #openstack-infra | 11:22 | |
| *** hashar_ has joined #openstack-infra | 11:27 | |
| *** hashar has quit IRC | 11:27 | |
| *** andre__ has joined #openstack-infra | 11:28 | |
| sdague | SergeyLukjanov: agreed | 11:31 |
| sdague | I'm making that change now | 11:31 |
| sdague | the < change will prevent us from using the git master in the gate | 11:31 |
| sdague | which is why that nova change doesn't work | 11:31 |
| openstackgerrit | Sean Dague proposed a change to openstack/requirements: Don't use oslo.messaging 1.3.0a8 https://review.openstack.org/77844 | 11:32 |
| *** johnthetubaguy1 has quit IRC | 11:32 | |
| *** Ryan_Lane has joined #openstack-infra | 11:34 | |
| openstackgerrit | Nikita Konovalov proposed a change to openstack-infra/storyboard: Add superuser check https://review.openstack.org/77859 | 11:36 |
| *** Ryan_Lane has quit IRC | 11:38 | |
| *** talluri has quit IRC | 11:39 | |
| *** talluri has joined #openstack-infra | 11:39 | |
| *** talluri has quit IRC | 11:42 | |
| *** talluri has joined #openstack-infra | 11:42 | |
| *** talluri has quit IRC | 11:46 | |
| *** yamahata has joined #openstack-infra | 11:50 | |
| *** hashar_ has quit IRC | 11:53 | |
| *** vkozhukalov has joined #openstack-infra | 11:55 | |
| *** VINOD has joined #openstack-infra | 11:55 | |
| *** lcestari_ has joined #openstack-infra | 11:57 | |
| *** heyongli has joined #openstack-infra | 11:59 | |
| *** VinayMahuli has quit IRC | 12:01 | |
| *** ekarlso has quit IRC | 12:09 | |
| *** sahumada_ is now known as sahumada | 12:09 | |
| *** andre__ has quit IRC | 12:10 | |
| *** andreaf has joined #openstack-infra | 12:10 | |
| *** yamahata has quit IRC | 12:10 | |
| openstackgerrit | Sergey Kolekonov proposed a change to openstack-infra/jenkins-job-builder: Added send-to options support to email-ext plugin https://review.openstack.org/73601 | 12:18 |
| *** eharney has quit IRC | 12:22 | |
| *** dizquierdo has joined #openstack-infra | 12:33 | |
| *** miqui has joined #openstack-infra | 12:36 | |
| *** david-lyle has quit IRC | 12:39 | |
| *** saju_m has quit IRC | 12:40 | |
| *** mwagner_lap has quit IRC | 12:40 | |
| *** hashar has joined #openstack-infra | 12:41 | |
| *** zhiyan has left #openstack-infra | 12:43 | |
| *** zhiyan has joined #openstack-infra | 12:43 | |
| *** mfink has quit IRC | 12:46 | |
| *** jamielennox is now known as jamielennox|away | 12:50 | |
| *** saju_m has joined #openstack-infra | 12:51 | |
| *** sandywalsh_ has joined #openstack-infra | 12:55 | |
| sdague | fungi: you up yet? | 13:00 |
| *** pdmars has joined #openstack-infra | 13:01 | |
| *** pdmars has quit IRC | 13:01 | |
| *** pdmars has joined #openstack-infra | 13:02 | |
| *** markmcclain has joined #openstack-infra | 13:03 | |
| *** afazekas has quit IRC | 13:03 | |
| *** noorul has joined #openstack-infra | 13:04 | |
| noorul | hello | 13:04 |
| noorul | anyone running devstack on debian here? | 13:04 |
| *** saju_m has quit IRC | 13:04 | |
| *** eharney has joined #openstack-infra | 13:05 | |
| *** dhellmann_ is now known as dhellmann | 13:11 | |
| sdague | noorul: doubtful | 13:12 |
| noorul | sdague: It stopped working for me | 13:12 |
| sdague | noorul: you are going to need to be more specific | 13:13 |
| noorul | https://github.com/openstack-dev/devstack/blob/master/lib/cinder#L496 | 13:13 |
| sdague | we started enforcing -o errexit, so my assumption is there was some bad code in the debian only paths | 13:13 |
| noorul | It fails there | 13:13 |
| sdague | noorul: ok, why? | 13:14 |
| sdague | what's the actual error message? | 13:14 |
| noorul | I could only find start command in upstream, but I never used it | 13:14 |
| jaypipes | mayu: ping. I replied to you via email about the above error. | 13:14 |
| sdague | ok, so that never actually worked in debian before, but we used to ignore the fact that it wasn't working | 13:14 |
| sdague | noorul: dpkg -S /sbin/stop | 13:15 |
| sdague | upstart: /sbin/stop | 13:15 |
| *** miqui_ has joined #openstack-infra | 13:15 | |
| sdague | so it sounds like a different path is needed for debian if you aren't running upstart | 13:15 |
| *** miqui has quit IRC | 13:15 | |
| noorul | It is not easy to install upstart on wheezy and I never had it installed and devestack was working | 13:16 |
| sdague | noorul: it wasn't working | 13:16 |
| sdague | it was failing, and not telling yuo | 13:16 |
| noorul | sdake: devstack was working | 13:16 |
| noorul | sdake: stopped working from last couple of days | 13:17 |
| *** flaper87 has quit IRC | 13:17 | |
| *** flaper87 has joined #openstack-infra | 13:17 | |
| *** jp_at_hp has quit IRC | 13:17 | |
| *** jp_at_hp has joined #openstack-infra | 13:17 | |
| sdague | noorul: so someone that runs debian is going to need to provide a patch that fixes that piece of logic | 13:17 |
| *** eharney has quit IRC | 13:18 | |
| *** eharney has joined #openstack-infra | 13:18 | |
| sdague | because that piece of logic never worked on debian | 13:18 |
| noorul | Is it because of -o errexit ? | 13:18 |
| sdague | and now we actually stop when he hit errors | 13:18 |
| sdague | yes | 13:18 |
| sdague | instead of ignoring the errors | 13:18 |
| sdague | and assuming everything worked, when it didn't | 13:18 |
| sdague | and you only discovering that very late | 13:18 |
| noorul | huh | 13:18 |
| *** dkranz has joined #openstack-infra | 13:19 | |
| noorul | that means if go back everything should work fine | 13:19 |
| noorul | s/if/if I/ | 13:19 |
| sdague | it means this will fail silently | 13:20 |
| noorul | I understand that | 13:21 |
| noorul | thank you! | 13:21 |
| sdague | so it would be good to actually fix this, but that requires a debian person helping figure out what the working code is there | 13:21 |
| *** markmcclain has quit IRC | 13:23 | |
| *** ekarlso has joined #openstack-infra | 13:23 | |
| *** miqui_ is now known as miqui | 13:23 | |
| noorul | debian tgt lacks init script, so might not be straight forward | 13:23 |
| sdague | noorul: completely? | 13:24 |
| sdague | how does the start call work then? | 13:24 |
| noorul | which start call? | 13:24 |
| *** dcramer__ has quit IRC | 13:25 | |
| noorul | https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=577925 | 13:26 |
| *** freyes_ has joined #openstack-infra | 13:26 | |
| noorul | sdague: Can I add scripts to devstack? | 13:30 |
| *** mfink has joined #openstack-infra | 13:30 | |
| *** proffalken has quit IRC | 13:30 | |
| *** weshay has joined #openstack-infra | 13:31 | |
| noorul | sdague: The bug has a script attached. I can add that to devstack and invoke it in the case of debian. | 13:31 |
| sdague | noorul: I'd really like to not do that. So this means cinder never worked in debian devstack right? | 13:33 |
| sdague | because there was never an init script | 13:33 |
| sdague | so we could never run it | 13:34 |
| sdague | we should ask zigo about fixing the package, he's been doing the rest of the openstack packages | 13:34 |
| zigo | sdague: What? :) | 13:35 |
| zigo | What is there to fix? | 13:35 |
| noorul | if we fix this problem, cinder might run | 13:36 |
| sdague | zigo: apparently tgt in debian doesn't have init scripts | 13:36 |
| *** dhellmann is now known as dhellmann_ | 13:36 | |
| noorul | zigo: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=577925 | 13:36 |
| sdague | which was not noticed until we turned on -o errexit in devstack | 13:36 |
| zigo | sdague: I've tried *very hard* to fix that in the stable release, but the releaes team *REFUSED* that I update the package in Stable. | 13:36 |
| *** thuc has joined #openstack-infra | 13:36 | |
| zigo | That's not fixable, unfortunately. | 13:37 |
| zigo | Unless you engage (yourself) into a political debate with the (IMO stupid in this case) Debian release team. | 13:37 |
| zigo | sdague: However, I could do a backport, if you think that's enough. | 13:38 |
| zigo | FYI, I was the one fixing the bug in Sid / Testing. | 13:38 |
| noorul | zigo: to wheezy-backports ? | 13:40 |
| zigo | Yeah. | 13:40 |
| noorul | That would help! | 13:40 |
| *** markmc has quit IRC | 13:40 | |
| *** markmc has joined #openstack-infra | 13:40 | |
| *** thuc has quit IRC | 13:41 | |
| sdague | so I'll let you and noorul figure out the right approach. Because getting involved in debian politics is something I'd like to avoid :) | 13:41 |
| zigo | Ok, will do a backport then. | 13:41 |
| noorul | zigo: thank you! once that is one may be I can tweak devstack and send a patch. | 13:42 |
| noorul | s/one/done | 13:42 |
| *** krtaylor has joined #openstack-infra | 13:44 | |
| *** thuc has joined #openstack-infra | 13:47 | |
| *** thuc has quit IRC | 13:48 | |
| *** markmcclain has joined #openstack-infra | 13:48 | |
| *** thuc has joined #openstack-infra | 13:48 | |
| *** proffalken has joined #openstack-infra | 13:48 | |
| *** dcramer__ has joined #openstack-infra | 13:49 | |
| zigo | noorul: & sdague: I uploaded tgt_1.0.17-1.1~bpo70+1, it's going to take a few days to get approved by the backports FTP masters. | 13:49 |
| *** thuc_ has joined #openstack-infra | 13:49 | |
| zigo | Hopefully, they wont ask the stupid questions ... | 13:50 |
| sdague | great, thanks folks | 13:50 |
| zigo | In the mean time, the package is available here: http://archive.gplhost.com/pub/backports/tgt/ | 13:51 |
| *** hashar has quit IRC | 13:51 | |
| zigo | (amd64 only) | 13:51 |
| zul | hah nova on ubuntu reached a milestone today: nova is officially evil on ubuntu: http://bazaar.launchpad.net/~ubuntu-server-dev/nova/icehouse/revision/666 | 13:52 |
| *** hashar has joined #openstack-infra | 13:52 | |
| *** thuc has quit IRC | 13:53 | |
| zigo | sdague: By the way, upstart will slowly die in Debian, so I don't think it's a good idea to rely on start/stop, better use "service", IMO. | 13:54 |
| zigo | "service" will work whatever the init system is: sysv-rc, systemd, OpenRC (which I maintain), or upstart. | 13:54 |
| sdague | zigo: agree | 13:54 |
| *** nosnos has quit IRC | 13:56 | |
| *** rfolco has joined #openstack-infra | 13:56 | |
| *** VINOD has quit IRC | 13:58 | |
| noorul | zigo: The init script on the patch shows output http://paste.openstack.org/show/72174/ | 14:00 |
| *** jooools1 has quit IRC | 14:00 | |
| *** jooools has joined #openstack-infra | 14:01 | |
| *** heyongli has quit IRC | 14:01 | |
| *** CaptTofu has quit IRC | 14:01 | |
| *** jooools has quit IRC | 14:02 | |
| zigo | noorul: Well, load the kernel module that it needs then! :) | 14:02 |
| zigo | I got no idea what kernel module that is though. | 14:03 |
| fungi | mikal: the reason the review link didn't get posted to bug 1287542 is that the bug was opened against nova but the proposed patch is to openstack/requirements | 14:04 |
| *** afazekas has joined #openstack-infra | 14:04 | |
| fungi | sdague: need some expediting of a fix for the oslo.messaging issue? | 14:04 |
| sdague | fungi: probably | 14:05 |
| sdague | but I think we're going to fix it in nova | 14:05 |
| noorul | zigo: me too | 14:05 |
| openstackgerrit | Nikita Konovalov proposed a change to openstack-infra/storyboard: Load superusers from a yaml file https://review.openstack.org/77884 | 14:06 |
| fungi | sdague: great! in that case, seeing nothing else obviously fubar this morning, i'll get back to my regularly scheduled improvements | 14:07 |
| *** wchrisj has joined #openstack-infra | 14:08 | |
| sdague | that's our biggest fubar | 14:08 |
| *** annegentle has joined #openstack-infra | 14:09 | |
| sdague | so I have no idea if there are others :) | 14:09 |
| openstackgerrit | Nikita Konovalov proposed a change to openstack-infra/storyboard: Fix session usage in db_api https://review.openstack.org/77885 | 14:09 |
| *** CaptTofu has joined #openstack-infra | 14:09 | |
| *** jooools has joined #openstack-infra | 14:10 | |
| fungi | sdague: if any are smaller than that, i'm sure people will have no problem taking care of them ;) | 14:10 |
| *** mbacchi has joined #openstack-infra | 14:12 | |
| *** pblaho has quit IRC | 14:13 | |
| sdague | fungi: so this https://review.openstack.org/#/c/77874 will be the thing we want to promote | 14:13 |
| sdague | because nova code can't pass pep8 until we do | 14:13 |
| fungi | okay, reqs it is | 14:13 |
| sdague | but it needs to clear check first | 14:14 |
| *** smarcet has joined #openstack-infra | 14:14 | |
| fungi | got it, so duplicate code in o.mess and o-inc, fixed in the latter and re-sync'd to nova? | 14:16 |
| fungi | s/fixed in/removed from/ | 14:16 |
| *** mriedem has joined #openstack-infra | 14:17 | |
| *** julim has joined #openstack-infra | 14:17 | |
| sdague | fungi: yep | 14:18 |
| sdague | the analysis is also on list. DinaBelova did the heavy lifting in figuring it out | 14:18 |
| *** mayu_ has joined #openstack-infra | 14:19 | |
| fungi | DinaBelova: great job! | 14:20 |
| mayu_ | jaypipes:hi | 14:20 |
| DinaBelova | fungi, thanks :) | 14:20 |
| jaypipes | mayu: did you get my latest email? | 14:20 |
| DinaBelova | U found this issue when it was not released, but it caused so many errors only after that | 14:20 |
| mayu_ | jaypipes:yes | 14:22 |
| sdague | noorul: https://review.openstack.org/77889 - so I think that's the right fix regardless | 14:24 |
| sdague | then when the init scripts show up, it should just work | 14:24 |
| *** jeckersb_gone is now known as jeckersb | 14:25 | |
| mayu_ | jaypipes: jenkins_jobs command not found on the master | 14:26 |
| jaypipes | mayu: then I don't believe that bash install_master.sh has run successfully, since JJB would be installed... | 14:27 |
| *** jgallard has joined #openstack-infra | 14:28 | |
| *** dizquierdo is now known as dizquierdo_afk | 14:29 | |
| *** banix has joined #openstack-infra | 14:30 | |
| *** thuc_ has quit IRC | 14:31 | |
| mayu_ | jaypipes: I check just now, it jenkins-jobs not jenkins_jobs | 14:31 |
| *** thuc has joined #openstack-infra | 14:32 | |
| *** homeless has joined #openstack-infra | 14:32 | |
| jaypipes | mayu: doh! sorry about that :( | 14:33 |
| * jaypipes always gets that mixed up .. | 14:33 | |
| mayu_ | jaypipes: I had restarted jenkins service without jenkins-jobs, is that function ? | 14:34 |
| mayu_ | jaypipes: or i must exec jenkins-jobs command ? | 14:34 |
| jaypipes | mayu: no, jenkins-jobs is not same as jenkins... you will need to do: | 14:34 |
| jaypipes | sudo jenkins-jobs --flush-cache update /etc/jenkins_jobs/config | 14:35 |
| *** changbl has quit IRC | 14:35 | |
| mayu_ | jaypipes:ok, I will do that | 14:35 |
| jaypipes | ok | 14:35 |
| *** mfer has joined #openstack-infra | 14:36 | |
| *** thuc has quit IRC | 14:36 | |
| *** dims has quit IRC | 14:43 | |
| *** thomasem has joined #openstack-infra | 14:44 | |
| *** dims has joined #openstack-infra | 14:46 | |
| *** zehicle_at_dell has quit IRC | 14:46 | |
| *** banix has quit IRC | 14:55 | |
| *** odyssey4me has quit IRC | 14:59 | |
| *** zns has joined #openstack-infra | 14:59 | |
| *** gokrokve_ has quit IRC | 15:01 | |
| *** sandywalsh has quit IRC | 15:01 | |
| *** gokrokve has joined #openstack-infra | 15:01 | |
| *** jnoller has joined #openstack-infra | 15:02 | |
| ttx | sdague: some nova changes got enqueued in gate... Is your workaround in already ? | 15:05 |
| *** dkranz has quit IRC | 15:06 | |
| markmc | ttx, fwiw, hopefully there'll be a new release of oslo.messaging in 30 minutes | 15:06 |
| markmc | mordred and I were just talking about this gate asymmetry yesterday | 15:07 |
| mordred | what did I do | 15:07 |
| mordred | ? | 15:07 |
| ttx | was just wondering if we shouldn't fail fast the nova changes that got enqueued before they wreck havoc | 15:07 |
| ttx | or if the workaround is already in place | 15:07 |
| markmc | mordred, oslo.messaging change that broke nova's pep8 jobs, but only when oslo.messaging was released | 15:08 |
| mordred | jeblair: heads up - there may be a bug somewhere in zuul merger related to branch creation. we've seen twice now the merger flip out about not being able to find a branch it darned well should have been able to find - that has been solved by removing the cached repo and restarting the merger | 15:08 |
| fungi | ttx: i believe if the nova oslo-incubator sync patch has been approved, so if i promote it to the top of the gate it should shield us from further nova bombs | 15:09 |
| mordred | jeblair: I mention it because branch creation isn't something we do a lot, so could totally be something that's latent that we haven't tripped | 15:09 |
| mordred | jeblair: I don't have a repeatable test case yet though | 15:09 |
| markmc | ttx, the first change in the queue is https://review.openstack.org/77874 | 15:09 |
| mordred | markmc: ah yes. that old thing | 15:09 |
| ttx | markmc: ah, great | 15:10 |
| markmc | ttx, so, assuming that doesn't get kicked out, we're ok | 15:10 |
| ttx | sdague, fungi: ignore me | 15:10 |
| fungi | ttx: nevermind--there are no nova changes ahead of 77874,2 in the gate anyway | 15:10 |
| ttx | better safe than sorry I guess | 15:10 |
| fungi | er, markmc just said that. sorry | 15:10 |
| *** mayu_ has quit IRC | 15:11 | |
| jeblair | mordred: good to know. we did just create a branch on keystone yesterday | 15:12 |
| *** jergerber has joined #openstack-infra | 15:12 | |
| jeblair | mordred: do you have someone tracking down that bug? | 15:12 |
| mordred | jeblair: hopefully | 15:13 |
| *** freyes_ has quit IRC | 15:13 | |
| jeblair | mordred: maybe you could open a bug in launchpad and paste some tracebacks and debug logs? i can help take a look. | 15:13 |
| mordred | jeblair: will do | 15:13 |
| *** lcheng_ has joined #openstack-infra | 15:13 | |
| mordred | jeblair: the potentially _more_ distressing thing is that when it happens, the merger ceases processing further actino | 15:13 |
| mordred | bug/traceback coming | 15:14 |
| sdague | fungi: so if we get a reset - 77874,2 should promote | 15:15 |
