| sdague | jeblair: so... the jobs that runs are very different than other jobs | 00:00 |
|---|---|---|
| *** rongze has quit IRC | 00:00 | |
| sdague | but if you have a better fingerprint, have at :) | 00:00 |
| jeblair | sdague: i don't understand your answer. | 00:01 |
| openstackgerrit | Clark Boylan proposed a change to openstack-infra/config: Import zuul-packaging into openstack-infra https://review.openstack.org/56107 | 00:01 |
| clarkb | jeblair: fungi: mordred: ^ pabelanger's change merge conflicted. I am going to self approve the resolution change shortly | 00:02 |
| sdague | the neutron large ops job runs a completely different payload than other jobs, so it was probably a short circuit | 00:02 |
| clarkb | but do feel free to rereview :) | 00:02 |
| *** fbo_away is now known as fbo | 00:02 | |
| sdague | basically, it is effective unique, but I agree, using job name is less than ideal | 00:02 |
| jeblair | sdague: fyi we now have 100.0 devstack tempest jobs. | 00:03 |
| sdague | nice | 00:03 |
| sdague | .0 | 00:03 |
| clarkb | we are going to delete about half of them though | 00:03 |
| sdague | very precise :) | 00:03 |
| sdague | heh | 00:04 |
| jeblair | sdague: that's 3 significant digits, not 1. :) | 00:04 |
| *** NobodyCam_afk is now known as NobodyCam | 00:04 | |
| clarkb | (not quite half as some of them are extras from my changes that aren't being run yet) | 00:04 |
| jeblair | clarkb: no, that's a local count in a clean jjb run. | 00:04 |
| clarkb | oh wow | 00:04 |
| fungi | clearly we need MORE | 00:05 |
| jeblair | sdague: so anyway, i bring it up because 1250168 only matches 1 of the 7 jobs that could trigger it. | 00:05 |
| jeblair | sdague: and listing all 7 is clearly not the right answer to that. :) | 00:06 |
| sdague | ok, cool | 00:06 |
| sdague | sorry, my brain is lost in trying to implement locally nested facets with the er results, because we can't get that from er | 00:06 |
| jeblair | sdague: and now my brain is lost too. | 00:07 |
| clarkb | sdague: where clauses would be so nice | 00:07 |
| clarkb | and sub query like things | 00:07 |
| sdague | clarkb: yeh | 00:08 |
| sdague | oh well | 00:08 |
| *** datsun_F40PH has quit IRC | 00:08 | |
| *** hogepodge has joined #openstack-infra | 00:08 | |
| ArxCruz | clarkb: around? | 00:09 |
| *** mrodden has quit IRC | 00:09 | |
| clarkb | ArxCruz: yes | 00:09 |
| ArxCruz | clarkb: so, do you want me to add a patch to ssh -t in nodepool ? | 00:09 |
| clarkb | ArxCruz: yes, I think that will be required | 00:09 |
| ArxCruz | using paramiko | 00:09 |
| clarkb | ArxCruz: jeblair may have better ideas, but I don't want to rely on special images as nodepool is building the special images for us | 00:10 |
| ArxCruz | clarkb: okay, it's when nodepool try to connect through ssh in the vm to install stuff correct ? | 00:10 |
| clarkb | ArxCruz: yes | 00:10 |
| clarkb | ArxCruz: the other alternative is to ssh in as root and do all of the things there | 00:10 |
| ArxCruz | clarkb: I see, so, o-infra is already testing fedora ? | 00:10 |
| clarkb | ArxCruz: I think some of what nodepool does happens ssh'd in as root | 00:10 |
| ArxCruz | afaik this only happens in fedora, not in precise | 00:10 |
| clarkb | ArxCruz: we tried adding fedora hosts late last week so that dprince could start that and this is when we saw the problem | 00:11 |
| ArxCruz | clarkb: I see, I got this problem my my FC too, but this there's nothing to do with my patch correct? It's a new patch, I wondering why you had to rollback | 00:11 |
| clarkb | ArxCruz: we had to rollback the change that added fedora nodes to nodepool because they were failing image builds | 00:12 |
| clarkb | and taking up resources | 00:12 |
| ArxCruz | the easiest way to fix in my side was enable tty in visudoers | 00:12 |
| ArxCruz | clarkb: ahhh, not mine patch, got it | 00:12 |
| clarkb | ArxCruz: right, but it is nodepool's job to do that for us and it can't :) | 00:12 |
| openstackgerrit | A change was merged to openstack-infra/config: Import zuul-packaging into openstack-infra https://review.openstack.org/56107 | 00:12 |
| ArxCruz | clarkb: tomorrow I will work on a patch :) | 00:12 |
| clarkb | ArxCruz: awesome thanks | 00:12 |
| ArxCruz | yw | 00:13 |
| *** rcleere has quit IRC | 00:13 | |
| *** fbo is now known as fbo_away | 00:13 | |
| *** yamahata_ has quit IRC | 00:15 | |
| *** weshay has joined #openstack-infra | 00:16 | |
| *** dkliban has quit IRC | 00:17 | |
| *** wenlock_ has quit IRC | 00:17 | |
| pabelanger | clarkb, cool! | 00:18 |
| pabelanger | also | 00:18 |
| pabelanger | \o/ | 00:18 |
| clarkb | pabelanger: we are using you as the guinea pig to see if new manage-projects is happy | 00:18 |
| pabelanger | clarkb, fair enough | 00:19 |
| clarkb | waiting for puppet to fire and pull run manage-projects is pretty boring | 00:21 |
| fungi | brace for impending disappointment (maybe) | 00:21 |
| clarkb | jog0: https://jenkins02.openstack.org/job/gate-tempest-devstack-vm-neutron-large-ops-default/75/console large ops seems to be failing a bit | 00:22 |
| clarkb | for the new hpcloud region nodepool config change, I think I may apply that change to nodepool.o.o by hand to make it easy to deal with "reverts" | 00:23 |
| *** dims has joined #openstack-infra | 00:23 | |
| *** dcramer_ has quit IRC | 00:24 | |
| clarkb | manage-projects is running | 00:24 |
| *** rnirmal has quit IRC | 00:27 | |
| clarkb | seems to have populated data for zuul-packaging, but hasn't updated the acls | 00:27 |
| clarkb | waiting for the script to finish running before I hit it with a hammer | 00:27 |
| openstackgerrit | A change was merged to openstack-infra/config: Make job names shorter https://review.openstack.org/59564 | 00:28 |
| *** matsuhashi has joined #openstack-infra | 00:29 | |
| *** markwash has quit IRC | 00:30 | |
| pabelanger | clarkb, ping me once the repo is good, I have some patches already lined up | 00:30 |
| clarkb | hrm there is a cron that runs manage-projects too | 00:32 |
| clarkb | I bet that is one reason we have trouble with github | 00:32 |
| clarkb | too many cooks in the kitchen | 00:32 |
| fungi | clarkb: github project creation succeeded | 00:32 |
| *** nati_uen_ has joined #openstack-infra | 00:33 | |
| clarkb | fungi: I think everything worked, but applying the ACLs | 00:33 |
| fungi | also, i'm not again' a cron job running m-p but if we keep it we should have that and the puppet one share a lockfile | 00:33 |
| clarkb | fungi: or just remove the exec in puppet and ask people to wait half an hour | 00:34 |
| *** nati_ueno has quit IRC | 00:35 | |
| clarkb | rerunning it by hand populated the ACLs | 00:35 |
| clarkb | I forgot I had to add a -v to see the debug output though :( so I didn't see any data | 00:35 |
| fungi | disappointing. i wonder what the issue was | 00:35 |
| clarkb | I am guessing there may still be a race around group creation | 00:35 |
| clarkb | we might need to do something hacky and add a sleep | 00:36 |
| clarkb | pabelanger: in any case you are good to go | 00:36 |
| *** ^d has quit IRC | 00:36 | |
| clarkb | zuul clone looks good too | 00:37 |
| *** ^d has joined #openstack-infra | 00:37 | |
| *** CaptTofu has quit IRC | 00:38 | |
| pabelanger | clarkb, roger that | 00:38 |
| *** CaptTofu has joined #openstack-infra | 00:39 | |
| pabelanger | now that it is imported, I'll remove my repo and clone | 00:39 |
| openstackgerrit | Clark Boylan proposed a change to openstack-infra/jeepyb: Always sleep after group creation. https://review.openstack.org/59583 | 00:39 |
| clarkb | fungi: hacks like ^ might help | 00:39 |
| clarkb | I need to head home shortly to catch the beginning of the football game | 00:40 |
| clarkb | I plan on adding the new hpcloud region by hand to nodepool once there | 00:40 |
| clarkb | unless someone else wants to give it a shot | 00:40 |
| * clarkb AFKs | 00:41 | |
| fungi | awcrapdoubleunplus | 00:44 |
| fungi | i'll reverify the changes which got lost when zuul tried to fire nonexistent "dvsm" jobs in the past few minutes | 00:45 |
| fungi | at least the ~3 i spotted | 00:45 |
| *** nati_uen_ has quit IRC | 00:45 | |
| fungi | and i even saw the mention that we'd probably want to stop puppet on zuul.o.o for that change, then totally spaced it | 00:46 |
| *** nati_ueno has joined #openstack-infra | 00:46 | |
| *** sdake_ has quit IRC | 00:47 | |
| *** sdake has quit IRC | 00:47 | |
| *** hogepodge has quit IRC | 00:49 | |
| *** matsuhashi has quit IRC | 00:50 | |
| clarkb | oh whoops | 00:50 |
| fungi | zaro: on 57333 i was going to add the three new entries for your change to hiera... what do i need to do to generate the uuid-like identifier for jenkins_dev_credentials_id? | 00:50 |
| *** matsuhashi has joined #openstack-infra | 00:50 | |
| dims | message:"No workspace found, files cannot be copied." <<< Just logged a bug 1257125 | 00:51 |
| uvirtbot | Launchpad bug 1257125 in openstack-ci "No workspace found, files cannot be copied" [Undecided,New] https://launchpad.net/bugs/1257125 | 00:51 |
| fungi | dims: have an example change? | 00:52 |
| fungi | i'm trying to figure out how to get to the ful log for a failure matching that from the logstash reference you linked in the bug | 00:53 |
| fungi | (there's a good chance it's my fault) | 00:53 |
| *** michchap_ has quit IRC | 00:53 | |
| *** matsuhas_ has joined #openstack-infra | 00:53 | |
| *** michchap has joined #openstack-infra | 00:53 | |
| fungi | ahh, found one | 00:53 |
| *** matsuhashi has quit IRC | 00:54 | |
| *** krotscheck has quit IRC | 00:54 | |