| fungi | sdague: so that we stop blocking the check pipeline you mean? | 15:15 |
| fungi | sdague: because at this point there's nothing ahead of it in the gate which would benefit from promoting that change as far as i can tell | 15:15 |
| sdague | fungi: yeh, just to stop blocking check | 15:17 |
| fungi | sdague: okay, i'll try to catch the next reset and kick that forward | 15:18 |
| *** mgagne has joined #openstack-infra | 15:18 | |
| jeblair | mordred: oh, that's okay then. we have 2 mergers so as long as we don't create 2 branches we're okay! ;) | 15:18 |
| jeblair | mordred: i will check to see if one of our mergers broke after we created the keystone branch | 15:18 |
| mordred | jeblair: hehe | 15:18 |
| *** jgrimm has joined #openstack-infra | 15:20 | |
| jeblair | mordred: both of our mergers are still processing things | 15:21 |
| mordred | cool | 15:21 |
| mordred | jeblair: fwiw, we have added a bunch of branches and it's only failed a couple of times - and I'm not 100% sure it's due to branch creation - it's just the first correlation that sprang to mind | 15:22 |
| *** mgagne has quit IRC | 15:22 | |
| *** chandan_kumar has joined #openstack-infra | 15:24 | |
| sdague | markmc: so I think the real fix for here is something that runs all the conf generators for projects | 15:25 |
| sdague | as a d-g job | 15:25 |
| sdague | that would have caught this | 15:25 |
| fungi | side by side comparison of the git server load averages is pretty interesting... http://fungi.yuggoth.org/tmp/git-cacti.html (01 and 02 haven't been replaced yet... i'll have 02 switching over in an hour or so) | 15:25 |
| sdague | but it would also cause and interesting new issue | 15:25 |
| markmc | sdague, this change could easily have broken nova's unit tests | 15:26 |
| sdague | oslo-incubator config options would be frozen | 15:26 |
| *** dizquierdo_afk is now known as dizquierdo | 15:26 | |
| sdague | markmc in what way? | 15:26 |
| markmc | sdague, running nova's unit/pep8 jobs with oslo.messaging master would be needed to fix that asymmetry | 15:26 |
| sdague | markmc: not entirely | 15:26 |
| mordred | unittests/pep8 in devstack ftw | 15:26 |
| *** yamahata has joined #openstack-infra | 15:26 | |
| markmc | sdague, if oslo.messaging had registered that config option at runtime, it would have broken nova | 15:27 |
| markmc | sdague, it's pretty weird that it didn't happen at runtime, really | 15:27 |
| jeblair | fungi: i assume we're seenig 05 slowly ramp up as it increases unpacked refs | 15:27 |
| sdague | because nothing tests that | 15:27 |
| markmc | sdague, it's not the same as the issue with adding new config options breaking the config generator test | 15:27 |
| sdague | markmc: as far as I can tell the whole ca path is unexercised in the gate | 15:28 |
| *** david-lyle has joined #openstack-infra | 15:28 | |
| markmc | sdague, ca path? | 15:28 |
| fungi | jeblair: that's what i suspect. which leads to my earlier suggestion that we should consider repacking more often... any objection to a change for daily repack runs? | 15:28 |
| sdague | the config option that was hit here | 15:28 |
| sdague | ca_file | 15:28 |
| markmc | yes, that's why I removed it from oslo.messaging | 15:28 |
| sdague | we don't run d-g jobs that actually exercise any of those | 15:28 |
| sdague | which is why it was only found in pep8 | 15:29 |
| sdague | when we walk all the config options | 15:29 |
| * markmc understands how it happened | 15:29 | |
| markmc | point is | 15:29 |
| markmc | we could easily find ourselves merging something into oslo.messaging | 15:29 |
| markmc | which only breaks unit tests or pep8 jobs | 15:29 |
| markmc | not devstack | 15:29 |
| sdague | sure | 15:29 |
| markmc | and we only catch that when we release | 15:29 |
| jeblair | fungi: nope | 15:29 |
| * fungi shall submit that for review forthwith | 15:30 | |
| sdague | markmc: so can we actually test setting the tag? | 15:30 |
| sdague | so we actually test the release would work, before we cut the release | 15:31 |
| markmc | maybe | 15:31 |
| markmc | or do unit tests and pep8 on master oslo.messaging | 15:31 |
| mordred | sdague: do you mean test the git tag before pushing the git tag? | 15:32 |
| sdague | mordred: yeh | 15:32 |
| mordred | sdague: it would have to be manual testing | 15:32 |
| sdague | markmc: so if we do unit tests & pep8 on master | 15:32 |
| mordred | sdague: I've wanted for a while the ability to propose a tag to gerrit for review | 15:32 |
| mordred | jeblair: oh! so... terrible idea | 15:32 |
| sdague | it means that running unit tests *not* in a devstack environment would have broken here instead | 15:33 |
| mordred | jeblair: oh- nevermind. I realized in typing it out that it won't work | 15:33 |
| sdague | it seems like the fundamental concept in zuul is to test a new state of the world, before it happens | 15:34 |
| sdague | to make sure that it's good before we expose it to everyone | 15:34 |
| markmc | sdague, huh? why would it have broken that? | 15:34 |
| sdague | markmc: so with this change | 15:34 |
| markmc | sdague, ok, got it | 15:35 |
| markmc | sdague, yeah, this one was weird | 15:35 |
| markmc | sdague, essentially an API change that nova needed to be ported to | 15:35 |
| *** lcheng_ has quit IRC | 15:35 | |
| markmc | sdague, which is why I removed those options from oslo.messaging now | 15:35 |
| sdague | yep | 15:35 |
| sdague | but I think this is going to just happen more an more with oslo libs that also include oslo-incubator code | 15:36 |
| markmc | sdague, we would never have been able to merge this change in the first place | 15:36 |
| sdague | markmc: yep | 15:36 |
| sdague | markmc: and actually, you'd never be able to change those options ever again | 15:36 |
| sdague | because you'd have 2 trees that needed to update simultaneously | 15:36 |
| *** dcramer__ has quit IRC | 15:37 | |
| *** ryanpetrello has joined #openstack-infra | 15:37 | |
| sdague | the only way to update would be to remove the code from one, update the other, then bring it back in on the first tree | 15:37 |
| markmc | sdague, we shouldn't overthink the options thing - it shouldn't happen | 15:37 |
| markmc | sdague, oslo libs should not include options from oslo-incubator | 15:37 |
| sdague | markmc: what about lockutils | 15:37 |
| markmc | sdague, only reason I let it pass in this case was they're not used at runtime | 15:37 |
| markmc | sdague, lockutils should become a library before any oslo libraries use it | 15:38 |
| sdague | this is actually a real issue, which we were going through the other day | 15:38 |
| sdague | because keystone client can't include lockutils | 15:38 |
| sdague | because of the config option collision | 15:38 |
| *** coolsvap has joined #openstack-infra | 15:38 | |
| sdague | oslo libs aren't the only place this comes up | 15:38 |
| markmc | yes, it shouldn't until lockutils is a library | 15:38 |
| markmc | my guess is keystone client doesn't even want the code which uses the options | 15:39 |
| sdague | the client libraries hit this as well | 15:39 |
| markmc | keystone client would use the oslo lockutils library | 15:39 |
| markmc | if it's not willing to depend on such a library, it shouldn't be proposing to use oslo-incubator code | 15:39 |
| *** wenlock has joined #openstack-infra | 15:39 | |
| sdague | so keystone client currently pulls in memorycache, which has config | 15:40 |
| * markmc sighs | 15:41 | |
| sdague | maybe I'm wrong, but I actually think this is a real issue around config bits in the incubator putting us in funny wedges | 15:42 |
| openstackgerrit | Jeremy Stanley proposed a change to openstack-infra/config: Move git02.openstack.org to a new server https://review.openstack.org/77915 | 15:42 |
| markmc | sdague, there is a real issue with libraries using code with config options from oslo-incubator, yes | 15:42 |
| markmc | sdague, and the answer is they shouldn't do that | 15:42 |
| sdague | markmc: ok, well we're a ways from that | 15:43 |
| markmc | sdague, "a ways" == "keystoneclient is using memorycache" ? | 15:43 |
| markmc | sdague, or more? | 15:43 |
| *** thedodd has joined #openstack-infra | 15:44 | |
| *** lcheng_ has joined #openstack-infra | 15:44 | |
| *** vkozhukalov has quit IRC | 15:45 | |
| *** mgagne has joined #openstack-infra | 15:46 | |
| sdague | actually, currently only looks like keystoneclient | 15:47 |
| sdague | but oslo.messaging has a need in the future? or was including that a mistake? | 15:47 |
| *** markmcclain has quit IRC | 15:47 | |
| markmc | see https://review.openstack.org/77891 | 15:48 |
| *** andre__ has joined #openstack-infra | 15:48 | |
| sdague | right, I saw you pulled it | 15:48 |
| markmc | I knew it was an issue when I merged using sslutils | 15:48 |
| markmc | but because the config options weren't actually used at runtime, I figured we'd be ok | 15:48 |
| openstackgerrit | Jeremy Stanley proposed a change to openstack-infra/config: Repack git repositories daily https://review.openstack.org/77919 | 15:48 |
| markmc | which proved incorrect | 15:48 |
| sdague | so it sounds like we actually need a job which checks for that in projects we consider libraries | 15:49 |
| fungi | i've disabled git02 in haproxy, in preparation for dns changes | 15:50 |
| *** noorul has left #openstack-infra | 15:52 | |
| *** johnthetubaguy has joined #openstack-infra | 15:54 | |
| *** sparkycollier has joined #openstack-infra | 15:54 | |
| *** ryanpetrello has quit IRC | 15:54 | |
| *** markmc has quit IRC | 15:55 | |
| *** nati_ueno has joined #openstack-infra | 15:57 | |
| *** johnthetubaguy1 has joined #openstack-infra | 15:57 | |
| *** coolsvap1 has joined #openstack-infra | 15:58 | |
| *** johnthetubaguy has quit IRC | 15:58 | |
| *** coolsvap has quit IRC | 16:00 | |
| *** pcrews has joined #openstack-infra | 16:01 | |
| *** ociuhandu_ has joined #openstack-infra | 16:01 | |
| *** hogepodge has joined #openstack-infra | 16:02 | |
| *** rcleere has joined #openstack-infra | 16:02 | |
| *** atiwari has joined #openstack-infra | 16:04 | |
| ekarlso | anyone looked at writing a angularjs page for zuul ? | 16:07 |
| *** ociuhandu_ has quit IRC | 16:08 | |
| *** dkranz has joined #openstack-infra | 16:11 | |
| *** dhellmann_ is now known as dhellmann | 16:12 | |
| jeblair | ekarlso: that would be great. check out the status page in the zuul repo itself | 16:13 |
| jeblair | ekarlso: i'd love it if we could use that directly and just add openstack styling, headers, and footers | 16:14 |
| *** markmcclain has joined #openstack-infra | 16:15 | |
| *** andre__ has quit IRC | 16:15 | |
| openstackgerrit | Spencer Krum proposed a change to openstack-infra/config: Add PuppetBoard to openstack infra https://review.openstack.org/77928 | 16:16 |
| jeblair | hooray! | 16:16 |
| nibalizer | fungi: https://review.openstack.org/77928 | 16:16 |
| * nibalizer raises roof | 16:16 | |
| nibalizer | im not entirely sure thats gonna work, but this way there is code for all of us to ponder, and maybe im less of a blocker | 16:17 |
| nibalizer | AaronGr: ^^ | 16:17 |
| jeblair | nibalizer: does the puppetdb have any sekrit information in it? | 16:17 |
| jeblair | nibalizer: (like the hiera secrets) | 16:18 |
| AaronGr | nibalizer: did you add something to your puppetboard module to handle the older apache module? | 16:18 |
| nibalizer | jeblair: i don't think so | 16:19 |
| *** rcleere has quit IRC | 16:19 | |
| dhellmann | sdague: I'm back online, if you want to talk about that library testing thing | 16:19 |
| nibalizer | jeblair: we should turn off the live querying page of puppetboard, which I can do in a commit later today i have to run now | 16:19 |
| nibalizer | AaronGr: i removed puppetlabs-apache as a requirement for my puppetlabs modul | 16:19 |
| nibalizer | and bumped major verion numbers as a way of saying sorry | 16:20 |
| AaronGr | nibalizer: ok, what about the parts of the apache module that aren't supported? i had to .erb most of the vhost to get it online with the o.o apache module | 16:20 |
| sdague | dhellmann: scrollback | 16:20 |
| nibalizer | also as AaronGr pointed out that creates a non ssl endpoint | 16:20 |
| dhellmann | sdague: I'm caught up | 16:21 |
| nibalizer | AaronGr: the puppetboard::apache::conf class .erbs a vhost file for you | 16:21 |
| anteaya | nibalizer: commented | 16:21 |
| nibalizer | whcih i think will be enough/work ? | 16:21 |
| nibalizer | anteaya: thanks! | 16:21 |
| anteaya | np, thanks for doing this | 16:21 |
| fungi | nibalizer: yay! | 16:21 |
| nibalizer | anteaya: im having a lot of fun | 16:22 |
| AaronGr | nibalizer: that was the only breakage i found setting it up, yeah | 16:22 |
| anteaya | nibalizer: I'm so glad to hear that! | 16:22 |
| nibalizer | ya after this lands you all will have to cook up a new project for me | 16:22 |
| dhellmann | sdague: I'll be in #openstack-oslo if you want to chat without interfering with business here | 16:22 |
| fungi | nibalizer: we already have dozens of new projects cooling for you on the window sill, no need to cook up more | 16:23 |
| fungi | sdague: 77874,2 just merged. rechecks for everybody! | 16:24 |
| anteaya | gate reset | 16:25 |
| fungi | jeblair: i think we got another one of those false merge failures from gerrit... 77874,2 merged successfully according to gerrit. 70136,25 (the change behind it) restarted testing | 16:25 |
| fungi | i'm looking in the zuul log now to confirm | 16:25 |
| jeblair | fungi: that isn't immediately recalling to mind a problem of which i am aware... i may need some reminding... | 16:26 |
| *** VijayTripathi has joined #openstack-infra | 16:26 | |
| anteaya | jeblair: mordred was mentioning it earlier | 16:26 |
| fungi | anteaya: no, something different | 16:26 |
| anteaya | oh sorry | 16:26 |
| *** e0ne has joined #openstack-infra | 16:27 | |
| *** johnthetubaguy1 is now known as johnthetubaguy | 16:27 | |
| *** dkliban has quit IRC | 16:28 | |
| fungi | jeblair: it's been suggested by people who stare at the zuul status page more steadily than i that sometimes zuul restarts testing on changes immediately behind one that merges, more often when it's in a large set of changes all merging in rapid succession. and without a concrete example the best guess was just that sometimes gerrit incorrectly reports the merge result as negative | 16:28 |
| *** thuc has joined #openstack-infra | 16:28 | |
| *** hogepodge has quit IRC | 16:28 | |
| *** e0ne_ has quit IRC | 16:29 | |
| fungi | this is the first one where i was paying attention to the series of change numbers and saw one merge successfully and all those behind it restart testing | 16:29 |
| fungi | so looking now to see whether i can confirm it from the zuul logs | 16:29 |
| jeblair | fungi: ah. when i have stared at the page and observed that, i have usually noted that the problem is that the change did not actually merge because gerrit said it required a recursive merge | 16:30 |
| fungi | it's possible there was a recursive merge required on a change between 77874,2 and 70136,25 which is why i'm checking to make sure they really were adjacent | 16:31 |
| jeblair | cool, let me know what you find | 16:31 |
| fungi | 2014-03-04 16:24:21,721 INFO zuul.DependentPipelineManager: Resetting builds for change <Change 0x7fda290c5ed0 70136,25> because the item ahead, <QueueItem 0x7fda28a56650 for <Change 0x7fda2a0afb90 77874,2> in gate>, failed to merge | 16:32 |
| fungi | bingo | 16:32 |
| * anteaya wonders what a recursive merge is, and if there is anything to read about it | 16:32 | |
| fungi | now to the debug logs for actual gerrit responses | 16:32 |
| *** mrodden1 has quit IRC | 16:33 | |
| *** zehicle_at_dell has joined #openstack-infra | 16:34 | |
| *** mbacchi has quit IRC | 16:34 | |
| *** changbl has joined #openstack-infra | 16:35 | |
| jeblair | 2014-03-04 16:23:20,327 DEBUG zuul.Gerrit: Waiting for <Change 0x7fda2a0afb90 77874,2> to appear in git repo | 16:35 |
| jeblair | 2014-03-04 16:24:21,720 DEBUG zuul.Gerrit: Change <Change 0x7fda2a0afb90 77874,2> did not appear in the git repo | 16:35 |
| fungi | gerrit being slow? | 16:35 |
| fungi | this would fit with the "many merges in rapid succession" triggering pattern i suspect | 16:36 |
| fungi | nothing extremely anomalous on the cacti graph for review.o.o in that timeframe, but maybe it was so brief as to be washed out in the 5-minute sample | 16:38 |
| *** gmoro has quit IRC | 16:39 | |
| *** harlowja_away is now known as harlowja | 16:40 | |
| *** harlowja has quit IRC | 16:41 | |
| fungi | replication_timeout = 60 and replication_retry_interval = 5 so it tried a dozen times over the course of a minute for gerrit to make that ref available? | 16:41 |
| jeblair | fungi: possibly a few less since i think it takes a while to transfer the refs for nova... it's even possible it tried only once if it took > 60 seconds | 16:42 |
| jeblair | we should explore that | 16:42 |
| fungi | it did eventually replicate, because the cgit servers have it... http://git.openstack.org/cgit/openstack/nova/commit/?id=5eafe1c | 16:42 |
| nibalizer | fungi: excellent | 16:42 |
| fungi | jeblair: i wonder if repacking the refs on review.o.o would also help | 16:43 |
| fungi | more often, i mean | 16:43 |
| jeblair | fungi: i don't think we should pack the refs on gerrit itself; however, this check actually is against /p/project which should hit the local mirror (where we do pack refs, i think) | 16:44 |
| openstackgerrit | Spencer Krum proposed a change to openstack-infra/config: Add PuppetBoard to openstack infra https://review.openstack.org/77928 | 16:44 |
| jeblair | fungi: to be honest, i'm not sure this feature is strictly necessary for us anymore... | 16:45 |
| jeblair | fungi: zuul always fetches from gerrit ssh directly, and builds fetch from zuul | 16:46 |
| fungi | got it | 16:46 |