| fungi | and yeah, that was me--thanks for filing that, dims. at least now i know what it's going to look like for the others which haven't reported yet | 00:54 |
| dims | cool :) | 00:55 |
| dims | biab | 00:55 |
| *** dangers is now known as danger_fo_away | 00:55 | |
| *** marun has joined #openstack-infra | 00:55 | |
| *** dolphm has quit IRC | 00:56 | |
| *** dcramer_ has joined #openstack-infra | 00:59 | |
| *** reaper has quit IRC | 01:00 | |
| dims | fungi, job rename issues? | 01:01 |
| fungi | dims: well, me issues. for merging a job rename change hastily | 01:01 |
| *** jerryz has quit IRC | 01:01 | |
| fungi | thankfully it only seems to have impacted a small handful of jobs, but that was one of them | 01:02 |
| *** marun has quit IRC | 01:02 | |
| dims | nice thanks | 01:02 |
| *** sdake has joined #openstack-infra | 01:03 | |
| *** ryanpetrello has joined #openstack-infra | 01:03 | |
| fungi | looks like the last of that hanmdful of changes has reported and been rechecked/reverified as needed | 01:08 |
| *** herndon_ has joined #openstack-infra | 01:13 | |
| *** sdake has quit IRC | 01:15 | |
| ekarlso | anyone here with git kungfu skills ? | 01:17 |
| clarkb | sure, whats up | 01:18 |
| ekarlso | trying to rebase https://review.openstack.org/#/c/58316/, so I do a git review -d <changeset> then do git rebase master | 01:18 |
| ekarlso | but it borks Failed to merge in the changes.... so I fix it up and do "git add" on the files | 01:18 |
| ekarlso | basically struggling with rebase at all :( | 01:19 |
| *** beagles has quit IRC | 01:19 | |
| clarkb | after you git add all of the files that show up as unmerged in `git status` you need to `git rebase --continue` | 01:20 |
| *** yamahata_ has joined #openstack-infra | 01:20 | |
| *** mrodden has joined #openstack-infra | 01:23 | |
| ekarlso | clarkb: yeh, and then after fixing the "both modified and doing add on them | 01:26 |
| *** nosnos has joined #openstack-infra | 01:26 | |
| ekarlso | I do --continue and get No changes - did you forget to use 'git add'? | 01:26 |
| clarkb | ekarlso: what does git status say? | 01:27 |
| *** rongze has joined #openstack-infra | 01:27 | |
| clarkb | ekarlso: it sounds like you reverted all of the changes | 01:27 |
| clarkb | in the fixup you went back to the base instead of the new code | 01:27 |
| ekarlso | hmmm | 01:28 |
| ekarlso | shouldn't it be | 01:28 |
| ekarlso | git review -d <changeset> | 01:28 |
| ekarlso | git rebase master | 01:28 |
| ekarlso | ? | 01:28 |
| clarkb | yes | 01:29 |
| clarkb | ekarlso: but when you resolve the conflicts you can essentially go back to having what was in master | 01:29 |
| ekarlso | hmmmms, my head doesn't wrap around that atm /: | 01:30 |
| clarkb | ekarlso: so with a rebase you are taking two commit trees and pasting one on top of the other | 01:30 |
| ekarlso | yeah | 01:30 |
| clarkb | ekarlso: this process is unable to automagically happen because there are conflicts. when resolving those conflicts it is possible to essentially noop the paste making the new code on master not change anything | 01:31 |
| *** rongze has quit IRC | 01:31 | |
| clarkb | you are doing master + something else. when resolving the conflicts something else can become nil | 01:31 |
| *** yamahata_ has quit IRC | 01:34 | |
| ekarlso | clarkb: is "git rebase" the way to make my change up to speed ? | 01:35 |
| clarkb | yes it moves your change from the old base to the new base | 01:36 |
| *** jcooley_ has joined #openstack-infra | 01:37 | |
| ekarlso | so rebase master, then git add ? | 01:37 |
| clarkb | you have to resolve any git conflicts in between | 01:38 |
| clarkb | git status will show you which files are unmerged | 01:38 |
| ekarlso | http://pastebin.com/EP7kaMC5 < clarkb, I know how to resolve the both modifieds but what's up with "added by us" ? | 01:39 |
| *** sjing has joined #openstack-infra | 01:39 | |
| clarkb | ekarlso: they are new files added by your change | 01:40 |
| fungi | possibly the unspoken part here is that you need to edit the files which are marked as "both modified" and then git add those once they look they way that you'll want them in the final merged version | 01:40 |
| clarkb | they are not in master | 01:41 |
| *** sarob_ has joined #openstack-infra | 01:41 | |
| fungi | differences between the two will be denoted by start and end markers like <<<<<<<< and ======== and >>>>>>>> | 01:41 |
| fungi | inside the files you need to edit | 01:42 |
| clarkb | fungi: you going to be around for a little bit longer? if so I am going to try the hpcloud nodepool change shortly | 01:42 |
| fungi | clarkb: yes, will be right here | 01:42 |
| *** CaptTofu has quit IRC | 01:42 | |
| ekarlso | clarkb: they are very much in master the added by us files... | 01:42 |
| *** CaptTofu has joined #openstack-infra | 01:43 | |
| ekarlso | fungi: I fixed the confliects and did a git add on the both modified files | 01:43 |
| *** sarob has quit IRC | 01:43 | |
| clarkb | hmm is master "us" in this case? | 01:43 |
| clarkb | it might be | 01:43 |
| ekarlso | also after resolving and trying --continue again I get: No changes - did you forget to use 'git add'?... | 01:44 |
| ekarlso | I guess my git-fu isn't good :( | 01:44 |
| clarkb | ekarlso: right, that implies that you resolved the conflicts by setting them to what is in master | 01:44 |
| clarkb | so there are no changes | 01:44 |
| fungi | ekarlso: you needed to edit designate/tests/__init__.py and designate/tests/test_backend/test_nsd4slave.py (saving them in your editor), then git add each of them, then git rebase --continue | 01:45 |
| *** jcooley_ has quit IRC | 01:45 | |
| ekarlso | fungi: and yes I did .. | 01:45 |
| *** ryanpetrello has quit IRC | 01:45 | |
| fungi | okay, just making sure | 01:45 |
| clarkb | ekarlso: also check out git for four year olds or whatever schwerns talk is called, it is very good for understanding how git works | 01:45 |
| *** sarob_ has quit IRC | 01:45 | |
| fungi | http://2010.osdc.com.au/proposal/196/git-ages-4-and | 01:46 |
| fungi | aww, that conference took down the video linked there | 01:46 |
| ekarlso | well, not today.. :) it's allmost 3 am and I wanted to get my change rebased before bed :) | 01:46 |
| *** kui has joined #openstack-infra | 01:47 | |
| openstackgerrit | Clark Boylan proposed a change to openstack-infra/config: Add new HP Cloud region https://review.openstack.org/56260 | 01:49 |
| clarkb | fungi: ^ that is just a rebase (too many other changes would have been made to nodepool.o.o without the rebase | 01:49 |
| clarkb | fungi: applying that now | 01:50 |
| *** slong is now known as summerlong | 01:50 | |
| fungi | clarkb: yep, looks like what we discussed, i just hadn't circled back around to it since the azs got collapsed into one | 01:51 |
| fungi | lgtm | 01:52 |
| clarkb | the image build is going | 01:53 |
| *** nibalizer has quit IRC | 01:57 | |
| clarkb | puppet is running | 02:00 |
| *** reed has quit IRC | 02:00 | |
| clarkb | if this pans out I am going to have to request a bunch of quota in that region tomorrow | 02:00 |
| openstackgerrit | Paul Belanger proposed a change to openstack-infra/config: Add zuul-packaging to #openstack-infra https://review.openstack.org/59595 | 02:00 |
| *** jcooley_ has joined #openstack-infra | 02:01 | |
| *** MarkAtwood has quit IRC | 02:01 | |
| *** jcooley_ has quit IRC | 02:02 | |
| *** nibalizer has joined #openstack-infra | 02:03 | |
| ekarlso | nighty clarkb and thanks ! :) | 02:03 |
| *** mriedem has joined #openstack-infra | 02:05 | |
| *** rongze has joined #openstack-infra | 02:06 | |
| *** yaguang has joined #openstack-infra | 02:07 | |
| fungi | clarkb: what was the upshot of floating-ip quota discussions for that region? | 02:08 |
| clarkb | fungi: I think we are just going to ask for a bunch of quota | 02:08 |
| openstackgerrit | A change was merged to openstack-infra/devstack-gate: Remove kernel-headers apt-get call https://review.openstack.org/54386 | 02:08 |
| clarkb | and see what happens :) | 02:08 |
| fungi | not unlike... all previous quota requests i guess. sounds fine by me | 02:08 |
| fungi | now that i am not a network architect at an internet service provider, i've completely lost touch with ipv4 runout | 02:09 |
| *** nati_ueno has quit IRC | 02:09 | |
| clarkb | well I don't know if you noticed, but HP has two /8s | 02:09 |
| clarkb | I think it is silly to hold back on IPs but whatever | 02:09 |
| fungi | yes. in fact, i'd noticed that well before i worked for them. legacy grants | 02:09 |
| fungi | aww, arin's stopped printing their ipv4 death clock on their main page | 02:11 |
| clarkb | I get a public address at my desk which is silly | 02:11 |
| *** esker has joined #openstack-infra | 02:13 | |
| *** vkozhukalov has joined #openstack-infra | 02:19 | |
| fungi | ahh, enough hunting finds it at https://www.arin.net/resources/request/ipv4_countdown.html | 02:19 |
| fungi | arin still has 1.57 /8 blocks unallocated in aggregate | 02:20 |
| clarkb | image builds take forever | 02:21 |
| *** CaptTofu has quit IRC | 02:21 | |
| *** CaptTofu has joined #openstack-infra | 02:21 | |
| *** dkliban has joined #openstack-infra | 02:25 | |
| *** ryanpetrello has joined #openstack-infra | 02:27 | |
| *** matsuhas_ has quit IRC | 02:31 | |
| *** matsuhashi has joined #openstack-infra | 02:31 | |
| *** sdake has joined #openstack-infra | 02:32 | |
| clarkb | image build still going... | 02:37 |
| clarkb | I am assuming we are waiting on the snapshot to happen because the log isn't saying much anymore | 02:37 |