| jeblair | fungi: so strictly speaking, we don't actually care if the replicas are up to date, because the next builds are going to pull from zuul which will be | 16:46 |
| jeblair | (when we wanted to make sure that we didn't start any new jobs until jenkins could pull current data from replicas this was more important) | 16:47 |
| *** mrodden has joined #openstack-infra | 16:47 | |
| jeblair | however, clarkb and mordred do care that their replicas are up to date | 16:47 |
| fungi | jeblair: anyway, it looks like we do actually repack gerrit's backend repos... http://git.openstack.org/cgit/openstack-infra/config/tree/modules/gerrit/manifests/cron.pp#n16 | 16:47 |
| jeblair | since they have that weird firewall thing going on | 16:47 |
| fungi | weekly | 16:47 |
| jeblair | fungi: oh, yeah, we do repack it, but i don't think we rack refs | 16:47 |
| mordred | stupid firewall | 16:48 |
| *** gyee has joined #openstack-infra | 16:48 | |
| jeblair | pack refs even | 16:48 |
| fungi | and also the local mirror http://git.openstack.org/cgit/openstack-infra/config/tree/modules/openstack_project/manifests/gerrit.pp#n335 | 16:48 |
| fungi | ahh, repack but not pack refs. got it | 16:49 |
| jeblair | if we don't pack refs on the local mirror, i think we could if we wanted. | 16:49 |
| fungi | we don't currently, just repack, no pack-refs there | 16:49 |
| openstackgerrit | garyk proposed a change to openstack/requirements: Update oslo.vmware to be >= 0.2 https://review.openstack.org/77937 | 16:50 |
| jeblair | i'm trying to get some timing information to see how long this normally takes | 16:51 |
| fungi | though we do tend to use that local mirror less and less anyway. these days it's either latency-tolerant operations in which case we get refs from git.o.o, or it's latency-sensitive in which case we're hitting gerrit's internal copies directly, right? | 16:51 |
| jeblair | fungi: true; we might want to see how much other people are using it though | 16:51 |
| fungi | sure | 16:51 |
| *** SumitNaiksatam has quit IRC | 16:53 | |
| *** gmoro has joined #openstack-infra | 16:53 | |
| *** krotscheck has joined #openstack-infra | 16:56 | |
| sdague | fungi: any chance - devstack-precise-check-rax-iad-2049871 is still alive and holdable? | 16:56 |
| *** harlowja has joined #openstack-infra | 16:56 | |
| sdague | there is a grenade issue that's cropping up that I want to figure out | 16:57 |
| jeblair | fungi: the previous nova change was detected as merged in 47 seconds; typical times for other projects are <5sec, usually only 1 or 2. | 16:58 |
| fungi | sdague: if the job already ended, pretty much zero chance | 16:58 |
| fungi | but checking | 16:58 |
| fungi | sdague: nope. 2049871 no longer exists | 16:59 |
| *** markmc has joined #openstack-infra | 16:59 | |
| jeblair | change before that was 46 seconds. so yeah, we're pretty close to the limit on nova | 16:59 |
| sdague | fungi: ok | 16:59 |
| fungi | jeblair: makes sense. nova is a beast | 17:00 |
| jeblair | i'm going to profile how long a single check takes to see how many polls that time represents | 17:00 |
| openstackgerrit | Brad P. Crochet proposed a change to openstack-infra/jenkins-job-builder: Added support for Exclusion plugin https://review.openstack.org/77940 | 17:01 |
| sdague | fungi: ok, the bummer is this takes a while to show up. And i doubt I'll be lucky enough to find it | 17:01 |
| *** tomhe has joined #openstack-infra | 17:02 | |
| *** sparkycollier has quit IRC | 17:02 | |
| *** dkliban has joined #openstack-infra | 17:03 | |
| *** rcleere has joined #openstack-infra | 17:03 | |
| *** andre__ has joined #openstack-infra | 17:05 | |
| anteaya | okay, just had a toe to toe with the tail-f guy in -infra, just in case they come back to gerrit admins wanting to get voting turned on for their system, please refer them to me | 17:06 |
| anteaya | otherwise it might be a time sink for you | 17:06 |
| *** sparkycollier has joined #openstack-infra | 17:07 | |
| anteaya | sorry that conversation took place in -neutron | 17:07 |
| jeblair | anteaya: thanks | 17:09 |
| anteaya | jeblair: np | 17:09 |
| *** markwash has joined #openstack-infra | 17:10 | |
| *** dkliban is now known as dkliban_lunch | 17:10 | |
| jeblair | fungi: it looks like it normally takes 10-15 seconds to download the refs for nova, but i've seen it take as much as 25 | 17:11 |
| *** VijayTripathi has quit IRC | 17:11 | |
| fungi | i can imagine if gerrit gets good and busy, that goes up | 17:12 |
| *** khyati has joined #openstack-infra | 17:12 | |
| jeblair | fungi: yeah, i also wonder if with all our replication targets, replication takes a little bit longer sometimes.. | 17:12 |
| fungi | and we've got a couple more nova changes about to merge in 10 minutes or so if we need more current data points | 17:12 |
| *** khyati has quit IRC | 17:12 | |
| tomhe | jeblair: We're running Zuul with the new merger processes and observe that sometimes Zuul triggers jobs in the dependent pipeline manager with a ZUUL_COMMIT that belongs to some other change currently in the same pipeline. Have you observed anything similar on the Openstack Zuul instance? | 17:13 |
| *** nati_ueno has quit IRC | 17:13 | |
| *** rwsu has joined #openstack-infra | 17:14 | |
| *** dprince has joined #openstack-infra | 17:14 | |
| jeblair | tomhe: we don't generally use zuul_commit, so that could escape our notice | 17:14 |
| *** afazekas has quit IRC | 17:14 | |
| *** luisg has joined #openstack-infra | 17:14 | |
| tomhe | jeblair: OK. We use the git plugin in Jenkins to checkout the code and I believe that I need the ZUUL_COMMIT as branch | 17:15 |
| *** dcramer__ has joined #openstack-infra | 17:15 | |
| tomhe | jeblair: Tried FETCH_HEAD but the git plugin doesn't like that | 17:15 |
| jeblair | tomhe: yep. give me a few more minutes to finish up what i'm working on, then i'll look into this with you | 17:15 |
| *** hogepodge has joined #openstack-infra | 17:15 | |
| tomhe | jeblair: thanks | 17:16 |
| *** jcoufal has joined #openstack-infra | 17:16 | |
| *** luisg has quit IRC | 17:17 | |
| jeblair | fungi: hitting https://git.o.o produces times in the 9-15 second range, so fairly similar. hitting http://git01 (no ssl) is a bit faster, mostly around 7 secs, git04 is mostly around 6 secs | 17:17 |
| *** andre__ has quit IRC | 17:17 | |
| *** e0ne has quit IRC | 17:17 | |
| *** luisg has joined #openstack-infra | 17:17 | |
| *** whoops has joined #openstack-infra | 17:17 | |
| *** luis_ has quit IRC | 17:17 | |
| jeblair | so we might see a small improvement with packing refs in the local replica | 17:18 |
| tomhe | jeblair: We now have a Zuul_Check_Commit job in the gate pipeline that detects ZUUL_COMMIT != FETCH_HEAD, so we should be able to get more data now. | 17:18 |
| annegentle | anyone have the IRC channel for the OpenStack board meeting handy? | 17:18 |
| fungi | annegentle: #openstack-board | 17:18 |
| fungi | annegentle: with side conversations in #openstack-foundation | 17:18 |
| jeblair | fungi: i still feel like this is an unecessary check for us, and it's even a fairly fragile check for clarkb and mordred (since it doesn't necessarily check that replication has completed to the replica they actually care about) | 17:21 |
| jeblair | clarkb, mordred, fungi: i think we should consider removing it, or otherwise think about what it really should be doing | 17:21 |
| fungi | jeblair: i'm inclined to concur | 17:22 |
| mordred | jeblair: fine by me | 17:22 |
| jeblair | cause right now it's effective purpose is "make sure that replication has happened to the local apache replica on review.o.o which we pretty much don't use in zuul" | 17:22 |
| *** derekh has joined #openstack-infra | 17:23 | |
| *** derekh has quit IRC | 17:23 | |
| *** sarob_ has joined #openstack-infra | 17:23 | |
| jeblair | tomhe: brb | 17:24 |
| tomhe | jeblair: No problem. Thanks. I'm on Europe time and fixing dinner for my kids. I'll check back soon. | 17:25 |
| *** CaptTofu has quit IRC | 17:28 | |
| *** yassine has quit IRC | 17:28 | |
| openstackgerrit | Davanum Srinivas (dims) proposed a change to openstack-infra/config: Save md5 and sha1 checksums with artifacts https://review.openstack.org/76933 | 17:28 |
| mordred | dhellmann: someone just proposed adding tools/requirements_style_check.sh to storyboard | 17:28 |
| dhellmann | mordred: !? | 17:29 |
| mordred | dhellmann: it has a comment from you in it, so I'm assuming you've been involved in it in some way? | 17:29 |
| mordred | dhellmann: https://review.openstack.org/#/c/77085/1/tools/requirements_style_check.sh | 17:29 |
| dhellmann | mordred: I think that was the check script I wrote for use in the incubator repo | 17:29 |
| mordred | dhellmann: do you know where it came frmo? | 17:29 |
| dhellmann | that isn't really meant to be used elsewhere | 17:29 |
| mordred | yeah | 17:29 |
| mordred | I was trying to track down where people were coming from with it :) | 17:30 |
| mordred | because yeah - alphabetized requirements == fail | 17:30 |
| dhellmann | it runs as part of the pep8 check in the incubator | 17:30 |
| *** reed has joined #openstack-infra | 17:30 | |
| *** sc68cal has quit IRC | 17:30 | |
| *** andre__ has joined #openstack-infra | 17:31 | |
| mordred | k. well, it makes sense in the incubator | 17:31 |
| *** sc68cal has joined #openstack-infra | 17:31 | |
| reed | gusy, anybody can pushs this one in? It's urgent *and* trivial https://review.openstack.org/#/c/77458/ | 17:31 |
| dhellmann | I plan to remove it as soon as we graduate the 20+ libraries we'll be releasing in juno | 17:31 |
| dhellmann | oh, btw, heads up -- we're going to need more git repos soon :-) | 17:31 |
| reed | who's gusy? damn, worse typist ever | 17:31 |
| mordred | woot | 17:32 |
| fungi | reed: done | 17:32 |
| reed | thanks fungi | 17:32 |
| jeblair | dhellmann: no problem, mordred has promised that'll all be fixed up by mid march :) | 17:32 |
| dhellmann | jeblair: cool | 17:32 |
| *** derekh has joined #openstack-infra | 17:32 | |
| jeblair | and even if it isn't, we'd take the hit and do it manually | 17:33 |
| dhellmann | these won't be coming all at one time, but I do expect them to come pretty regularly early in the cycle | 17:33 |
| dhellmann | I need to think about how to solve the issue we had recently with library releases breaking things in unit tests, though | 17:33 |
| *** cody-somerville has joined #openstack-infra | 17:33 | |
| *** markwash has quit IRC | 17:33 | |
| fungi | dhellmann: it would be possible to run unit tests of all depending projects on each commit of the dependent library | 17:34 |
| fungi | but that would be a lot of jobs, potentially | 17:34 |
| dhellmann | fungi: yeah, we started setting up tools like that for oslotest | 17:34 |
| dhellmann | yeah, 20+ repos running against each other and the apps, that's a lot of jobs | 17:35 |
| fungi | though in theory those libraries are individually small and have fewer changes than, say, nova | 17:36 |
| *** sc68cal has quit IRC | 17:36 | |
| dhellmann | we could add 1 job to each app that pulls in libs from source and then runs the apps unit tests, but that still lets us break apps | 17:36 |
| dhellmann | fungi: oh, man, I hope so | 17:36 |
| *** sc68cal has joined #openstack-infra | 17:36 | |
| dhellmann | I expect a lot of them to be fairly static, actually | 17:36 |
| *** bhuvan has joined #openstack-infra | 17:36 | |
| *** bhuvan_ has joined #openstack-infra | 17:36 | |
| *** andre__ has quit IRC | 17:36 | |
| fungi | in which case lots of jobs for them isn't a huge resource concern | 17:37 |
| jeblair | that makes having a lot of jobs easier to swallow | 17:37 |
| jeblair | devstack-gate has quite a number of jobs, for instance. | 17:37 |
| fungi | more of a blocking issue since it's easy for other breakage in a unit test to block development on that library | 17:37 |
| *** cadenzajon has joined #openstack-infra | 17:38 | |
| fungi | multiply the chance of some spontaneous outside issue impacting a random project's unit test viability by the number of projects you're testing the lib against and it could get hairy | 17:38 |
| *** nicedice has joined #openstack-infra | 17:38 | |
| jeblair | tomhe: you said in an email that this happens when you have 2 changes for different branches of the same project... i'm going to see if i can repro a test case | 17:38 |
| dhellmann | fungi: that smells like gridlock | 17:38 |
| fungi | dhellmann: could run them non-voting and use them solely for advisory data (that's what we do with a lot of the backward-compat tests for d-g) | 17:40 |
| tomhe | jeblair: I thought then that it might be related to cherry-picked commits but today we have seen it on regular commits. I'll go back and see if I can verify that this also targeted different branches. | 17:40 |
| dhellmann | fungi: that may work, I think I can rely on the oslo team to look at failures to see if they are real | 17:40 |
| fungi | dhellmann: though it does leave the risk that someone ignores a legitimate failure that patch is causing and crams it through, or is blind to a legitimate issue because there's another current problem in the affected project. but maybe that wouldn't be a frequent issue in practice | 17:41 |
| dhellmann | fungi: we could also have them be voting for the more critical libs like messaging and db, I guess | 17:41 |
| dhellmann | although if the others aren't critical, being blocked may not be a problem | 17:42 |
| fungi | pragmatism for the win | 17:42 |
| dhellmann | fungi: right | 17:43 |
| fungi | does oslo.test already implement the cross-testing you would need for this (e.g. check out nova master, tweak the tox venv to install a locally-provided oslo.messaging, then run nova unit tests)? | 17:44 |
| *** jaypipes has quit IRC | 17:45 | |
| dhellmann | fungi: it does, but the script lives in the oslo.test library, so that would need to move to the infra repo, and probably be changed slightly to know which library needs to be updated (I think oslotest is hard-coded now) | 17:45 |
| fungi | okay, just making sure i understand the intent of oslotest | 17:45 |
| dhellmann | fungi: the job definitions are still up for review | 17:45 |
| fungi | cool beans | 17:45 |
| dhellmann | yeah, that's just the base class and some fixtures used in unit tests | 17:45 |
| *** jraim_ has joined #openstack-infra | 17:45 | |
| tomhe | jeblair: I can confirm now that the changes are targeted for different branches. | 17:46 |
| dhellmann | fungi: see https://review.openstack.org/#/c/76381/2 and https://review.openstack.org/#/c/76654/ and http://git.openstack.org/cgit/openstack/oslo.test/tree/tools/run_cross_tests.sh | 17:46 |
| fungi | dhellmann: yep, hoping to get to those two today | 17:47 |
| jeblair | tomhe: yeah, i think i have a reproduction by making a slight change to the "test_build_configuration_multi_branch" test; relevant output is here: http://paste.openstack.org/show/72251/ | 17:47 |
| dhellmann | fungi: oh, I thought we were waiting until the feature freeze lifted? I'm fine with going ahead, I had just put it on the back-burner for now | 17:47 |
| fungi | dhellmann: well, wanted to look at them, not approve them ;) | 17:47 |
| dhellmann | fungi: ah, ok :-) | 17:47 |
| *** CaptTofu has joined #openstack-infra | 17:48 | |
| jeblair | tomhe: A and C are on the same branch, B is on a different one | 17:48 |
| tomhe | jeblair: Spot on | 17:48 |
| *** VijayTripathi has joined #openstack-infra | 17:48 | |
| *** mwagner_lap has joined #openstack-infra | 17:49 | |
| clarkb | jeblair: which check? | 17:49 |
| *** andre__ has joined #openstack-infra | 17:50 | |
| jeblair | clarkb: zuul waiting for the change to show up in the apache replica before considering a change merged | 17:50 |
| *** hashar has quit IRC | 17:51 | |
| *** vish1 has joined #openstack-infra | 17:52 | |
| clarkb | jeblair: I think we can live without that check. Its prior to merge that we need the intermediate refs to show up | 17:52 |
| *** chandan_kumar has quit IRC | 17:53 | |
| *** zns has quit IRC | 17:53 | |
| *** che-arne has quit IRC | 17:53 | |
| *** jraim has quit IRC | 17:53 | |
| *** vishy has quit IRC | 17:53 | |
| *** bradm has quit IRC | 17:53 | |
| *** vish1 is now known as vishy | 17:53 | |
| *** jraim_ is now known as jraim | 17:53 | |
| clarkb | which is my proposed change that should get tests written when feature freeze is over | 17:53 |
| *** mbacchi has joined #openstack-infra | 17:54 | |
| pleia2 | zaro: testing your change, thumbs up on fedora, trying centos now | 17:54 |
| clarkb | pleia2: can you also test precise for completeness (if that isn't too much trouble) | 17:55 |
| pleia2 | clarkb: will do | 17:55 |
| clarkb | pleia2: thank you | 17:55 |
| *** gokrokve has quit IRC | 17:55 | |
| openstackgerrit | A change was merged to openstack-infra/config: Add track chairs mailing list https://review.openstack.org/77458 | 17:55 |
| *** sarob_ has quit IRC | 17:55 | |
| *** gokrokve has joined #openstack-infra | 17:56 | |
| *** sarob_ has joined #openstack-infra | 17:56 | |
| openstackgerrit | A change was merged to openstack-infra/storyboard: Decoupled API story and task creation. https://review.openstack.org/77668 | 17:56 |
| * clarkb catches up on sb | 17:57 | |
| *** jlibosva has quit IRC | 17:58 | |
| *** mayu has quit IRC | 17:58 | |
| jeblair | tomhe: i think i found the bug | 17:59 |
| tomhe | jeblair: Great | 17:59 |
| openstackgerrit | A change was merged to openstack-infra/storyboard-webclient: Task, story, and project fixes https://review.openstack.org/77013 | 17:59 |
| *** bhuvan__ has joined #openstack-infra | 17:59 | |
| *** bhuvan___ has joined #openstack-infra | 17:59 | |
| tomhe | jeblair: That was incredibly fast | 17:59 |
| Alex_Gaynor | If anyone's interested, https://caniusepython3.com/check/b77d1b26-8866-434d-8d98-c620d56122bf shows the current status of our deps with respect to py3kj | 18:00 |