| clarkb | all of the old hpcloud region image builds started after new region and have appeared to complete before new region | 02:38 |
| clarkb | mordred: ^ | 02:38 |
| clarkb | so that doesn't bode well | 02:39 |
| *** ryanpetrello has quit IRC | 02:43 | |
| *** nati_ueno has joined #openstack-infra | 02:43 | |
| clarkb | fungi: what do you think? | 02:45 |
| clarkb | I don't see anything ni the logs | 02:45 |
| clarkb | spoke too soon, we appear to have an image now | 02:47 |
| fungi | mmm | 02:47 |
| fungi | clearly i disappeared for just the right amount of time | 02:48 |
| clarkb | waiting for some slaves to show up now | 02:48 |
| fungi | so it seems... slower so far? | 02:48 |
| clarkb | image builds are slower | 02:49 |
| clarkb | I am guessing that is due to time to create snapshot | 02:49 |
| *** senk has joined #openstack-infra | 02:49 | |
| clarkb | since the scripty bits seems to fly along just fine | 02:49 |
| clarkb | then the whole process hangs | 02:49 |
| clarkb | there is a region b slave building for jenkins02 right now | 02:53 |
| *** nati_ueno has quit IRC | 02:53 | |
| *** dkranz has joined #openstack-infra | 02:54 | |
| fungi | i guess we'll have more meaningful stats soon enough, in that case | 02:55 |
| clarkb | https://jenkins02.openstack.org/computer/devstack-precise-hpcloud-region-b-774933/ | 02:56 |
| *** sdake_ has joined #openstack-infra | 02:56 | |
| *** sdake_ has quit IRC | 02:56 | |
| *** sdake_ has joined #openstack-infra | 02:56 | |
| *** CaptTofu has quit IRC | 02:56 | |
| clarkb | and https://jenkins02.openstack.org/job/check-tempest-dsvm-neutron/44/console | 02:56 |
| *** CaptTofu has joined #openstack-infra | 02:57 | |
| *** dcramer_ has quit IRC | 02:57 | |
| clarkb | if that test comes out looking ok, I am going to leave nodepool.o.o as is, or you can approve the change that adds the region and I will restart puppet there | 02:58 |
| *** nosnos has quit IRC | 02:58 | |
| *** nosnos has joined #openstack-infra | 02:59 | |
| fungi | i can approve at that point | 03:00 |
| *** loq_mac has joined #openstack-infra | 03:00 | |
| mordred | clarkb: awesome. I'm sure the team will love to hear that :) | 03:01 |
| clarkb | mordred: are you watching the game? I am enjoying this | 03:03 |
| *** dkliban has quit IRC | 03:04 | |
| *** dkranz has quit IRC | 03:04 | |
| clarkb | there are 5 jobs running on region-b hosts, I have all of them open and will keep an eye on them | 03:04 |
| *** wenlock_ has joined #openstack-infra | 03:05 | |
| *** chandankumar has joined #openstack-infra | 03:06 | |
| *** nati_ueno has joined #openstack-infra | 03:07 | |
| *** gyee has quit IRC | 03:14 | |
| *** nati_ueno has quit IRC | 03:17 | |
| clarkb | fungi: https://jenkins02.openstack.org/job/gate-swift-dsvm-functional/9/ passed | 03:23 |
| clarkb | fungi: that isn't a tempest test though, we can wait a little longer for tempest things to finish | 03:23 |
| fungi | yeah, i'm continuing to keep tabs on the neutron one you linked earlier | 03:24 |
| clarkb | it seems a tad bit slower | 03:25 |
| *** dstanek_afk has joined #openstack-infra | 03:26 | |
| fungi | eh | 03:26 |
| fungi | perhaps | 03:27 |
| clarkb | the swift functional test typically runs in ~14 minutes according to jenkins, it took ~18-19 on new host | 03:27 |
| clarkb | but that could just be normal variability, need more numbers | 03:27 |
| fungi | sounds on par with rackspace | 03:27 |
| *** jcooley_ has joined #openstack-infra | 03:29 | |
| *** mrodden has quit IRC | 03:29 | |
| *** markwash has joined #openstack-infra | 03:31 | |
| *** mrodden has joined #openstack-infra | 03:36 | |
| mriedem | i've never seen this gate error before, not sure how to handle | 03:37 |
| mriedem | http://logs.openstack.org/78/35178/10/check/check-tempest-dsvm-full/0dbf076/logs/devstack-gate-setup-workspace-new.txt | 03:37 |
| mriedem | Unable to find ref refs/zuul/master/Zd2b3b9d225c04027a545a4dcfc9101f8 for openstack/glance | 03:37 |
| mriedem | is that an infra problem? | 03:37 |
| clarkb | mriedem: yes, there is a bug under openstack-ci somewhere | 03:38 |
| mriedem | looking | 03:38 |
| clarkb | the web server in front of those git repos can't keep up | 03:38 |
| clarkb | fungi: the neutron tests are pretty slow :( | 03:39 |
| fungi | mriedem: that message doesn=t usually indicate an error though | 03:39 |
| mriedem | fungi: just trying to look for something on that one to reverify against | 03:40 |
| mriedem | console.html didn't show much of anything | 03:40 |
| clarkb | fungi: are you sure? I thought all projects get the ref when zuul does a merge | 03:40 |
| fungi | the setup script asks forgiveness rather than permission when identifying which projects have zuul refs prepared | 03:41 |
| pabelanger | fungi, re: https://review.openstack.org/#/c/46339/ | 03:41 |
| pabelanger | If there is nothing else in the home/gerrit2/review_site/lib/ folder, you could setup puppet to purge the files | 03:41 |
| pabelanger | vs selecting specific files and setting them to absent | 03:42 |
| clarkb | fungi: ah gotcha, only projects with changes in them need to have zuul refs | 03:42 |
| pabelanger | I don't know what else is in the dir to properly comment | 03:42 |
| *** matsuhashi has quit IRC | 03:42 | |
| *** dims has quit IRC | 03:43 | |
| mriedem | fungi: is there something i'm not seeing in this log? http://logs.openstack.org/78/35178/10/check/check-tempest-dsvm-full/0dbf076/console.html | 03:44 |
| fungi | pabelanger: i'd have to check... i think there could be other stuff, but if memory serves (if it's the change i'm thinking about) then that might be the remainder | 03:45 |
| *** nati_ueno has joined #openstack-infra | 03:45 | |
| fungi | mriedem: no idea--i'm not on a computer where i can browse those logs easily | 03:46 |
| *** sdake has quit IRC | 03:46 | |
| *** sdake has joined #openstack-infra | 03:46 | |
| *** sdake has quit IRC | 03:46 | |
| *** sdake has joined #openstack-infra | 03:46 | |
| pabelanger | fungi, okay, comment left. Might be cleaner with purge to maintain it. Just noticed it passing by | 03:46 |
| fungi | clarkb: looks like the neutron job succeeded | 03:46 |
| clarkb | fungi: the neutron check test passed, I thnik we should go ahead and approve the config change, can you do that? | 03:46 |
| clarkb | jinx | 03:46 |
| fungi | yup--doing | 03:47 |
| clarkb | mriedem: it would be something in the setup workspace log that shows why it failed, that scripted exit 1'd for some reason | 03:47 |
| fungi | pabelanger: thanks | 03:48 |
| mriedem | clarkb: yeah, it exited with 1 after this "Unable to find ref refs/zuul/master/Zd2b3b9d225c04027a545a4dcfc9101f8 for openstack/glance" | 03:48 |
| clarkb | mriedem: yeah notice the RPC failed message, I think that was the problem | 03:48 |
| mriedem | yup | 03:49 |
| openstackgerrit | A change was merged to openstack-infra/config: Add new HP Cloud region https://review.openstack.org/56260 | 03:49 |
| mriedem | clarkb: i couldn't find an existing bug so i opened bug 1257155 | 03:49 |
| uvirtbot | Launchpad bug 1257155 in openstack-ci "check-tempest-dsvm-full fails in the gate" [Undecided,New] https://launchpad.net/bugs/1257155 | 03:49 |
| mriedem | so i could reverify against it quick | 03:50 |
| *** loq_mac has quit IRC | 03:52 | |
| *** loq_mac has joined #openstack-infra | 03:57 | |
| *** esker has quit IRC | 03:58 | |
| *** loq_mac has quit IRC | 03:58 | |
| *** esker has joined #openstack-infra | 03:59 | |
| *** ^d has quit IRC | 04:00 | |
| *** ^d has joined #openstack-infra | 04:01 | |
| *** resker has joined #openstack-infra | 04:03 | |
| *** esker has quit IRC | 04:03 | |
| clarkb | fungi: puppet is running on nodepool.o.o again | 04:04 |
| *** wenlock_ has quit IRC | 04:04 | |
| clarkb | I think we are good I am going to watch new orleans continue to lose | 04:04 |
| fungi | clarkb: awesome. enjoy | 04:04 |
| *** loq_mac has joined #openstack-infra | 04:05 | |
| *** loq_mac has quit IRC | 04:05 | |
| *** resker has quit IRC | 04:08 | |
| *** loq_mac has joined #openstack-infra | 04:11 | |
| *** weshay has quit IRC | 04:17 | |
| mriedem | clarkb: it'll only really be fun if carolina beats new orleans at home in a week | 04:17 |
| clarkb | no no this is fun :) | 04:18 |
| mriedem | clarkb: are you in seattle? | 04:19 |
| clarkb | yes | 04:19 |
| mriedem | well then, congratulations | 04:20 |
| mriedem | i'm in minnesota and well, you know how their season is going :( percy came back from injury just long enough to trounce the vikings and then get injured again. | 04:20 |
| *** michchap has quit IRC | 04:21 | |
| *** michchap has joined #openstack-infra | 04:21 | |
| *** rcleere has joined #openstack-infra | 04:25 | |
| *** matsuhashi has joined #openstack-infra | 04:25 | |
| *** rongze has quit IRC | 04:31 | |
| *** CaptTofu has quit IRC | 04:34 | |
| *** CaptTofu has joined #openstack-infra | 04:35 | |
| openstackgerrit | Jeremy Stanley proposed a change to openstack-infra/config: Updates to static.o.o volume management docs https://review.openstack.org/59615 | 04:38 |
| *** herndon_ has quit IRC | 04:43 | |
| lifeless | mordred: any answer for my q on https://review.openstack.org/#/c/58106/ ? | 04:43 |
| lifeless | russellb: btw I meant to ping you about reviewstats for bandwidth, but ELIFEGOTBUSY | 04:44 |
| *** mriedem has quit IRC | 04:53 | |
| *** DinaBelova has joined #openstack-infra | 04:56 | |
| *** DennyZhang has joined #openstack-infra | 04:57 | |
| *** sarob has joined #openstack-infra | 04:57 | |
| *** DinaBelova has quit IRC | 04:57 | |