| *** ndipanov has quit IRC | 18:00 | |
| markmc | mordred, I'd never actually looked at openstack/openstack - very nice | 18:00 |
| markmc | mordred, why is Author: jenkins until recently? | 18:00 |
| jeblair | tomhe: being able to reproduce complex scenarious in the functional test suite is _really_ handy | 18:00 |
| anteaya | Alex_Gaynor: thank you | 18:00 |
| *** gokrokve has quit IRC | 18:00 | |
| *** bradm has joined #openstack-infra | 18:00 | |
| tomhe | jeblair: I get that | 18:00 |
| jeblair | tomhe: (and there was already a test for this scenario, it just wasn't checking ZUUL_COMMIT | 18:00 |
| pleia2 | anteaya: ooh, exciting to see the puppetboard changes coming in \o/ | 18:01 |
| *** jlibosva has joined #openstack-infra | 18:01 | |
| *** gokrokve has joined #openstack-infra | 18:01 | |
| anteaya | pleia2: yes I am happy too, nibalizer is doing a great job | 18:01 |
| *** jcoufal_ has joined #openstack-infra | 18:02 | |
| markmc | mordred, ah, they're merge commits | 18:03 |
| *** bhuvan_ has quit IRC | 18:03 | |
| openstackgerrit | ldbragst proposed a change to openstack-dev/hacking: Add hacking check for inline comments with spaces https://review.openstack.org/77950 | 18:03 |
| *** bhuvan has quit IRC | 18:04 | |
| *** jcoufal has quit IRC | 18:05 | |
| openstackgerrit | ldbragst proposed a change to openstack-dev/hacking: Add hacking check for inline comments with spaces https://review.openstack.org/77950 | 18:05 |
| *** vkozhukalov has joined #openstack-infra | 18:05 | |
| jeblair | isn't that a pep8 check? ^ | 18:05 |
| clarkb | jeblair: pretty sure it is | 18:06 |
| clarkb | I want ot say that has bitten me ewll before hacking became a thing | 18:06 |
| * clarkb looks | 18:06 | |
| fungi | i thought so... pep8 (the tool) complains at me for no space after the # and not having at least two spaces before if inline with code | 18:06 |
| Alex_Gaynor | dhellmann: what would it take to get a ceilometerclient release so taht the troverclassifers accurately reflect the supported versions? | 18:07 |
| clarkb | oh I see, pep8 makes you do 'code #foo' | 18:08 |
| *** morganfainberg_Z is now known as morganfainberg | 18:08 | |
| *** jgallard has quit IRC | 18:08 | |
| clarkb | that change would make you do 'code # foo' | 18:08 |
| clarkb | honestly ugh why do we care? and it isn't consistent due to TODOs and FIXMEs | 18:08 |
| fungi | oh, that. right | 18:08 |
| * clarkb leaves a review | 18:08 | |
| fungi | [beware of hobgoblins] | 18:08 |
| *** che-arne has joined #openstack-infra | 18:10 | |
| openstackgerrit | A change was merged to openstack-infra/storyboard: Fix session usage in db_api https://review.openstack.org/77885 | 18:10 |
| *** jamielennox|away is now known as jamielennox | 18:10 | |
| *** yamahata__ has joined #openstack-infra | 18:10 | |
| jeblair | tomhe: you jinxed me by saying this was fast. getting the test right is taking some time. ;) | 18:10 |
| tomhe | jeblair: my bad | 18:11 |
| *** dkliban_lunch is now known as dkliban | 18:11 | |
| openstackgerrit | Michael Krotscheck proposed a change to openstack-infra/storyboard: Update developer documentation https://review.openstack.org/74713 | 18:12 |
| *** Ryan_Lane has joined #openstack-infra | 18:13 | |
| *** freyes_ has joined #openstack-infra | 18:13 | |
| *** yamahata has quit IRC | 18:14 | |
| *** hogepodge_ has joined #openstack-infra | 18:17 | |
| *** andre__ has quit IRC | 18:18 | |
| *** hogepodge has quit IRC | 18:18 | |
| *** hogepodge_ is now known as hogepodge | 18:18 | |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/zuul: Fix ZUUL_COMMIT in certain conditions https://review.openstack.org/77953 | 18:18 |
| jeblair | tomhe: ^ there we go! | 18:19 |
| tomhe | jeblair: Thanks! | 18:19 |
| nibalizer | anteaya: :) | 18:20 |
| tomhe | jeblair: And we really appreciate the new merger architecture. It really improved the perfomance. | 18:20 |
| anteaya | nibalizer: we are very happy with your work, well done | 18:20 |
| *** andre__ has joined #openstack-infra | 18:20 | |
| jeblair | tomhe: np, thank you for finding this and providing a succinct test case! | 18:20 |
| anteaya | jeblair: is there going to be a patch for removing the check for apache repo, or have I missed it? I have been trying to keep an eye out for it | 18:21 |
| *** bhuvan___ has quit IRC | 18:21 | |
| *** bhuvan__ has quit IRC | 18:21 | |
| jeblair | anteaya: probably not today | 18:21 |
| anteaya | jeblair: k, thanks | 18:21 |
| *** nati_ueno has joined #openstack-infra | 18:22 | |
| tomhe | jeblair: Think I'll fetch that change right away :) | 18:22 |
| *** nati_ueno has quit IRC | 18:22 | |
| *** nati_ueno has joined #openstack-infra | 18:23 | |
| openstackgerrit | A change was merged to openstack-infra/config: Run PyPy jobs on py3k-precise single-use slaves https://review.openstack.org/77732 | 18:24 |
| fungi | i'll babysit that ^ | 18:24 |
| *** gokrokve has quit IRC | 18:25 | |
| *** gokrokve has joined #openstack-infra | 18:25 | |
| fungi | recent retesting says it should be completely non-impacting | 18:25 |
| *** homeless has quit IRC | 18:27 | |
| openstackgerrit | A change was merged to openstack-infra/devstack-gate: add some helpfulness to our log output https://review.openstack.org/76349 | 18:28 |
| *** sweston has joined #openstack-infra | 18:28 | |
| *** gokrokve has quit IRC | 18:29 | |
| *** sparkycollier has quit IRC | 18:29 | |
| *** bhuvan has joined #openstack-infra | 18:29 | |
| *** bhuvan_ has joined #openstack-infra | 18:29 | |
| *** jaypipes has joined #openstack-infra | 18:30 | |
| *** browne has joined #openstack-infra | 18:33 | |
| *** markwash has joined #openstack-infra | 18:35 | |
| browne | hi all, anyone know how to go about setting up gerritbot? In particular, I want it to monitor project openstack/oslo.vmware to write messages to #openstack-vmware | 18:35 |
| *** jaypipes has quit IRC | 18:36 | |
| fungi | browne: submit a change for review... add a openstack-vmware section in http://git.openstack.org/cgit/openstack-infra/config/tree/modules/gerritbot/files/gerritbot_channel_config.yaml | 18:37 |
| clarkb | browne: edit https://git.openstack.org/cgit/openstack-infra/config/tree/modules/gerritbot/files/gerritbot_channel_config.yaml to include the channel and project See http://ci.openstack.org/irc.html note there are channel ownership requirements too | 18:37 |
| *** e0ne has joined #openstack-infra | 18:37 | |
| anteaya | clarkb: not sure if you saw this: http://lists.openstack.org/pipermail/openstack-infra/2014-March/000954.html chuck wants a ci account | 18:38 |
| anteaya | clarkb: just mentioning it since he said he will be away next week | 18:38 |
| clarkb | anteaya: I did, I think we have been batching those up periodically though. | 18:38 |
| browne | fungi: thanks! | 18:38 |
| browne | clarkb: thanks! | 18:38 |
| anteaya | clarkb: k | 18:38 |
| fungi | browne: and /msg chanserv access #openstack-vmware add openstackinfra +AFRfiorstv | 18:39 |
| fungi | browne: or if you're not a channel founder, then whoever is should add that | 18:39 |
| browne | fungi: ok, i'm not the owner, but will find him/her | 18:39 |
| *** e0ne has quit IRC | 18:39 | |
| openstackgerrit | A change was merged to openstack-infra/storyboard: Update developer documentation https://review.openstack.org/74713 | 18:41 |
| *** andre__ has quit IRC | 18:42 | |
| *** khyati has joined #openstack-infra | 18:43 | |
| *** homeless has joined #openstack-infra | 18:44 | |
| openstackgerrit | A change was merged to openstack-infra/config: Remove old elasticsearch cluster members https://review.openstack.org/76051 | 18:45 |
| fungi | clarkb: jeblair: SergeyLukjanov: mordred: git02 replacement is ready to go into service, if anybody wants to vote for https://review.openstack.org/77915 | 18:46 |
| mordred | looking | 18:46 |
| fungi | after that, it leaves just git01, which i'll knock out after the team meeting | 18:47 |
| *** mrmartin has joined #openstack-infra | 18:47 | |
| fungi | thanks! | 18:47 |
| sdague | it's a cool thing when the gate isn't totally exploding, and we merge 13 in row | 18:47 |
| fungi | this has been going very smoothly. hooray for high-availability | 18:48 |
| SergeyLukjanov | fungi, cool | 18:48 |
| jeblair | sdague: we identified a source of false resets in zuul earlier; we probably won't patch it out until after i3 though. | 18:48 |
| sdague | jeblair: cool | 18:48 |
| sdague | what was it? | 18:48 |
| *** andre__ has joined #openstack-infra | 18:49 | |
| *** jlibosva has quit IRC | 18:49 | |
| jeblair | sdague: zuul waits for changes to propogate to a gerrit replica before confirming that a change has merged. that can sometimes time out. it's probably not necessary anymore because of other improvements | 18:49 |
| *** gokrokve has joined #openstack-infra | 18:50 | |
| openstackgerrit | A change was merged to openstack-infra/config: Move git02.openstack.org to a new server https://review.openstack.org/77915 | 18:50 |
| jeblair | it has happened 3 times UTC-today | 18:50 |
| *** changbl has quit IRC | 18:50 | |
| jeblair | and 8 times yesterday | 18:51 |
| *** johnthetubaguy has quit IRC | 18:51 | |
| openstackgerrit | Morgan Fainberg proposed a change to openstack-infra/config: Add in scaffolding for sample_config update periodic task https://review.openstack.org/77789 | 18:52 |
| devananda | how would ya'll feel about a temporary repo for housing the nova "ironic" driver, so we can continue development pace on it during the nova FF | 18:52 |
| fungi | jeblair: sdague: it's the sort of subtle issue which we'd have been unlikely to spot without a high volume of changes for very large projects passing through the gate and merging, which i find an incredible turn of events on its own | 18:52 |
| openstackgerrit | Morgan Fainberg proposed a change to openstack-infra/config: Enable periodic sample config job for Keystone https://review.openstack.org/77790 | 18:52 |
| sdague | jeblair: cool, interesting | 18:52 |
| sdague | glad it was found | 18:52 |
| jeblair | fungi: found it, yay! :) | 18:52 |
| clarkb | fungi ever attentive | 18:53 |
| jeblair | devananda: would a feature branch of the nova project work? | 18:53 |
| clarkb | devananda: ther eis no such thing as a temporary repo. | 18:53 |
| * fungi got very lucky and happened to look over/remember some random numbers at just the right time | 18:53 | |
| morganfainberg | SergeyLukjanov, corrected that config option to use "proposal" instead of bare-precise | 18:53 |
| *** jaypipes has joined #openstack-infra | 18:53 | |
| sdague | fungi: awesome | 18:53 |
| devananda | jeblair: the driver is self-contained at this point, we dont need to branch all of nova | 18:53 |
| devananda | clarkb: then github? | 18:54 |
| jeblair | devananda: where is it now? | 18:54 |
| *** changbl has joined #openstack-infra | 18:54 | |
| devananda | jeblair: https://review.openstack.org/#/c/71429/9 and https://review.openstack.org/#/c/51328/29 | 18:54 |
| clarkb | devananda: just pointing out that we don't delete repositories. They are basically permanent | 18:54 |
| jeblair | devananda: so it's in nova? | 18:54 |
| SergeyLukjanov | morganfainberg, thx, I'll take a look on it soon | 18:54 |
| *** lcestari_ has quit IRC | 18:54 | |
| jeblair | devananda: if it's currently in nova, why not a feature branch on nova then? | 18:54 |
| morganfainberg | SergeyLukjanov thnx! | 18:54 |
| devananda | jeblair: maybe i'm missing something. how does that help? | 18:55 |
| openstackgerrit | A change was merged to openstack-infra/storyboard: Clean up doc, private func, removed unused file https://review.openstack.org/77179 | 18:55 |
| fungi | devananda: nova cores could approve changes proposed to that feature branch while nova master is frozen | 18:55 |
| anteaya | 51328 has a comment that this should be developed out of tree | 18:56 |
| *** thedodd has quit IRC | 18:56 | |
| fungi | devananda: and then that feature branch could get a merge commit review after the thaw | 18:56 |
| devananda | fungi: at this point, i'm not expecting nova to pay attention to it for a while | 18:56 |
| anteaya | would a nova branch be considered out of tree | 18:56 |
| devananda | maybe that's cynical, but I would like the ironic team to be able to iterate quickly | 18:56 |
| devananda | and the tripleo team to be able to use it | 18:56 |
| devananda | ASAP | 18:56 |
| devananda | rather than wait for nova-core to review it | 18:57 |
| jeblair | devananda: is this code going into nova? | 18:57 |
| *** e0ne has joined #openstack-infra | 18:57 | |
| devananda | which they have said won't happen until there is CI -- and we are having a lot of trouble testing any CI system with multiple patches in-flight in nova | 18:57 |
| devananda | jeblair: that's the goal | 18:57 |
| mrmartin | fungi: re, if you have some time, could you check this patch: https://review.openstack.org/#/c/77621/ we need this for close the openstackid dev project | 18:57 |
| fungi | mrmartin: so that supplants part of https://review.openstack.org/76407 ? | 18:58 |
| mordred | devananda: I think the chicken and egg is that to develop the CI system that's needed, you need the patches to nova, but you can't land the patches in nova until you have CI for them, yeah? | 18:59 |
| devananda | right | 18:59 |
| devananda | AND | 18:59 |
| mordred | how important is it for developing the ci | 18:59 |
| devananda | tripleo wants to use our driver | 18:59 |
| mrmartin | fungi: yes, I made a new patch, because some modification required in openstack_project to accept ssl cert content | 18:59 |
| fungi | to make an apple pie from scratch, you must first invent the universe^Winfrastructure | 18:59 |
| mordred | that other patches to nova continue to land? | 19:00 |
| fungi | meeting time? | 19:00 |
| mordred | and/or - is is possible to put the driver in teh ironic tree? | 19:00 |
| jeblair | meeting time! | 19:00 |
| jeblair | devananda: can we pick this up later? | 19:00 |
| devananda | jeblair: sure | 19:00 |
| fungi | oh, keystone's running over | 19:00 |
| *** thedodd has joined #openstack-infra | 19:00 | |
| mordred | I believe that nova virt driver config can reference random python paths etc | 19:00 |
| devananda | mordred: that's an interesting idea | 19:00 |
| mordred | devananda: it won't protect you from virt driver interface issues | 19:01 |
| mordred | but damn if that's regularly killing you there is something _Else_ That's grossly problematic :) | 19:01 |
| *** andre__ has quit IRC | 19:03 | |
| devananda | mordred: no. that sounds like it will work. thanks | 19:03 |
| mordred | devananda: cool. | 19:04 |
| openstackgerrit | A change was merged to openstack-infra/gerrit: replace submodule url reference from relative to absolute https://review.openstack.org/73125 | 19:05 |
| *** zns has joined #openstack-infra | 19:06 | |
| *** luis_ has joined #openstack-infra | 19:06 | |
| *** pblaho has joined #openstack-infra | 19:09 | |
| *** jooools has quit IRC | 19:09 | |
| *** pblaho has quit IRC | 19:09 | |
| *** pblaho has joined #openstack-infra | 19:09 | |
| dhellmann | Alex_Gaynor: we usually talk about doing a release at our weekly meeting, and I think eglynn typically manages them | 19:10 |
| *** luisg has quit IRC | 19:10 | |
| *** e0ne has quit IRC | 19:16 | |
| *** e0ne has joined #openstack-infra | 19:16 | |
| *** gokrokve_ has joined #openstack-infra | 19:17 | |
| *** gokrokve has quit IRC | 19:21 | |
| *** andreaf has quit IRC | 19:21 | |
| sdague | man, we're currently basically limitted in merge solely by the number of nodes in nodepool | 19:22 |
| *** SumitNaiksatam has joined #openstack-infra | 19:23 | |
| *** sweston has quit IRC | 19:24 | |
| anteaya | nice | 19:24 |
| *** amcrn has quit IRC | 19:24 | |
| anteaya | 900 nodes, we popped up again, I missed the additions - last I heard we were around 750 | 19:26 |
| anteaya | thanks to whoever gave us 150 more | 19:26 |
| fungi | rax raised our quota | 19:27 |
| sdague | cool. | 19:27 |
| fungi | weeks ago | 19:28 |
| clarkb | fungi: anteaya: https://review.openstack.org/#/c/71323/ should be included in this friday's manage-projects run. It simplifies our jeepyb configs quite a bit but I don't want it to merge without some testing (because taking a step back on the manage-projects front would be bad) | 19:28 |
| fungi | clarkb: logstash is keeping up swimmingly | 19:28 |
| clarkb | fungi: anteaya can you note ^ someplace appropriate? | 19:28 |
| anteaya | nice, thanks rax | 19:29 |
| sdague | fungi: so we seem to be narrowing down on running nodes with more in deleting and building. Any chance some of those are sticking? Just because allocating a new node in check looks like it's about 1hr 15m delay | 19:29 |
| anteaya | clarkb: can do | 19:29 |
| *** lcestari has quit IRC | 19:29 | |
| *** bada has joined #openstack-infra | 19:29 | |
| clarkb | anteaya: danke | 19:29 |
| clarkb | sdague: part of that is the gate is higher priority | 19:30 |
| sdague | clarkb: sure | 19:30 |
| krtaylor | anteaya, re: 900 nodes - is that cores? or? | 19:30 |
| clarkb | its 900 * 4 cores | 19:30 |
| anteaya | krtaylor: looking at the graph in zuul | 19:30 |
| sdague | clarkb: well rax are 8 cores | 19:31 |
| clarkb | sdague: oh right | 19:31 |
| sdague | so somewhere between 4k - 7k cores | 19:31 |
| clarkb | because performance | 19:31 |
| clarkb | or something | 19:31 |
| sdague | clarkb: yeh, I mostly wanted to make sure that a few builders weren't stalling, because I know we saw some of that in the past. And any free resource we have today would be goodness | 19:32 |
| fungi | sdague: a lot of those showing building are the 190 in tripleo cloud which keep failing and getting retried because it's down | 19:33 |
| fungi | and the deleting state nodes are just because our providers don't respond reliably or quickly to nova delete calls | 19:33 |
| sdague | ok | 19:33 |
| sdague | no problem then | 19:33 |
| fungi | but i'll also check whether we're hitting the 5000 nodes built a day limit on any of the rackspace regions in a bit | 19:33 |
| sdague | yeh, because we're flat out | 19:34 |
| *** dizquierdo has quit IRC | 19:35 | |
| anteaya | guess I didn't notce the increase in quota since it has been a while since we had to use all of them at once | 19:35 |
| *** vkozhukalov has quit IRC | 19:36 | |
| *** jcoufal_ has quit IRC | 19:37 | |
| *** yolanda_ has quit IRC | 19:37 | |
| *** pblaho has quit IRC | 19:38 | |
| *** bada has quit IRC | 19:39 | |
| *** ildikov_ has quit IRC | 19:39 | |
| *** lcestari has joined #openstack-infra | 19:42 | |
| *** yolanda_ has joined #openstack-infra | 19:42 | |
| clarkb | blast 71323 has merge conflicts | 19:44 |
| *** Ryan_Lane has quit IRC | 19:44 | |
| clarkb | Iwill swing around on that beforefriday I hope | 19:44 |
| *** ryanpetrello has joined #openstack-infra | 19:44 | |
| *** Ryan_Lane has joined #openstack-infra | 19:44 | |
| *** thomasem_ has joined #openstack-infra | 19:45 | |
| *** Ryan_Lane has quit IRC | 19:45 | |
| *** thomasem has quit IRC | 19:46 | |
| *** wchrisj has quit IRC | 19:47 | |
| *** arborism has joined #openstack-infra | 19:47 | |
| *** arborism is now known as amcrn | 19:47 | |
| *** yolanda_ has quit IRC | 19:48 | |
| *** wchrisj has joined #openstack-infra | 19:49 | |
| openstackgerrit | Alessandro Pilotti proposed a change to openstack/requirements: Sets the min python-novaclient version to 2.16.0 https://review.openstack.org/77981 | 19:50 |
| *** lcheng_ has quit IRC | 19:52 | |
| *** freyes_ has quit IRC | 19:55 | |
| *** cadenzajon has quit IRC | 19:57 | |
| SpamapS | fungi: we _might_ have resurrected tripleo ci | 19:59 |
| SpamapS | fungi: I see new nodes spawning.. | 19:59 |
| *** unicell has quit IRC | 20:00 | |
| clarkb | SpamapS: you got those ethernet ports sorted? | 20:00 |
| pleia2 | any thoughts on bugday next week? :) | 20:00 |
| anteaya | so https://etherpad.openstack.org/p/new-projects-2014-03-07 | 20:00 |
| fungi | SpamapS: cheers! sounds like you need a beer | 20:00 |
| clarkb | pleia2: I think we should do one | 20:00 |
| *** Sukhdev has joined #openstack-infra | 20:00 | |
| anteaya | is the next rounds, was waiting for open discussion to post and missed it | 20:00 |
| SpamapS | fungi: several | 20:01 |
| SpamapS | clarkb: yeah.. ugh | 20:01 |
| openstackgerrit | Michael Krotscheck proposed a change to openstack-infra/storyboard: Remove Branch and Milestone legacy tables https://review.openstack.org/77187 | 20:01 |
| *** denis_makogon_ has joined #openstack-infra | 20:01 | |
| nibalizer | kk thanks yall | 20:01 |
| SergeyLukjanov | jeblair, fungi, clarkb, mordred, re savanna renaming - we're still waiting for reply from the foundation lawyers, after that we'll have 2 business days long final voting and we'll be ready to rename the world | 20:02 |
| anteaya | pleia2: I won't be here for a bug day next week | 20:02 |
| anteaya | but that shouldn't hold you back | 20:02 |
| mordred | SergeyLukjanov: kk | 20:02 |
| *** gyee has quit IRC | 20:03 | |
| anteaya | SergeyLukjanov: what are is the link to the current options? | 20:04 |
| *** VijayTripathi has quit IRC | 20:04 | |
| anteaya | s/what/where | 20:04 |
| SergeyLukjanov | anteaya, let me find ;) | 20:04 |
| clarkb | anteaya: can you address https://review.openstack.org/#/c/74130/1 really quickly? | 20:04 |
| *** Ryan_Lane has joined #openstack-infra | 20:04 | |
| SergeyLukjanov | anteaya, http://civs.cs.cornell.edu/cgi-bin/results.pl?id=E_5dd4f18fde38ce8e, so, we've decided to chose the name from the top 5 options | 20:05 |
| anteaya | SergeyLukjanov: thanks | 20:05 |
| SergeyLukjanov | anteaya, and now waiting for their review | 20:05 |
| anteaya | SergeyLukjanov: about your comment on 74130 | 20:05 |
| anteaya | the reason I had selected one item to log is that was fungi | 20:06 |
| anteaya | s request | 20:06 |
| *** rlandy has quit IRC | 20:06 | |
| anteaya | since all three items would be too much output | 20:06 |
| anteaya | if you want me to change to return code, I have no issue with that, and can do so | 20:06 |
| anteaya | but I was requested to just change the logging status of one item | 20:07 |
| anteaya | I don't care which it is | 20:07 |
| SergeyLukjanov | anteaya, it's quite strange for me to log command that we're running but not logging return code, I see only pros of doing it, not cons, anyway, I'm ok with merging as is and think about further improvements later | 20:07 |
| anteaya | and return code is captured in debug level | 20:07 |
| fungi | anteaya: well, i don't think i asked for that specifically. i said we should analyze which things we log at debug are low-enough volume they could be switched to info. if memory serves you asked if you could start with just one log call as an example and i said sure | 20:07 |
| *** esker has joined #openstack-infra | 20:07 | |
| anteaya | okay | 20:07 |
| anteaya | so fungi how do you feel about return codes getting logged at info, in addition to commands? | 20:08 |
| fungi | mainly because i didn'tdon't have time to go over the debug logs from it at the moment to decide what should be what (and that's why you were setting up your own gerrit to test against) | 20:08 |
| anteaya | I will do it if you think it is okay | 20:08 |
| clarkb | mordred: can you address https://review.openstack.org/#/c/77311/2 ? | 20:08 |
| anteaya | I can set up gerrit but I can't get manage projects configured to run | 20:08 |
| clarkb | anteaya: I think they should be paired | 20:08 |
| fungi | anteaya: seems fine to me. i don't really know how often that does'doesn't get logged | 20:09 |
| *** denis_makogon_ is now known as denis_makogon | 20:09 | |
| clarkb | the return code and command output go together | 20:09 |
| anteaya | I will make the change | 20:09 |
| SergeyLukjanov | anteaya, thx | 20:09 |
| SergeyLukjanov | clarkb, ++ | 20:09 |
| fungi | i figured 74130 was just a test example of using the new -v vs -d functionality, which is why i hadn't +2'd it | 20:10 |
| fungi | to provide some means of confirming we got info level logged with -v | 20:10 |
| fungi | and was a placeholder for a change which actually set reasonable log levels on all those debug calls | 20:11 |
| mordred | clarkb: yah | 20:11 |
| clarkb | fungi: I think I am mostly caught up on jeepyb reviews. I approved the one that looked to have your seal of approval | 20:13 |
| fungi | clarkb: awesome--hopefully next friday will be equally enlightening in that regard | 20:13 |
| openstackgerrit | Monty Taylor proposed a change to openstack-infra/jeepyb: Support resetting back to a non-master branch https://review.openstack.org/77311 | 20:14 |
| openstackgerrit | Anita Kuno proposed a change to openstack-infra/jeepyb: Change return code and command said log.info https://review.openstack.org/74130 | 20:14 |
| *** jcoufal has joined #openstack-infra | 20:14 | |
| anteaya | personally I would hold off on approving 74130 until we see how it performs, as I said I have not figured out how to configure my gerrit to run manage-projects | 20:15 |
| *** ryanpetrello has quit IRC | 20:24 | |
| *** lcestari has quit IRC | 20:25 | |
| mordred | clarkb: there you go | 20:25 |
| *** hashar has joined #openstack-infra | 20:29 | |
| *** Ryan_Lane has quit IRC | 20:31 | |
| *** markwash has quit IRC | 20:32 | |
| fungi | openstackgerrit: is struggling suddenly... suspecting another netsplit | 20:34 |
| openstackgerrit | Jeremy Stanley proposed a change to openstack-infra/config: Move git01.openstack.org to a new server https://review.openstack.org/77994 | 20:35 |
| openstackgerrit | A change was merged to openstack-infra/jeepyb: Only add upstream remotes if we track upstream https://review.openstack.org/77298 | 20:35 |
| fungi | or it's just slow | 20:36 |
| openstackgerrit | Jeremy Stanley proposed a change to openstack-infra/config: Add storyboard SSL cert/key/chain https://review.openstack.org/76407 | 20:41 |
| *** sparkycollier has joined #openstack-infra | 20:42 | |
| *** jcooley_ has quit IRC | 20:43 | |
| *** jaypipes has quit IRC | 20:44 | |
| *** ociuhandu has quit IRC | 20:45 | |
| *** markwash has joined #openstack-infra | 20:45 | |
| *** sabari_ has joined #openstack-infra | 20:46 | |
| *** sabari__ has joined #openstack-infra | 20:46 | |
| *** jnoller has quit IRC | 20:47 | |
| *** esker has quit IRC | 20:48 | |
| anteaya | what changed in nova pep8? | 20:49 |
| anteaya | 3 nova patches failing in the gate on pep8 | 20:49 |
| *** sabari_ has quit IRC | 20:50 | |
| anteaya | check_uptodate.sh: nova.conf.sample is not up to date. | 20:50 |
| *** cody-somerville has quit IRC | 20:50 | |
| fungi | anteaya: yep, that's fairly common | 20:50 |
| anteaya | really? | 20:50 |
| anteaya | seems odd to have hit three in a row | 20:50 |
| fungi | something in a library being imported by nova just added new options and now all nova changes will fail until a change is merged to update the sample config | 20:51 |
| anteaya | russellb: ^^ | 20:51 |
| fungi | something added kombu_reconnect_delay | 20:51 |
| *** rfolco has quit IRC | 20:53 | |
| *** Ryan_Lane has joined #openstack-infra | 20:55 | |
| anteaya | so far I am not seeing the patch that did so | 20:55 |
| *** smurugesan has joined #openstack-infra | 20:55 | |
| SergeyLukjanov | anteaya, fungi, here is the patch https://github.com/openstack/oslo.messaging/commit/fcd51a67d18a9e947ae5f57eafa43ac756d1a5a8 | 20:56 |
| SergeyLukjanov | released as part of 1.3.0a9 | 20:56 |
| SergeyLukjanov | it looks like it's a bit incorrect approach with check_autoupdate - it could break any consumers of some projects | 20:57 |
| anteaya | ah I was looking in -nova | 20:57 |
| anteaya | dhellmann: ^^ | 20:57 |
| fungi | anteaya: if it were in nova, in theory it would never have been able to merge | 20:57 |
| *** mrda_away is now known as mrda | 20:57 | |
| *** jcoufal has quit IRC | 20:58 | |
| dhellmann | anteaya, fungi : is this the oslo.messaging release? | 20:58 |
| fungi | dhellmann: looks like it, yes | 20:58 |
| anteaya | fungi: because nova uses global requirements? | 20:58 |
| dhellmann | I thought we landed patches in nova to fix that | 20:58 |
| *** sabari__ has quit IRC | 20:58 | |
| dhellmann | or a patch, anyway | 20:58 |
| fungi | nova now wants its sample config updated | 20:58 |
| dhellmann | oh, ok | 20:58 |
| dhellmann | someone just needs to do that in nova | 20:58 |
| fungi | yep | 20:58 |
| *** sparkycollier has quit IRC | 20:58 | |
| dhellmann | and then turn off the check and build that file during packaging so we're not maintaining it by hand any more | 20:58 |
| *** smurugesan has quit IRC | 20:59 | |
| fungi | but as we've seen in the past, now all nova changes will fail (including resetting gating) until they stop checking that in their pep8 jobs | 20:59 |
| anteaya | so kombu is in global-requirements: http://git.openstack.org/cgit/openstack/requirements/tree/global-requirements.txt#n38 | 20:59 |
| *** smurugesan has joined #openstack-infra | 20:59 | |
| *** smurugesan has quit IRC | 20:59 | |
| fungi | maybe this will spur them on to start building those | 20:59 |
| *** sabari2 has joined #openstack-infra | 21:01 | |
| *** bhuvan has quit IRC | 21:01 | |
| *** sabari2 is now known as sabari_ | 21:01 | |
| *** wenlock has quit IRC | 21:01 | |
| mordred | jeblair: in teh board meeting, we've decided to rewrite openstack in go | 21:01 |
| *** bhuvan_ has quit IRC | 21:01 | |
| fungi | i heard that | 21:01 |
| geekinutah | mordred: finally | 21:01 |
| *** wenlock has joined #openstack-infra | 21:01 | |
| fungi | pretty amazing. the board members are going to do all the real work on that too | 21:01 |
| SergeyLukjanov | fungi, anteaya, dhellmann, here is a patch for periodic config update proposal job - https://review.openstack.org/#/c/77789/ by morganfainberg | 21:01 |
| dstufft | go get is the worst | 21:01 |
| dstufft | just fyi | 21:01 |
| fungi | dstufft: even the name... eww | 21:02 |
| jeblair | dansmith: feb 22 was the day freenode was completely ususable | 21:02 |
| annegentle | jeblair: your risk analysis may be right on, by the way, probably is. I've just seen some reactive behavior in non-profits I've worked with that makes me ask more questions | 21:02 |
| dhellmann | SergeyLukjanov: the file should be build when nova packages are built, and never checked into git | 21:02 |
| annegentle | jeblair: re: oft | 21:02 |
| annegentle | oftc | 21:02 |
| *** sabari_ is now known as sabari | 21:02 | |
| jeblair | annegentle: i definitely understand the concern; i share it. i don't want to have to move back if we move! :) | 21:02 |
| dansmith | jeblair: okay, I just get a beep from my proxy every time it tries to reconnect and I expect I would have remembered that storm | 21:02 |
| dansmith | jeblair: but fair enough | 21:02 |
| *** bhuvan_ has joined #openstack-infra | 21:03 | |
| SergeyLukjanov | dhellmann, yup, I agree that it's the best case, IIRC there were some concerns about how to impl it, am I right? | 21:03 |
| *** bhuvan__ has joined #openstack-infra | 21:03 | |
| jeblair | dansmith: on that day i ended up on a netsplit server; i was never disconnected from it, but it wasn't connected to the hub | 21:03 |
| *** jcoufal has joined #openstack-infra | 21:03 | |
| wenlock | jeblair, on irc, i didn't understand the answer to why openstack doesn't run a private irc for openstack.org, why is that? | 21:03 |
| jeblair | dansmith: so i tried reconnecting and found all of the servers in rotation were unresponsive | 21:03 |
| *** sandywalsh_ has quit IRC | 21:03 | |
| dansmith | jeblair: okay, well, that's what I mean by "not down" but that's fine | 21:03 |
| dhellmann | SergeyLukjanov: no idea :-/ | 21:03 |
| morganfainberg | dhellmann, i disagree, when we are generating documents or if we are using a milestone | 21:03 |
| dstufft | wenlock: because running your own irc network is the worst | 21:04 |
| openstackgerrit | A change was merged to openstack-infra/config: Add OpenstackID SSL Certs https://review.openstack.org/77621 | 21:04 |
| dansmith | jeblair: by examining the A records, or by just reconnecting a bunch of times? because they do insert bad records on purpose for DoS avoidance I think | 21:04 |
| morganfainberg | dhellmann, or even master, we want something that is actually in the tree | 21:04 |
| jeblair | dansmith: i tried all of the A records | 21:04 |
| fungi | dansmith: yeah, i tried every server in the round-robin, all were either unresponsive, had nobody or only 1 other nick in channel, except for dixon which kept refusing my connection because it had too many connections already. when i finally managed to squeeze onto dixon that's where everyone else was | 21:04 |
| dansmith | jeblair: okay, well, my real concern is whether OFTC is better or smaller, that's all | 21:04 |
| dansmith | jeblair: because, I've talked to freenode admins before and .. you know. | 21:04 |
| dstufft | afaik OFTC is just less of a target | 21:05 |
| jeblair | dansmith: i don't know -- what did you learn talking to them? you can pm me if you want | 21:05 |
| fungi | and doesn't have the legacy/stigma of opn | 21:05 |
| morganfainberg | dhellmann, i think building only at packaging time is, while a good approach, leaving a gap in coverage. | 21:05 |
| dansmith | jeblair: no, I mean.. I've talked to them before about things and don't ever want to have to do that again.. evar :) | 21:05 |
| dhellmann | morganfainberg: if those other jobs need the file, they can build it too, my main point was to remove it from git and stop verifying that we're keeping that version up to date | 21:06 |
| dansmith | jeblair: in that, OFTC has a leg up from a "dealing with the admins" PoV :) | 21:06 |
| jeblair | dstufft, dansmith: it's definitely smaller, and i think it's less of a target. even if they get hit by ddos less for only those reasons, as opposed to something more intrinsic, that may not still be a bad thing. | 21:06 |
| *** cadenzajon has joined #openstack-infra | 21:06 | |
| dansmith | jeblair: it's just a question of whether it's worth the pain or not | 21:06 |
| fungi | dansmith: you talked to them about things and they actually got back to you (positive or negative answers aside)? | 21:06 |
| dansmith | jeblair: because if amazon is DoSing freenode from AWS to mess with us, then OFTC will crumble | 21:06 |
| morganfainberg | dhellmann, by that standpoint, translations should be sideband as well? | 21:06 |
| *** bhuvan___ has joined #openstack-infra | 21:06 | |
| dansmith | :P | 21:06 |
| dhellmann | morganfainberg: translations can't be generated from some other file, though, they are created by people | 21:07 |
| dansmith | fungi: yeah, they responded with mocking and insults, IIRC :) | 21:07 |
| jeblair | dansmith: ah, yeah, that's been great. we're years into trying to get some freenode things sorted. oftc admins have been very responsive | 21:07 |
| dstufft | if amazon is DoSing from AWS then there's no IRC network that will handle it ;P | 21:07 |
| dhellmann | morganfainberg: the sample config is based entirely on other artifacts we have under source control | 21:07 |
| fungi | dstufft: we'll put one on internet2 | 21:07 |
| dansmith | dstufft: well, I'm joking of course, but that's the point | 21:07 |
| *** bhuvan has joined #openstack-infra | 21:08 | |
| *** sabari has quit IRC | 21:08 | |
| dstufft | you see | 21:08 |
| dstufft | what freenode needs is just one big IRC server | 21:08 |
| *** sabari has joined #openstack-infra | 21:08 | |
| dstufft | then there's no more netsplits | 21:08 |
| jeblair | dstufft: it should be in java | 21:09 |
| jeblair | i hear it's good for really big servers | 21:09 |
| fungi | jeblair: after we rewrite java in go | 21:09 |
| *** bhuvan__ has quit IRC | 21:09 | |
| *** sparkycollier has joined #openstack-infra | 21:09 | |