| *** senk has quit IRC | 05:00 | |
| *** matsuhashi has quit IRC | 05:00 | |
| *** nono1 has joined #openstack-infra | 05:01 | |
| nono1 | Question regarding weekly meetup - Is this done on the same IRC channel #openstack-infra ?? | 05:01 |
| *** wenlock_ has joined #openstack-infra | 05:02 | |
| clarkb | no it is in #openstack-meeting | 05:03 |
| nono1 | so to attend the mtg, i go on webchat for #openstack-meeting ? | 05:05 |
| lifeless | IRC, yes. | 05:05 |
| nono1 | that means, no need to mention "openstack-infra" anywhere? but just "#openstack-meeting?" | 05:06 |
| *** jcoufal has joined #openstack-infra | 05:07 | |
| *** matsuhashi has joined #openstack-infra | 05:08 | |
| nono1 | 1 more q - IRC channel doesn't have Audio, right? then for the meeting how does Audio work? Is there any google hangout link for Tuesday mtgs? | 05:08 |
| *** dstanek has joined #openstack-infra | 05:09 | |
| *** ryanpetrello has joined #openstack-infra | 05:11 | |
| nono1 | pls someone answer to my query about Audio pls... | 05:12 |
| openstackgerrit | Khai Do proposed a change to openstack-dev/pbr: show diff between pip installed packages and requirements https://review.openstack.org/54639 | 05:12 |
| pleia2 | nono1: it's all text-based, no audio | 05:12 |
| *** ryanpetrello has quit IRC | 05:12 | |
| *** ryanpetrello has joined #openstack-infra | 05:13 | |
| *** senk has joined #openstack-infra | 05:16 | |
| *** nono1 has quit IRC | 05:16 | |
| *** markwash has quit IRC | 05:16 | |
| *** ryanpetrello has quit IRC | 05:17 | |
| *** pcrews has quit IRC | 05:19 | |
| *** nicedice has quit IRC | 05:23 | |
| *** crank has quit IRC | 05:26 | |
| *** matsuhashi has quit IRC | 05:27 | |
| fungi | wow, that was a surreal series of questions | 05:29 |
| *** crank has joined #openstack-infra | 05:31 | |
| *** DennyZhang has quit IRC | 05:31 | |
| *** jaypipes has quit IRC | 05:33 | |
| *** loq_mac has quit IRC | 05:34 | |
| *** ^demon has joined #openstack-infra | 05:41 | |
| *** ^d has quit IRC | 05:43 | |
| *** ^demon has quit IRC | 05:45 | |
| *** SergeyLukjanov has joined #openstack-infra | 05:48 | |
| *** sarob has quit IRC | 05:51 | |
| *** sarob has joined #openstack-infra | 05:51 | |
| *** jaypipes has joined #openstack-infra | 05:54 | |
| *** sarob has quit IRC | 05:56 | |
| *** senk has quit IRC | 05:56 | |
| *** matsuhashi has joined #openstack-infra | 05:59 | |
| *** ArxCruz has quit IRC | 06:00 | |
| openstackgerrit | Khai Do proposed a change to openstack-infra/config: add nodepool to jenkins-dev server https://review.openstack.org/57333 | 06:01 |
| zaro | fungi: ^ | 06:01 |
| zaro | fungi: that jenkins_dev_credentials_id var was a copy/paste mistake. | 06:02 |
| *** markwash has joined #openstack-infra | 06:02 | |
| *** DennyZhang has joined #openstack-infra | 06:08 | |
| *** pcrews has joined #openstack-infra | 06:18 | |
| *** dstanek has quit IRC | 06:19 | |
| *** sarob has joined #openstack-infra | 06:22 | |
| *** dstanek has joined #openstack-infra | 06:25 | |
| *** pcrews has quit IRC | 06:26 | |
| *** SergeyLukjanov has quit IRC | 06:27 | |
| lifeless | devananda: config/modules/openstack_project/templates/nodepool/nodepool.yaml.erb is where the nodepool stuff is bTW | 06:31 |
| *** jpeeler has quit IRC | 06:34 | |
| *** fungi has quit IRC | 06:35 | |
| *** fungi has joined #openstack-infra | 06:36 | |
| *** jpeeler has joined #openstack-infra | 06:37 | |
| *** sarob has quit IRC | 06:38 | |
| *** dstanek has quit IRC | 06:43 | |
| *** yamahata_ has joined #openstack-infra | 06:46 | |
| *** vkozhukalov has quit IRC | 06:47 | |
| *** denis_makogon has joined #openstack-infra | 06:48 | |
| *** rcleere has quit IRC | 06:53 | |
| *** summerlong is now known as slong-gone | 07:01 | |
| *** yamahata_ has quit IRC | 07:07 | |
| katyafervent | Hi! Do you guys know where chat log can be found? | 07:08 |
| *** bingbu has joined #openstack-infra | 07:09 | |
| *** sdake_ has quit IRC | 07:09 | |
| *** ryanpetrello has joined #openstack-infra | 07:13 | |
| *** ryanpetrello has quit IRC | 07:19 | |
| *** sarob has joined #openstack-infra | 07:22 | |
| *** DinaBelova has joined #openstack-infra | 07:24 | |
| *** vipul has quit IRC | 07:32 | |
| *** vipul has joined #openstack-infra | 07:33 | |
| *** jcooley_ has quit IRC | 07:35 | |
| *** jcooley_ has joined #openstack-infra | 07:36 | |
| *** yolanda has joined #openstack-infra | 07:38 | |
| *** zenoway has joined #openstack-infra | 07:42 | |
| *** nprivalova has joined #openstack-infra | 07:46 | |
| *** DennyZhang has quit IRC | 07:52 | |
| *** fbo_away is now known as fbo | 07:52 | |
| *** _david_ has joined #openstack-infra | 07:53 | |
| *** sarob has quit IRC | 07:55 | |
| *** _david_ has quit IRC | 07:59 | |
| *** flaper87|afk is now known as flaper87 | 08:03 | |