| morganfainberg | dhellmann, personally i think it should be in the repo, but thats based on trying to be friendly to our users, they can look at git and see the "close" to state-of-the-art rather than relying on packaging or having to checkout and build. but if the standpoint is this should never be in git i don't have a good answer - packaging is not handled by us (we make an effort to throw packaging over the wall to the distros) | 21:09 |
| dstufft | write an IRC server in Twisted and make everyone sign in with their gerrit username! | 21:09 |
| *** bhuvan_ has quit IRC | 21:09 | |
| jeblair | does anyone here know any oftc ops? | 21:09 |
| morganfainberg | dhellmann, i'll adhere to it, but i don't agree. | 21:09 |
| jeblair | i'd like to abuse a personal relationship to ask them how much more awesome than freenode they are, if one exists. :) | 21:10 |
| dansmith | jeblair: hah | 21:10 |
| dhellmann | morganfainberg: if we keep it in the tree, we'll just have to keep doing this update dance, so as long as no one complains about that any more I'm ok with keeping it. :-) | 21:10 |
| morganfainberg | jeblair, hehe | 21:10 |
| morganfainberg | dhellmann, if we can move it to a periodic task i think we can avoid a lot of complaining. and i expect the pep8 check to go away completely (people are used to translations and requirements updates) :) | 21:11 |
| *** ociuhandu has joined #openstack-infra | 21:12 | |
| dhellmann | morganfainberg: I can go along with that, then. | 21:12 |
| morganfainberg | dhellmann, once i have this working for keystone i'll def work on the other projects to get them working the same way | 21:12 |
| *** bhuvan_ has joined #openstack-infra | 21:12 | |
| dhellmann | morganfainberg: you're doing it as a job in the infra repo? | 21:13 |
| *** bhuvan__ has joined #openstack-infra | 21:13 | |
| morganfainberg | dhellmann, yep | 21:13 |
| morganfainberg | dhellmann, same as how translations would be triggered, periodic task | 21:13 |
| morganfainberg | dhellmann, https://review.openstack.org/#/c/77789/ | 21:13 |
| dhellmann | morganfainberg: I'll take a look after the release meeting, then, cool | 21:13 |
| morganfainberg | dhellmann, and the keystone "turn on" would be https://review.openstack.org/#/c/77790/ | 21:13 |
| morganfainberg | dhellmann, sounds good | 21:14 |
| *** bhuvan has quit IRC | 21:14 | |
| *** bhuvan___ has quit IRC | 21:14 | |
| *** bada has joined #openstack-infra | 21:17 | |
| mordred | dhellmann, morganfainberg: reading scrollback | 21:18 |
| *** bhuvan___ has joined #openstack-infra | 21:18 | |
| *** bhuvan__ has quit IRC | 21:19 | |
| morganfainberg | mordred, on that topic, similarly if this works i can do man pages (at least keystone updates man pages at infrequent intervals) | 21:19 |
| mordred | morganfainberg: sorry - my brain is poorly processing this | 21:19 |
| dansmith | jeblair: classic response | 21:19 |
| mordred | morganfainberg: we're talking about generating config samples? | 21:19 |
| dansmith | jeblair: "we ain't special" | 21:19 |
| morganfainberg | mordred, no worries, periodic task to propose sample config changes | 21:19 |
| morganfainberg | mordred, yes | 21:19 |
| *** bhuvan_ has quit IRC | 21:20 | |
| *** flaper87 is now known as flaper87|afk | 21:20 | |
| morganfainberg | mordred, rather than having a pep8 check and needing to make sure everything is in sync for every review | 21:20 |
| morganfainberg | mordred, this is a bigger issue if we import opts from a library like oslo.messaging. an update to oslo.messaging would break all pep8 checks until someone fixed the sample.conf | 21:21 |
| clarkb | morganfainberg: there is one more thing that pep8 caught on that change | 21:21 |
| clarkb | mordred: ^ tab complete is hard | 21:21 |
| dhellmann | morganfainberg: expect the same problem with oslo.db | 21:21 |
| morganfainberg | dhellmann, ++ | 21:21 |
| dhellmann | and possibly other of the 20+ new libraries we'll be making during juno | 21:21 |
| morganfainberg | clarkb, did I miss a tab? | 21:21 |
| markmc | has anyone started working on making the config file generation install time? | 21:21 |
| clarkb | morganfainberg: you are fine, I fail | 21:22 |
| *** esker has joined #openstack-infra | 21:22 | |
| clarkb | morganfainberg: sorry for the disturbance | 21:22 |
| morganfainberg | clarkb, ah okie :P | 21:22 |
| *** esker has quit IRC | 21:22 | |
| morganfainberg | clarkb, hehe :) | 21:22 |
| *** esker has joined #openstack-infra | 21:23 | |
| *** bhuvan has joined #openstack-infra | 21:23 | |
| *** bhuvan_ has joined #openstack-infra | 21:23 | |
| *** jcooley_ has joined #openstack-infra | 21:23 | |
| morganfainberg | markmc, if there is a real push to move it into setup and infra would rather have that, i'll circle back instead. but i think there will be wondering why we don't offer a sample config people can look at in our repos | 21:24 |
| *** bhuvan___ has quit IRC | 21:24 | |
| fungi | morganfainberg: we offer a script which can build them easily | 21:24 |
| markmc | morganfainberg, yeah, sorry - just catching up on scrollback | 21:24 |
| fungi | morganfainberg: and in tarballs built from teh tip of every branch and tag | 21:24 |
| markmc | morganfainberg, we can publish the generated file elsewhere, doesn't have to be git | 21:24 |
| fungi | including them in the doc builds like we do with api docs, might be worthwhile? | 21:25 |
| morganfainberg | markmc, sure, it currently is in git and the "we should put them else where" has been nebulous - so i figure i'd make a stab at this and see where we landed | 21:25 |
| morganfainberg | fungi, for sure it should be built then if we don't have it in the repos | 21:26 |
| *** sarob__ has joined #openstack-infra | 21:26 | |
| *** sarob__ has quit IRC | 21:26 | |
| *** esker has quit IRC | 21:27 | |
| *** briancurtin has joined #openstack-infra | 21:27 | |
| fungi | the main audience for pre-generated sample config files donesn't seem to me to be substantially overlapping the sorts of people who go poking in the git repository first thing | 21:27 |
| *** oubiwan__ has joined #openstack-infra | 21:27 | |
| morganfainberg | markmc, fungi, dhellmann, sounds to me like it should just be a doc job and/or release thing instead? i like it in-tree, but if the project wants something else i'll concede and change tacks on it. | 21:28 |
| *** bhuvan__ has joined #openstack-infra | 21:28 | |
| *** bhuvan___ has joined #openstack-infra | 21:28 | |
| morganfainberg | s/project/infra | 21:28 |
| morganfainberg | and the rest of the audience here | 21:28 |
| dhellmann | morganfainberg: I am +1 to moving it out of git and +0 to having it updated periodically without a pep8 test | 21:29 |
| *** bhuvan has quit IRC | 21:29 | |
| fungi | morganfainberg: well, the api docs aren't in-tree either... but we provide an easy way to generate them and we also publish those and keep them up to date automatically, so sample configs seem vmuch more like that than translations or requirements files | 21:29 |
| briancurtin | for python-openstacksdk, we have a change stuck in the gate: the change is to add newlines to requirements.txt files, but the gate check for requirements is currently failing because of it -- any ideas? https://review.openstack.org/#/c/75852/ | 21:29 |
| *** eharney has quit IRC | 21:29 | |
| morganfainberg | dhellmann, my #1 goal is no pep8 tests | 21:29 |
| *** bhuvan_ has quit IRC | 21:29 | |
| morganfainberg | dhellmann, esp. heading into juno | 21:29 |
| dhellmann | morganfainberg: I've got your back, then. | 21:30 |
| morganfainberg | dhellmann, the rest is preference but not strong enough to make a stronger stance than personal preference | 21:30 |
| clarkb | pleia2: thank your for testing https://review.openstack.org/#/c/77683/ | 21:30 |
| dhellmann | morganfainberg: yep | 21:30 |
| clarkb | fungi: jeblair mordred SergeyLukjanov ^ is a good one to review. It iwll fix our centos image builds | 21:30 |
| morganfainberg | dhellmann, but the easiest way to get traction one way or another is propose something vs. complaining about it :) | 21:31 |
| dhellmann | morganfainberg: +1000 | 21:31 |
| *** julim has quit IRC | 21:31 | |
| fungi | morganfainberg: communication through source code. absolutely more productive | 21:32 |
| markmc | morganfainberg, dhellmann, in the immediate, short-term - how about we kill the "check config file is up to date" from the pep8 job | 21:32 |
| morganfainberg | markmc, it's per-project tox.ini | 21:33 |
| dhellmann | markmc: good idea | 21:33 |
| morganfainberg | markmc, but easy to propose those changes | 21:33 |
| openstackgerrit | Khai Do proposed a change to openstack-infra/config: Add jenkins job to build gerrit plugins https://review.openstack.org/70014 | 21:33 |
| markmc | or, better ... | 21:34 |
| morganfainberg | i am going to gues that removing that will have some push-back if we don't have an alternative in place | 21:34 |
| markmc | print out warning spew if it's out of date, but don't fail pep8 | 21:34 |
| *** bhuvan___ has quit IRC | 21:34 | |
| *** bhuvan__ has quit IRC | 21:34 | |
| morganfainberg | markmc, actually a non-voting task would be good for that | 21:34 |
| fungi | warnspam seems like a fine compromise to me while the proper solution is in progress | 21:34 |
| morganfainberg | fungi, ^ | 21:34 |
| morganfainberg | something that goes in "check" queue? | 21:35 |
| morganfainberg | that way it's more visible than warning in logs | 21:35 |
| sdague | fungi: so there is no way to get a job extra priority in check right? | 21:35 |
| fungi | non-voting check-only job, sure, but i think that solving the publishing problem with the help of someone from docs (Ajeager?) would probably not be much more work | 21:35 |
| fungi | sdague: one specific job? i don't think so--right now we can only ptioritize pipelines i think | 21:36 |
| morganfainberg | fungi, super, i'll abandon the periodic and get that proposed today | 21:36 |
| fungi | prioritize | 21:36 |
| sdague | because all of nova is blocked again | 21:36 |
| openstackgerrit | Elizabeth Krumbach Joseph proposed a change to openstack-infra/config: Add support for Fedora 20 to nodepool https://review.openstack.org/69510 | 21:36 |
| fungi | sdague: oh, you mean a change, not a job? | 21:36 |
| sdague | yeh | 21:36 |
| sdague | https://review.openstack.org/#/c/78009/ | 21:36 |
| sdague | it would be really good to get pep8 early on that | 21:37 |
| fungi | sdague: if you're comfortable that the review is solid, i can enqueue it in the gate and promote it (it's two steps, but will have the effect you want) | 21:37 |
| sdague | I am, russellb how do you feel about that | 21:37 |
| fungi | sdague: did you try running tox -epep8 on it yourself? should be good enough since the gate will do that too | 21:37 |
| morganfainberg | fungi, and make sure it doesn't require a tox target etc, so we can remove the pep8 bits everywehre. and then we can do the doc bit(s) / install. cc markmc, dhellmann | 21:37 |
| sdague | yes, I did | 21:37 |
| fungi | sdague: worst case, it will cause a gate reset. not as many as it's likely to prevent | 21:38 |
| sdague | yeh, not with the nova parts in there | 21:38 |
| fungi | i mean, it's literally comment lines in a sample configuration file | 21:39 |
| russellb | sdague: find with me | 21:39 |
| russellb | fine | 21:39 |
| sdague | russellb: cool, want to put your +A on https://review.openstack.org/#/c/78009/ | 21:39 |
| sdague | then we'll do it | 21:39 |
| sdague | bingo | 21:39 |
| sdague | fungi: lets take it to the gate | 21:40 |
| *** hashar has quit IRC | 21:41 | |
| fungi | sdague: russellb: it's up front now | 21:41 |
| sdague | thanks fungi | 21:41 |
| russellb | great thanks | 21:41 |
| fungi | https://jenkins04.openstack.org/job/gate-nova-pep8/3482/console | 21:42 |
| fungi | in case you want earliest possible results | 21:42 |
| sdague | thanks | 21:42 |
| openstackgerrit | Davanum Srinivas (dims) proposed a change to openstack-infra/config: Add notifications to #openstack-oslo channel https://review.openstack.org/76598 | 21:43 |
| openstackgerrit | Davanum Srinivas (dims) proposed a change to openstack-infra/config: Add Eavesdrop bot to #openstack-oslo https://review.openstack.org/76584 | 21:43 |
| openstackgerrit | Davanum Srinivas (dims) proposed a change to openstack-infra/config: Support filtering by review id(s) https://review.openstack.org/72446 | 21:43 |
| jeblair | sdague, dansmith: http://paste.openstack.org/show/72319/ | 21:44 |
| jeblair | sdague, dansmith: i think the short version is: oftc is smaller and they are not the target of ddos attacks, and as a result they are probably not currently in a position to repond as well as freenode if they did | 21:44 |
| clarkb | dims: zul says cloud archive will have libvirt updtes this week, do you still have a change handy that we can kick which will test cloud archive for us? | 21:45 |
| dims | clarkb, yep. i can kick the tires | 21:46 |
| mordred | jeblair: do you think that they are in a position to receive help from us or our member companies? | 21:46 |
| mordred | jeblair: like, I hear rackspace and dreamhost have some bang-up network ops folks... | 21:46 |
| jeblair | mordred: i don't know; but search for 'hyperfilter' | 21:47 |
| sdague | jeblair: so where does that data fit into your current leaning? | 21:47 |
| clarkb | dims: great, I will try to pay attenttion to cloud archive and ping you when the new libvirt is available | 21:47 |
| *** jpich has quit IRC | 21:48 | |
| fungi | mordred: jeblair: i can imagine that if the openstack project considered its irc network a significant (even necessary tool) we might be able to find additional services and support for the non-profit organization providing it | 21:48 |
| sdague | fungi: pep8 success! | 21:49 |
| fungi | sdague: now just hope it doesn't die on some random unrelated error ;) | 21:49 |
| sdague | yeh, exactly | 21:49 |
| mordred | fungi: right. that's sort of what I Was thinking | 21:49 |
| *** dkliban has quit IRC | 21:49 | |
| dansmith | jeblair: yeah, seems about right | 21:50 |
| clarkb | pleia2: reviewed the nodepool script updates, good point about the copy pasta locations. A couple small formatting things inline | 21:50 |
| pleia2 | clarkb: thanks | 21:50 |
| *** jcoufal has quit IRC | 21:50 | |
| jeblair | mordred: tomaw might be fixing the freenode group situation right now; i don't want to distract him. but in a minute, i'll ask. :) | 21:51 |
| mordred | jeblair: awesome. I was just reading that | 21:51 |
| *** jcoufal has joined #openstack-infra | 21:52 | |
| fungi | yeah, that conversation seems consistent with what i've observed on oftc over the eyars | 21:52 |
| fungi | years | 21:52 |
| sdague | so I think regardless, the CNAME record is probably sensible | 21:52 |
| clarkb | sdague: thats actually one of the few bits that I think I disagree with | 21:52 |
| *** zns has quit IRC | 21:52 | |
| sdague | clarkb: ok | 21:52 |
| sdague | why? | 21:53 |
| mordred | clarkb: why? | 21:53 |
| clarkb | people should know what they are talking to, but I don't feel strongly enough about it to say no | 21:53 |
| openstackgerrit | A change was merged to openstack-infra/config: Fix update java alternatives on redhat systems https://review.openstack.org/77683 | 21:53 |
| dansmith | clarkb: I think you're right, FWIW | 21:53 |
| sdague | ok, that's fair | 21:53 |
| fungi | clarkb: i agree, but i think the people who care which network they're talking to aren't teh audience for the cname anyway | 21:53 |
| dansmith | clarkb: if I'm already on freenode, and think that irc.openstack.org is a different network, I'm confused | 21:53 |
| sdague | that's true, and I wouldn't use the cname anyway, because I have other freenode connections | 21:54 |
| clarkb | same here which is why I don't feel strongly about it | 21:54 |
| clarkb | and it may help others that don't grok IRC | 21:54 |
| morganfainberg | sdague, ++ i have a bunch setup because of the issues w/ freenode | 21:54 |
| morganfainberg | sdague, a lot of people have moved to using specific server(s) | 21:54 |
| openstackgerrit | Elizabeth Krumbach Joseph proposed a change to openstack-infra/config: Add support for Fedora 20 to nodepool https://review.openstack.org/69510 | 21:54 |
| sdague | morganfainberg: honestly, I think I'm still using the generic CNAME | 21:54 |
| fungi | right, i'll be on freenode long after any switch because i'm on non-openstack-related channels here... just like i'm on oftc already for non-openstack-related channels (and on another network too) | 21:54 |
| sdague | fungi: yeh, for me, I'm not on oftc, all the open source I've worked on the last decade has been on freenode | 21:55 |
| clarkb | pleia2: thanks, +2 | 21:55 |
| pleia2 | \o/ | 21:55 |
| sdague | but I registered my nick just in case | 21:56 |
| clarkb | fungi: same here which is why I don't see this as a painful switch :) | 21:56 |
| openstackgerrit | A change was merged to openstack-infra/config: Add storyboard SSL cert/key/chain https://review.openstack.org/76407 | 21:57 |
| clarkb | I have four server connection in my client with almost 100 buffers open ... whats one more :) | 21:57 |
| sdague | well for non wechat folks, we need to spin up another znc cluster | 21:57 |
| pleia2 | /w 122 | 21:57 |
| pleia2 | :) | 21:57 |
| sdague | which is just puppet | 21:57 |
| fungi | it's been very, very many years since i had to figure out how to coerce my irc client to talk to multiple servers/networks and even join the same channels on more than one network | 21:57 |
| *** jcooley_ has quit IRC | 21:57 | |
| sdague | but it's different | 21:57 |
| clarkb | sdague: you don't | 21:57 |
| clarkb | znc can do multiple servers | 21:57 |
| dansmith | clarkb: it's not entirely baked, last I checked | 21:58 |
| *** gyee has joined #openstack-infra | 21:58 | |
| sdague | I think that with the 2 layer znc setup that dansmith and I have, it might get wonky | 21:58 |
| clarkb | dansmith: iirc morganfainberg has it going now | 21:58 |
| dansmith | yeah | 21:58 |
| dansmith | sdague: right | 21:58 |
| fungi | even irssi will talk directly to multiple servers/networks and has for more than a decade | 21:58 |
| clarkb | 2 layer? | 21:58 |
| *** dcramer__ has quit IRC | 21:58 | |
| sdague | clarkb: yeh, a znc for zncs | 21:58 |
| mordred | we could always run zephyr | 21:58 |
| dansmith | clarkb: all cool people are znc-on-znc, thought you knew | 21:59 |
| jeblair | mordred, clarkb, fungi, pleia2: i am told that i am now the group contact for openstack-* ! | 21:59 |
| sdague | tripleZ | 21:59 |
| sdague | or not | 21:59 |
| mordred | or, zephyr not as good and re-written in xml - jabber | 21:59 |
| mordred | jeblair: WOOT | 21:59 |
| fungi | jeblair: congratulations on your promotion! | 21:59 |
| mordred | jeblair: who should I mail scotch to? | 21:59 |
| pleia2 | jeblair: hooray! | 21:59 |
| clarkb | dansmith: darn I should learn to bake znc cakes | 21:59 |
| sdague | clarkb: one znc to connect to freenode | 21:59 |
| clarkb | jeblair: noway | 21:59 |
| sdague | a bunch of zncs to talk to that | 21:59 |
| fungi | i think the last irc client i used which was moderately fiddly for connecting to more than one server at a time was bitchx | 21:59 |
| sdague | each for a different client | 21:59 |
| SlickNik | dansmith: What's the hot sell of znc-on-znc? | 21:59 |
| jeblair | it turns out, all we had to do this whole time was ask for help on #oftc! :) | 21:59 |
| clarkb | jeblair: ha | 22:00 |
| mordred | jeblair: silly us | 22:00 |
| pleia2 | jeblair: hahah | 22:00 |
| sdague | so all clients have full scrollback of just what they did not see | 22:00 |
| dansmith | SlickNik: separate buffers for phone, laptop, desktop, tablet | 22:00 |
| nibalizer | jeblair: haaha | 22:00 |
| clarkb | sdague: I solve that problem by using one client :) | 22:00 |
| * fungi laughed out loud | 22:00 | |
| dansmith | sdague: mine is all puppeted so I wasn't going to complain about that :) | 22:00 |
| fungi | it's funey 'cause it's true | 22:00 |
| SlickNik | dansmith: ah, I can now see why that would be useful. :) | 22:00 |
| *** Sukhdev has quit IRC | 22:00 | |
| sdague | dansmith: mine is too :) | 22:00 |
| sdague | I stole your policy | 22:01 |
| dansmith | oh, now I remember | 22:01 |
| sdague | clarkb: yeh, some of us like prettier fonts :) | 22:01 |
| dansmith | irssi from my phone is not my idea of a good time :) | 22:01 |
| jeblair | some of us like metal fonts | 22:01 |
| sdague | jeblair: we can't all be as hardcore as yuo | 22:01 |
| clarkb | dansmith: it works really well, at least compared to andchat | 22:02 |
| fungi | the last replacement git server is ready to go into service now... https://review.openstack.org/77994 | 22:02 |
| clarkb | I haven't tried any other native android clients | 22:02 |
| dansmith | clarkb: no notifications in the top bar means it's a no-go for me | 22:02 |
| clarkb | dansmith: ah, so you can get that with weechat using the weechat protocol | 22:02 |
| *** bhuvan_ has joined #openstack-infra | 22:02 | |
| *** bhuvan has joined #openstack-infra | 22:02 | |
| clarkb | which is silly for other reasons | 22:02 |
| *** lcheng_ has joined #openstack-infra | 22:03 | |
| clarkb | but totally doable | 22:03 |
| dansmith | clarkb: or just use andchat :) | 22:03 |
| *** mrmartin has quit IRC | 22:03 | |
| dansmith | don't get me wrong, andchat is really annoying, but... | 22:03 |
| *** jhesketh_ has joined #openstack-infra | 22:03 | |
| nibalizer | dansmith: i know several friends using notify-my-android and pebble watches | 22:04 |
| jhesketh_ | Morning | 22:04 |
| dansmith | nibalizer: I'm just not cool enough I guess | 22:04 |
| *** zns has joined #openstack-infra | 22:04 | |
| *** gondoi has joined #openstack-infra | 22:05 | |
| nibalizer | dansmith: im not sure the pebble watch is cool, though it is nerd-cool | 22:05 |
| dansmith | heh | 22:05 |
| gondoi | hi, does anyone know what version of tox is installed on the jenkins setup? | 22:05 |
| clarkb | gondoi: 1.6.1 | 22:05 |
| gondoi | clarkb: thank you | 22:05 |
| openstackgerrit | David Lyle proposed a change to openstack/requirements: Adding support for Django 1.6 https://review.openstack.org/77015 | 22:05 |
| *** thomasem_ has quit IRC | 22:05 | |
| clarkb | gondoi: >=1.7.0 is broken for us right now since upstream made backward incompatible changes | 22:06 |
| gondoi | clarkb: yes I was running into the same thing :D | 22:07 |
| gondoi | thanks again | 22:07 |
| clarkb | and holger is back according to my irc logs, I will try pinging him tomorrow mroning | 22:08 |
| *** mbacchi has quit IRC | 22:08 | |
| sdague | so is the trippleo gate not registering inuse nodes correctly? | 22:09 |
| *** fbo is now known as fbo_away | 22:09 | |
| sdague | because the total number of nodes seems to have dropped back down below 750 | 22:10 |
| dtroyer | jeblair, clarkb: this review is wedged on requirements: https://review.openstack.org/#/c/75852/ it is trying to fix the problem that gate-python-openstacksdk-requirement complains about. Shall we temporarily remove gate-python-openstacksdk-requirement or can you just push it in somehow? | 22:10 |
| sdague | and I'm assuming that's because the trippleo nodes are not in build state | 22:10 |
| clarkb | dtroyer: you can make your requirements match | 22:11 |
| dtroyer | that is trying to add the newlines that the job complains is missing. am I missing something? | 22:12 |
| *** eharney has joined #openstack-infra | 22:12 | |
| dtroyer | oh jeez... | 22:13 |
| dtroyer | I am missing something | 22:13 |
| clarkb | oh I actually have no idea what is going on there | 22:13 |
| sdague | clarkb: upstream requirements.txt lost it's newlines? | 22:13 |
| clarkb | sdague: I don't think so | 22:14 |
| dtroyer | the version doesn't match... | 22:14 |
| clarkb | since upstream is global-requirements.txt | 22:14 |
| clarkb | dtroyer: but http://logs.openstack.org/52/75852/1/check/gate-python-openstacksdk-requirements/ff846a9/console.html is complaining about newlines | 22:14 |
| sdague | clarkb: yeh, the change adds newlines | 22:15 |
| sdague | unless someone did it on a wonky OS | 22:15 |
| sdague | and those aren't real newlines | 22:15 |
| clarkb | sdague: that is my guess | 22:15 |
| jeblair | i just registered #openstack-doc!!!!1111one!!1 | 22:15 |
| fungi | jeblair: wow. eleven | 22:15 |
| clarkb | in seattle we go to 12! | 22:16 |
| clarkb | jeblair: have a 12 | 22:16 |
| gondoi | clarkb: do you use devpi or any kind of caching with pip? | 22:16 |
| fungi | jeblair: how long had you been trying to wrest control of that channel? | 22:16 |
| SlickNik | lol@clarkb | 22:16 |
| gondoi | clarkb: also, do you know if it works for requirements pointed at a repo? | 22:16 |
| fungi | gondoi: in production we build a restricted-content mirror | 22:16 |
| jeblair | fungi: 2.5 years? | 22:17 |
| gondoi | fungi: does that mean you mirror pypi? | 22:17 |
| fungi | gondoi: yes, part of it anyway | 22:17 |
| clarkb | fungi: gondoi: but only sort of | 22:17 |
| fungi | gondoi: and requirements which use a hyperlink (to tarballs or repositories) are a bad idea at this point... pip 1.5 and later really will give you fits over that | 22:17 |
| clarkb | fungi: gondoi: it is a "mirror" but really more like a display that takes a video feed of pypi and modifies it to be prettier when you look at it | 22:17 |
| clarkb | our "mirror" doesn't mirror external package links but the packages themselves and it will build platform dependent wheels for the platform it is run on | 22:18 |
| *** dhellmann is now known as dhellmann_ | 22:18 | |
| fungi | gondoi: we mirror just the parts of pypi which we need to install openstack components and run tests on them | 22:18 |
| gondoi | fungi: what type of fits are you talking about... i'm having issues with tox and package installs failing inconsistently | 22:18 |
| fungi | gondoi: oh, inconsistency is usually a sign that one or more of the packages you're requiring are hosted on an intermittently broken host | 22:19 |
| gondoi | is there an approved pip cache location? we set jenkins tox to ~/cache/pip in our tox.ini | 22:19 |
| jeblair | and registered openstack-hyper-v and openstack-cinder too | 22:20 |
| fungi | gondoi: the fits pip 1.5 and later will give you over hyperlink requirements is that you need to pass extra flags to override its security enforcement (ssl, cert validation, et cetera) | 22:20 |
| *** smarcet has left #openstack-infra | 22:20 | |
| sdague | dtroyer: oops I raced you with newlines | 22:20 |
| sdague | you should stomp over me again | 22:21 |
| dtroyer | ok…od showed me 0x0a in the file... | 22:21 |
| fungi | any objections to approving https://review.openstack.org/77994 so i can bring the new git01 online? | 22:22 |
| clarkb | fungi: looking | 22:22 |
| fungi | that will conclude our git server farm replacement | 22:22 |
| clarkb | I say go for it | 22:22 |
| fungi | thanks clarkb! | 22:22 |
| openstackgerrit | A change was merged to openstack-infra/config: Add puppetdb server to cacti https://review.openstack.org/76583 | 22:23 |
| fungi | i'll make sure it's operating sanely under load before i disappear for dinner | 22:23 |
| clarkb | sdague: did ps2 not have newlines? | 22:24 |
| fungi | i've got one more catch-up rsync going on docs-draft, and then when i return from dinner i'll swap the mount between the old and new filesystems and do one last rsync (without --delete) to catch stragglers | 22:24 |
| clarkb | fungi: sounds good | 22:24 |
| sdague | clarkb: dtroyer and I raced | 22:24 |
| *** mfer has quit IRC | 22:24 | |
| sdague | we were both working on patches and pushed | 22:24 |
| fungi | once that's done i can delete the old docs-draft and expand the logs volume | 22:24 |
| clarkb | sdague: oh ha | 22:24 |
| clarkb | sdague: then ps1 was missing them afterall? | 22:24 |
| sdague | I think so | 22:24 |
| jeblair | and #openstack-manila | 22:25 |
| dtroyer | I also updated the versions within to match global… | 22:25 |
| openstackgerrit | A change was merged to openstack-infra/config: Move git01.openstack.org to a new server https://review.openstack.org/77994 | 22:26 |
| sdague | silly macs | 22:26 |
| fungi | in other news, the last static jenkins slaves have just been deleted from jenkins01 and jenkins02, so all our jenkins masters except jenkins.o.o are nodepool-only now | 22:27 |
| clarkb | woot | 22:27 |
| sdague | woot | 22:27 |
| fungi | we are twisting all the knobs this week, it seems | 22:28 |
| *** bhuvan has quit IRC | 22:28 | |
| *** bhuvan_ has quit IRC | 22:29 | |
| fungi | also, this is one SERIOUSLY long board meeting. i may have to miss the end of it to go eat something | 22:29 |
| sdague | isn't that the definition of those things? :) | 22:29 |
| clarkb | do these meetings get announced somewhere? I probably need to be on a new ml | 22:29 |
| clarkb | I never catch them until its too late | 22:29 |
| jeblair | clarkb: foundation@lists.o.o | 22:30 |
| fungi | clarkb: http://lists.openstack.org/pipermail/foundation/2014-February/001643.html | 22:30 |
| fungi | it's *extremely* low-traffic | 22:30 |
| clarkb | thanks that is indeed a ml I am not on | 22:30 |
| *** e0ne has quit IRC | 22:30 | |
| clarkb | and if it is low traffic I should just get on it | 22:30 |
| anteaya | are they using asterisk for the meeting? | 22:30 |
| jeblair | [far less than it should be :( ] | 22:30 |
| fungi | anteaya: unfortunately not | 22:30 |
| jeblair | anteaya: no, we'd need to do something to mute non-board members for that i think | 22:30 |
| anteaya | I thought that was why we set it up | 22:31 |
| fungi | anteaya: webex with a google hangout just in case it's needed | 22:31 |
| anteaya | ah | 22:31 |
| jeblair | anteaya: should be possible, but a little more work than we've done so far | 22:31 |
| * anteaya nods | 22:31 | |
| mordred | fungi: it's a full-day meeting | 22:31 |
| jeblair | anteaya, fungi: there should be toll free us/canada numbers | 22:31 |
| fungi | anteaya: though i just dial in, lurk in #openstack-board and pull up the etherpad (at least they're using some free/accessible stuff anyway!) | 22:32 |
| fungi | mordred: i figured | 22:32 |
| fungi | your remarks have been insightful as always | 22:32 |
| sdague | has there been shouting? | 22:32 |
| fungi | not as much as usual... | 22:33 |
| jeblair | anyone know if you can list channels you have registered or where are on the access list? | 22:33 |
| *** oubiwan__ has quit IRC | 22:33 | |
| ekarlso | is the zuul data available in some restish nice ish api ? | 22:33 |
| jeblair | ekarlso: it's a json blob | 22:34 |
| *** briancurtin has quit IRC | 22:34 | |
| ekarlso | jeblair: available where ? : ) | 22:34 |
| clarkb | is openstack-board for the meetings? | 22:34 |
| anteaya | yes, I have dialed in before, much point of doing so now? | 22:34 |
| anteaya | are they winding up? | 22:34 |
| jeblair | ekarlso: /status.json | 22:34 |
| sdague | jeblair: /msg nickserv listchans | 22:35 |
| sdague | I think | 22:35 |
| fungi | clarkb: yeah | 22:35 |
| jeblair | sdague: thanks, that's it! | 22:35 |
| mordred | wow. I own #openstack-bacon apparently | 22:35 |
| StevenK | Bwahaha | 22:36 |
| fungi | anteaya: it's been the defcore working group discussion/update for a few hours now. no idea when it's wrapping up | 22:36 |
| clarkb | fungi: its open now I am there :) | 22:36 |
| dolphm | mordred: congratulations | 22:36 |
| anteaya | fungi: ugh | 22:36 |
| pleia2 | mordred: you broke StevenK, he's all laughs now | 22:36 |
| anteaya | I think I get enough of that at the tc mmeetings | 22:36 |
| mordred | pleia2: :) | 22:36 |
| sdague | mordred: now you need to make an incubated project | 22:37 |
| fungi | i've heard that the tc never turns down any project for incubation, especially if they bring cookies | 22:37 |
| clarkb | ok I am on the list now | 22:38 |
| StevenK | Deploying bacon using Openstack? | 22:38 |
| clarkb | hopefully I will not miss the next meeting | 22:38 |
| jeblair | mordred: /msg chanserv access #openstack-bacon add openstackinfra +AFRfiorstv | 22:38 |
| fungi | StevenK: even better... openstack-powered bacon | 22:38 |
| StevenK | Haha | 22:38 |
| StevenK | Bacon, with the power of the cloud! | 22:38 |
| *** oubiwan__ has joined #openstack-infra | 22:38 | |
| sdague | bacon should be some sort of naming or relationship service | 22:38 |
| mordred | jeblair: done | 22:38 |
| sdague | like kevin bacon | 22:38 |
| anteaya | the /msg nickserv listchans works for me on freenode but not oftc | 22:38 |
| pleia2 | anteaya: yeah, they run different services | 22:39 |
| anteaya | k | 22:39 |
| fungi | or sir frances | 22:39 |
| *** rossella_s has quit IRC | 22:39 | |
| StevenK | The channel modes are also COMPLETLY different | 22:39 |
| fungi | grr, francis | 22:39 |
| StevenK | Because Freenode decided to change everything when they rewrote their ircd | 22:39 |
| fungi | they're using charybdis at this point, yeah? | 22:40 |
| StevenK | I stopped keeping track after the second rebase | 22:40 |
| fungi | i've been slowly working on moving my servers from dancer-ircd to charybdis | 22:40 |
| pleia2 | ircd-seven afaik | 22:40 |
| fungi | ahhh, that | 22:40 |
| StevenK | I've been on Freenode since 2001, and the ircd has changed like four times | 22:40 |
| openstackgerrit | A change was merged to openstack-infra/config: Rename oslo.sphinx to oslosphinx https://review.openstack.org/73709 | 22:41 |
| sdague | oof, we hit another one of those zuul resets | 22:41 |
| sdague | at least the nova change landed first | 22:41 |
| fungi | at least we don't lose any changes when that happens, we "just" lose an hour of valuable testing time and hundreds of servers | 22:42 |
| sdague | yeh | 22:42 |
| *** echohead_ has joined #openstack-infra | 22:42 | |
| fungi | and the nova change is what caused it, at least indirectly | 22:42 |
| sdague | ok | 22:42 |
| sdague | well the important thing is that merged | 22:42 |
| fungi | because when you look in the dictionary for the definition of huge, one of the entries is nova's git repository | 22:43 |
| clarkb | fungi++ | 22:43 |
| sdague | meh, compared to what? :) | 22:43 |
| jeblair | we could reprioritize that, but i sorta figured that all told it'd be better to avoid the restart | 22:43 |
| sdague | it's only 20k changes, right | 22:43 |
| fungi | i've been using clones of nova to smoke out any potential problems with the git servers | 22:43 |
| anteaya | okay I've got them on speaker phone, where is their etherpad fungi? | 22:43 |
| clarkb | jeblair: ya I agree | 22:43 |
| fungi | anteaya: https://etherpad.openstack.org/p/unofficial-openstack-bod-march4-2014 | 22:43 |
| clarkb | the impact is minimal from a human perspective right now | 22:44 |
| anteaya | thanks | 22:44 |
| sdague | clarkb: well the impact is what merges for i3 | 22:44 |
| fungi | jeblair: absolutely. i don't want to restart zuul in the middle of all this if we don't have to | 22:44 |
| clarkb | sdague: meh, at some point we have to stop merging 200 commits on the last day :) | 22:44 |
| *** jcooley_ has joined #openstack-infra | 22:44 | |
| sdague | only 100 so far today | 22:44 |
| *** mriedem has quit IRC | 22:44 | |
| clarkb | there were a bunch of weeks after the summit to do stuff in | 22:45 |
| *** browne has left #openstack-infra | 22:45 | |
| anteaya | and clarkb said over 100 yesterday | 22:45 |
| sdague | yeh, though with a growing number of integrated projects, that really means these are going to grow | 22:45 |
| *** CaptTofu has quit IRC | 22:45 | |
| sdague | even if each project gets better | 22:45 |
| sdague | which, honestly, this doesn't feel like the mad rush of havana | 22:46 |
| *** sweston has joined #openstack-infra | 22:46 | |
| clarkb | sdague: ya this is calm | 22:46 |
| clarkb | I like it | 22:46 |
| sdague | and 2 oslo.messaging releases today that required changes in nova and keystone to unbreak pep8 | 22:46 |
| sdague | has caused some setbacks | 22:46 |
| clarkb | I am doing code review even | 22:46 |
| *** pdmars has quit IRC | 22:46 | |
| sdague | heh | 22:46 |
| clarkb | sdague: I just reviewed your integrated python jobs change | 22:46 |
| sdague | I'm mostly keeping nova code moving | 22:47 |
| *** lcestari has joined #openstack-infra | 22:47 | |
| sdague | clarkb: cool | 22:48 |
| geekinutah | jogo: how do I know which filenames I can query in logstash? | 22:49 |
| geekinutah | for example, I'm trying to query screen-q-dhcp.txt, doesn't seem to be going so well.. | 22:49 |
| *** VijayTripathi has joined #openstack-infra | 22:49 | |
| geekinutah | nm, found it.... | 22:49 |
| *** bhuvan has joined #openstack-infra | 22:50 | |
| clarkb | geekinutah: they are listed at https://git.openstack.org/cgit/openstack-infra/config/tree/modules/openstack_project/files/logstash/jenkins-log-client.yaml | 22:50 |
| *** nicedice has quit IRC | 22:50 | |
| jogo | geekinutah: and if something is missing you can add it there ^ | 22:50 |
| clarkb | geekinutah: you can also write lgostash queries to figure it out in most cases | 22:50 |
| *** nicedice has joined #openstack-infra | 22:50 | |
| fungi | just as a heads up, i started some dns cleanup on openstack.org last week... current state of that in case anyone has input on records i didn't delete but can (i left any still pointing to reachable systems)... https://etherpad.openstack.org/p/dns-cleanup-2014-02-27 | 22:50 |
| geekinutah | clarkb, jogo: thx | 22:51 |
| *** denis_makogon has quit IRC | 22:51 | |
| fungi | all total, i deleted a couple hundred cruft records so far | 22:51 |
| jeblair | okay, openstackinfra is now a founder on all the channels i was a founder on; that should be all of the known channels, and that brings its total to 54 | 22:51 |
| *** dims has quit IRC | 22:52 | |
| fungi | jeblair: terrific! also good to know that the right place to request freenode support is oftc ;) | 22:52 |
| mordred | jeblair: thats awesome | 22:52 |
| clarkb | fungi: before you disappaer https://review.openstack.org/#/c/71947/ has a response for you | 22:53 |
| openstackgerrit | Sean Dague proposed a change to openstack-infra/config: add job for testing that hooks work https://review.openstack.org/77728 | 22:53 |
| * fungi feels so... special | 22:53 | |
| *** mfink has quit IRC | 22:53 | |
| sdague | jeblair: I think that's right now | 22:53 |
| *** nati_ueno has quit IRC | 22:54 | |
| jeblair | fungi: thanks for the cleanup | 22:54 |
| *** bhuvan_ has joined #openstack-infra | 22:55 | |
| clarkb | by the way, update your gnutls | 22:55 |
| fungi | yup | 22:55 |
| fungi | gnutls for teh failz | 22:55 |
| jeblair | sdague: missed one thing, left comment, let me know if you want more explanation | 22:55 |
| *** blamar has joined #openstack-infra | 22:56 | |
| *** nati_ueno has joined #openstack-infra | 22:56 | |
| *** VijayTripathi has quit IRC | 22:56 | |
| openstackgerrit | Sean Dague proposed a change to openstack-infra/config: add job for testing that hooks work https://review.openstack.org/77728 | 22:57 |
| sdague | fixed | 22:57 |
| sdague | clarkb: and would you be updating that over an ssl connection... :) | 22:57 |
| jeblair | cool | 22:57 |
| clarkb | sdague: yes apt is affected too | 22:58 |
| fungi | sdague: validating the openpgp rsa signature on the package | 22:58 |
| clarkb | sdague: but packages are signed so meh | 22:58 |
| sdague | :) | 22:59 |
| sdague | yeh, that's true | 22:59 |
| clarkb | fungi: are package signature affected too? | 22:59 |
| sdague | which is actually a better situation than the apple bug | 22:59 |
| sdague | clarkb: it shouldn't be | 22:59 |
| *** ctracey has joined #openstack-infra | 22:59 | |
| fungi | clarkb: shouldn't be, no | 22:59 |
| clarkb | sdague: I didn't think so | 22:59 |
| clarkb | ah good so it isn't great but not end of the world like | 22:59 |
| fungi | uses gnupg (1.x has its own internal crypto, 2.x links an extracted libcrypt) | 23:00 |
| *** jaypipes has joined #openstack-infra | 23:00 | |
| fungi | gnutls has some non-ssl/tls primitives exposed, but gnupg doesn't use them | 23:00 |
| pleia2 | yeah :( | 23:01 |
| *** bhuvan__ has joined #openstack-infra | 23:01 | |
| *** bhuvan___ has joined #openstack-infra | 23:01 | |
| fungi | all's quiet on the western front--time to grab some grub... bbiaw | 23:03 |
| anteaya | so safe to update using apt? | 23:04 |
| clarkb | anteaya: relatively so :) | 23:04 |
| anteaya | as safe as any other method to update? | 23:05 |
| *** arborism has joined #openstack-infra | 23:05 | |
| *** bhuvan_ has quit IRC | 23:05 | |
| *** bhuvan has quit IRC | 23:05 | |
| anteaya | clarkb: and sweston has confirmed brocade does indeed want a 3rd 3rd party testing account: https://bugs.launchpad.net/openstack-ci/+bug/1287348 | 23:05 |
| anteaya | sweston: meet clarkb | 23:05 |
| *** jeckersb is now known as jeckersb_gone | 23:05 | |
| clarkb | anteaya: thank you for the checking | 23:05 |
| sweston | hi clarkb | 23:06 |
| anteaya | clarkb: np, we had a good chat | 23:06 |
| clarkb | hi | 23:06 |
| * clarkb semi secretly wishes that every company wouldn't need a bajillion test accounts but oh well | 23:06 | |
| anteaya | so we have two contacts sweston and shivharis for brocade | 23:06 |
| sweston | clarkb: so we are setting up several systems for parallel testing, across business units | 23:06 |
| *** amcrn has quit IRC | 23:06 | |
| *** jcoufal has quit IRC | 23:06 | |
| sweston | clarkb: hehe, well not so semi secretly now, right? | 23:07 |
| sweston | clarkb: but we can have just one account if that would make it easier for the infra team | 23:07 |
| clarkb | sweston: no its fine | 23:07 |
| clarkb | I lost the battle long ago. There are valid reasons for different accounts | 23:07 |
| clarkb | its just a bit shocking that 2 months ago there were like 3 accounts total for this stuff and now everyone needs 3 | 23:08 |
| *** dims has joined #openstack-infra | 23:08 | |
| sweston | hehe. I think priorities shifted when requirements changed | 23:08 |
| anteaya | and like I was explaining to sweston we just need to know folks know about each other | 23:09 |
| anteaya | and not creating a situation we are expected to sort out | 23:09 |
| jeblair | maybe once we get the gerrit upgrade situation sorted out, we can look into a plugin that organizes and displays 3rd party testing results differently to try to better deal with the number of systems we have | 23:10 |
| clarkb | jeblair: ++ | 23:10 |
| sdague | jeblair: yeh, we definitely need a different UI on this | 23:10 |
| sdague | on the up side, people care! | 23:10 |
| jeblair | sdague: yay! | 23:11 |
| *** dstanek has quit IRC | 23:12 | |
| sweston | so I do have a question on another matter, if now is a decent time to ask? is launch-node.py meant to work with nova network? | 23:12 |
| mordred | sweston: it does | 23:12 |
| *** dkranz has quit IRC | 23:12 | |
| anteaya | mikal has this so far: http://www.rcbops.com/gerrit/reports/neutron-cireport.html | 23:12 |
| anteaya | and the code: https://github.com/rcbau/hacks/tree/master/gerrit | 23:13 |
| sweston | jeblair: I would be interested in helping you with that as well | 23:13 |
| clarkb | jeblair: fungi SergeyLukjanov mordred https://review.openstack.org/#/c/70511/ my comment there is probably somewhat important for changes of that nature | 23:13 |
| *** sabari has quit IRC | 23:13 | |
| clarkb | jeblair: fungi SergeyLukjanov mordred at the very least we should be pciking up on those thigns so we don't do the maven properties derp again | 23:13 |
| mordred | sweston: we use it against clouds taht run nova-network, and against clouds that run neutron | 23:13 |
| jeblair | sweston: that's great! ping us after we've upgraded to gerrit 2.8. we're trying to do upstream-first development with gerrit, so we don't end up running a local fork anymore | 23:15 |
| mrodden | gerrit > 2.6 scares me | 23:16 |
| mrodden | they made so many changes | 23:16 |
| sweston | mordred: aha, thank you. I am running it with Neutron and I had some issues when using it with tenant networks | 23:16 |
| jeblair | mrodden: they did, and i have some serious reservations about the new change screen in 2.9. but 2.8 on https://review-dev.openstack.org/#/q/status:open,n,z looks pretty promising | 23:17 |
| mrodden | i haven't looked at 2.9 at all yet | 23:17 |
| sweston | jeblair: sweet, that will be a great change!! I guess I can get more information at this weeks infra meeting? | 23:17 |
| *** bhuvan___ has quit IRC | 23:17 | |
| *** bhuvan__ has quit IRC | 23:17 | |
| jeblair | sweston: that was a few hours ago, but i'm sure we'll talk about it at the next meeting | 23:18 |
| sweston | mordred: but it helps to know how you are currently using it. | 23:18 |
| *** zns has quit IRC | 23:19 | |
| jeblair | actually the new change screen is in 2.8 as well, but we're not making it the default | 23:19 |
| *** bhuvan has joined #openstack-infra | 23:19 | |
| *** bhuvan_ has joined #openstack-infra | 23:19 | |
| *** dprince has quit IRC | 23:19 | |
| clarkb | jeblair: 2.9 makes it mandatory right? | 23:19 |
| mrodden | sounds like 2.10 it will be the only one | 23:20 |
| mrodden | in 2.10 | 23:20 |
| jeblair | yeah. we're going to need to fix some serious usability issues if we want to move to a version where it's mandatory | 23:20 |
| *** sparkycollier has quit IRC | 23:20 | |
| *** mkoderer has quit IRC | 23:21 | |
| * StevenK sobs about Gerrit some more | 23:21 | |
| sweston | jeblair: Okay. I am adding it to my calendar, so I will be there next week ... please let me know if I can be helpful until then | 23:21 |
| jeblair | sweston: cool. if you haven't already read them, we have some docs about how the infra team operates (we try to do everything in public and invite participation): http://ci.openstack.org/ | 23:22 |
| sweston | jeblair: awesome :-) | 23:23 |
| jeblair | sweston: and zaro is leading the effort to upgrade gerrit, so you might be interested in reviewing his patches: https://review.openstack.org/#/dashboard/6987 | 23:23 |
| jeblair | sweston: (zaro == Khai Do) | 23:23 |
| mordred | StevenK: what's wrong with gerrit | 23:24 |
| mrodden | the new change screen is weird... | 23:24 |
| clarkb | jeblair: I should check on my patch upstream, but the ant thing for buck made me really not want to touch it with a 10 foot pole | 23:24 |
| mrodden | and they have these.... icon things... | 23:24 |
| *** bhuvan_ has quit IRC | 23:24 | |
| *** bhuvan has quit IRC | 23:24 | |
| jeblair | sweston: getting familiar with the config repo so you can help review things there would be extremely helpful for us, and keep you plugged into what we're doing (and of course give you a voice) | 23:24 |
| StevenK | mordred: My major complaint is that they refuse our patches and change everything. | 23:24 |
| sweston | jeblair: I see there are only a few of them, hehe | 23:24 |
| mordred | StevenK: oh! that. yes. | 23:25 |
| jeblair | mrodden: yeah, we're not having avatars. | 23:25 |
| mordred | jeblair: we're not? but avatars make me feel prioritized as an ego | 23:25 |
| jeblair | sticking people's faces into source code is one of the many things that does not make it easier to read (and in fact does the opposite) | 23:25 |
| mrodden | someone got their social networking tools stuck in my code review tools | 23:25 |
| *** jergerber has quit IRC | 23:25 | |
| *** rcleere has quit IRC | 23:25 | |
| *** mfink has joined #openstack-infra | 23:26 | |
| mrodden | "how can you -2 my patch with my super awesome smile?!" | 23:26 |
| sweston | jeblair: excellent. that gives me plenty to do for this week :-) | 23:27 |
| jeblair | sweston: awesome! | 23:27 |
| mrodden | on openstack-infra/config, is there anything that verifies the jenkins job configs actually work in Jenkins? i only see the compare-xml and some zuul stuff | 23:27 |
| StevenK | mrodden: Haha | 23:27 |
| clarkb | mrodden: or to be less than StevenK, there isn't | 23:29 |
| jeblair | mrodden: the compare-xml and layout jobs exercise jjb on the change, so if the jjb syntax breaks, it should show up there. and theoretically, jjb shouldn't generate an invalid jenkins config | 23:29 |
| jeblair | mrodden: but the resulting job itself doesn't get run or anything | 23:29 |
| *** jcoufal has joined #openstack-infra | 23:30 | |
| mrodden | ok thats what i thought | 23:30 |
| jeblair | mrodden: so usually we land changes to add new jobs as non-voting if we have no idea if they will work, then exercise them, then make them voting | 23:30 |
| StevenK | clarkb: I was laughing at "How can you -2 my patch ...", not the question. | 23:30 |
| mrodden | ok | 23:30 |
| clarkb | StevenK: oh :P | 23:30 |
| mrodden | jeblair: yeah that was my next question "how does one test a job..." | 23:30 |
| mrodden | thanks | 23:30 |
| *** dcramer__ has joined #openstack-infra | 23:32 | |
| openstackgerrit | Morgan Fainberg proposed a change to openstack-infra/config: Add in non-voting sample-config check https://review.openstack.org/78038 | 23:32 |
| *** david-lyle has quit IRC | 23:34 | |
| openstackgerrit | Morgan Fainberg proposed a change to openstack-infra/config: Add in non-voting sample-config check https://review.openstack.org/78038 | 23:34 |
| clarkb | mordred: wenlock: re https://review.openstack.org/#/c/51425/ I think the only way that ever gets in is if we have an informal infra change freeze, and work only on getting that through | 23:37 |
| clarkb | mordred: wenlock: otherwise it will constantly be in rebase hell | 23:37 |
| *** prad has quit IRC | 23:38 | |
| *** jcoufal has quit IRC | 23:39 | |
| *** thedodd has quit IRC | 23:39 | |
| *** dprince has joined #openstack-infra | 23:40 | |
| morganfainberg | dhellmann_, as discussed https://review.openstack.org/#/c/78038/ cc fungi | 23:40 |
| *** jcoufal has joined #openstack-infra | 23:41 | |
| wenlock | clarkb, im fine for waiting ... as long as i don't have to wait for pigs to fly :D | 23:43 |
| clarkb | wenlock: well I think we need to be more proactive than waiting | 23:43 |
| clarkb | and schedule something | 23:43 |
| wenlock | clarkb, that would be fantastic, the next rebase is going to be rough for us... because it includes a big zuul update | 23:43 |
| wenlock | clarkb, i think i want to give our team some time to consume that latest zuul update.... | 23:44 |
| *** mwagner_lap has quit IRC | 23:44 | |
| wenlock | im also wondering if someone has solved the whole behind a firewall thing with pip in another way.... | 23:44 |
| *** markwash has quit IRC | 23:45 | |
| openstackgerrit | Khai Do proposed a change to openstack-infra/gerrit: Add a Change Owner group https://review.openstack.org/65133 | 23:45 |
| *** sarob_ has quit IRC | 23:46 | |
| *** sarob_ has joined #openstack-infra | 23:46 | |
| clarkb | wenlock: you can run a pip mirror | 23:47 |
| *** zns has joined #openstack-infra | 23:47 | |
| *** bhuvan has joined #openstack-infra | 23:48 | |
| krotscheck | clarkb, have a moment? I'm trying to figure out the best way to handle per-slave npm configuration. | 23:48 |
| *** blamar has quit IRC | 23:48 | |
| clarkb | krotscheck: sure | 23:48 |
| krotscheck | There's basically three places you can load it from. NPM's install directory (which is in a venv), NPM's global path (also in venv), or the current user's home directory (~jenkins) | 23:49 |
| wenlock | clarkb, doh, gotcha | 23:49 |
| *** whoops has quit IRC | 23:49 | |
| krotscheck | Given that we really only want the slaves hitting that mirror, I figured option 3 would be best. | 23:49 |
| wenlock | clarkb, i was hoping for something .... well portable | 23:49 |
| *** blamar has joined #openstack-infra | 23:49 | |
| krotscheck | ... but jeblair wants it to be a not-hidden file. | 23:50 |
| clarkb | krotscheck: you put it in /etc though | 23:50 |
| clarkb | iirc | 23:50 |
| clarkb | wenlock: ya that option isn't great | 23:51 |
| krotscheck | clarkb: Well, right now jenkinsuser puts it in ~/.npmrc | 23:51 |
| krotscheck | ... maybe that's ok. | 23:52 |
| wenlock | clarkb, so i guess we just need to schedule... i can propose some dates after getting back with our team.... i want them to make sure we consume latest zuul with that rebase | 23:52 |
| *** bhuvan has quit IRC | 23:52 | |
| *** lcheng_ has quit IRC | 23:52 | |
| *** mrodden has quit IRC | 23:53 | |
| clarkb | wenlock: for things on our end prior to release candidates happening is probably best | 23:53 |
| clarkb | jeblair: fungi SergeyLukjanov mordred ^ what do you think? | 23:54 |
| openstackgerrit | Michael Krotscheck proposed a change to openstack-infra/config: Add NPM mirror https://review.openstack.org/68818 | 23:54 |
| jeblair | i3 is this week, so next week is promising | 23:54 |
| jeblair | maybe best to put it on the meeting agenda for tuesday and we can schedule a time? | 23:55 |
| jeblair | (targeting next thurs or friday or something?) | 23:55 |
| *** sweston has quit IRC | 23:56 | |
| wenlock | jeblair, clarkb, that should work....maybe i just need to prep a rebase this week | 23:57 |
| clarkb | ++ | 23:57 |
| wenlock | i think im flying tuesday | 23:57 |
| wenlock | but that might be evening, so might be able to make the meeting | 23:58 |
| jeblair | wenlock: if you can't make it, send an email with a proposed time, link to the changes, and rough series of steps and we can arrange it over email | 23:58 |
| wenlock | jeblair, cool | 23:59 |
| *** bhuvan has joined #openstack-infra | 23:59 | |
| *** bhuvan_ has joined #openstack-infra | 23:59 | |
| wenlock | jeblair, ive been meaning to learn how to post on the mail list, so i might just do that | 23:59 |
| *** lcheng_ has joined #openstack-infra | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!