| *** avishayb has joined #openstack-infra | 08:12 | |
| *** osanchez has joined #openstack-infra | 08:14 | |
| *** pblaho has joined #openstack-infra | 08:20 | |
| *** sarob has joined #openstack-infra | 08:22 | |
| *** SergeyLukjanov has joined #openstack-infra | 08:23 | |
| *** che-arne has quit IRC | 08:25 | |
| *** vkozhukalov has joined #openstack-infra | 08:27 | |
| *** zenoway has quit IRC | 08:31 | |
| *** jcooley_ has quit IRC | 08:36 | |
| *** jcooley_ has joined #openstack-infra | 08:38 | |
| *** pblaho has quit IRC | 08:40 | |
| *** guohliu has joined #openstack-infra | 08:41 | |
| *** denis_makogon has quit IRC | 08:42 | |
| *** dizquierdo has joined #openstack-infra | 08:42 | |
| *** sjing has quit IRC | 08:49 | |
| *** guohliu has quit IRC | 08:49 | |
| *** ArxCruz has joined #openstack-infra | 08:50 | |
| *** jorisroovers has joined #openstack-infra | 08:53 | |
| *** nati_ueno has quit IRC | 08:56 | |
| *** sarob has quit IRC | 08:56 | |
| *** derekh has joined #openstack-infra | 08:57 | |
| *** dpyzhov has quit IRC | 08:58 | |
| *** guohliu has joined #openstack-infra | 09:01 | |
| *** jpich has joined #openstack-infra | 09:02 | |
| *** boris-42 has joined #openstack-infra | 09:03 | |
| *** jroovers has joined #openstack-infra | 09:05 | |
| *** jorisroovers has quit IRC | 09:05 | |
| *** Ryan_Lane has joined #openstack-infra | 09:08 | |
| *** yassine has joined #openstack-infra | 09:10 | |
| *** avishayb has quit IRC | 09:10 | |
| openstackgerrit | Tom Fifield proposed a change to openstack-infra/jeepyb: Add script to post a gerrit welcome message https://review.openstack.org/58900 | 09:11 |
| *** hashar has joined #openstack-infra | 09:12 | |
| *** ryanpetrello has joined #openstack-infra | 09:16 | |
| *** ryanpetrello has quit IRC | 09:20 | |
| *** sarob has joined #openstack-infra | 09:22 | |
| *** ilyashakhat_ has quit IRC | 09:31 | |
| *** dizquierdo has quit IRC | 09:32 | |
| *** ilyashakhat has quit IRC | 09:32 | |
| *** jroovers has quit IRC | 09:32 | |
| *** jcooley_ has quit IRC | 09:34 | |
| openstackgerrit | Sergey Lukjanov proposed a change to openstack/requirements: Bump python-savannaclient version to >= 0.4.0 https://review.openstack.org/59657 | 09:39 |
| *** che-arne has joined #openstack-infra | 09:42 | |
| *** markmc has joined #openstack-infra | 09:48 | |
| *** pblaho has joined #openstack-infra | 09:48 | |
| *** kushi_ has joined #openstack-infra | 09:50 | |
| *** matsuhashi has quit IRC | 09:51 | |
| *** kui has quit IRC | 09:52 | |
| *** bingbu has quit IRC | 09:53 | |
| *** matsuhashi has joined #openstack-infra | 09:54 | |
| *** kushi_ has quit IRC | 09:55 | |
| *** sarob has quit IRC | 09:55 | |
| jog0 | clarkb: both large ops seem to be failing | 09:56 |
| jog0 | we have a bug for it https://bugs.launchpad.net/neutron/+bug/1232303 | 09:58 |
| uvirtbot | Launchpad bug 1232303 in tempest "FAIL: tempest.scenario.test_large_ops.TestLargeOpsScenario.test_large_ops_scenario[compute,image]" [Undecided,New] | 09:58 |
| *** jishaomei has joined #openstack-infra | 10:02 | |
| openstackgerrit | A change was merged to openstack-infra/elastic-recheck: Update job names https://review.openstack.org/59573 | 10:04 |
| *** jcooley_ has joined #openstack-infra | 10:05 | |
| *** ilyashakhat has joined #openstack-infra | 10:06 | |
| openstackgerrit | Yuuichi Fujioka proposed a change to openstack-dev/hacking: cheking for metaclass to be Python 3.x compatible https://review.openstack.org/56890 | 10:08 |
| *** dizquierdo has joined #openstack-infra | 10:09 | |
| *** jishaomei has quit IRC | 10:09 | |
| *** jcooley_ has quit IRC | 10:12 | |
| *** boris-42 has quit IRC | 10:13 | |
| *** nprivalova has quit IRC | 10:20 | |
| *** marun has joined #openstack-infra | 10:20 | |
| *** pblaho has quit IRC | 10:21 | |
| *** pblaho has joined #openstack-infra | 10:22 | |
| *** sarob has joined #openstack-infra | 10:22 | |
| *** Ryan_Lane has quit IRC | 10:23 | |
| *** yamahata_ has joined #openstack-infra | 10:26 | |
| *** wenlock_ has quit IRC | 10:26 | |
| *** rongze has joined #openstack-infra | 10:28 | |
| *** giulivo has joined #openstack-infra | 10:28 | |
| *** fbo is now known as fbo_away | 10:28 | |
| giulivo | guys , need some pointer to upstream code or someone who can volunteer some of his time explaining me | 10:28 |
| giulivo | for which jobs we use the various [sections] in the tempest's tox.ini file | 10:29 |
| giulivo | specifically, when is that we run testenv:testr-full when testenv:all and when testenv:smoke | 10:30 |
| giulivo | ? | 10:30 |
| *** jcoufal has quit IRC | 10:34 | |
| *** dizquierdo has quit IRC | 10:37 | |
| *** guohliu has quit IRC | 10:42 | |
| *** nprivalova has joined #openstack-infra | 10:42 | |
| *** yamahata_ has quit IRC | 10:43 | |
| *** rongze has quit IRC | 10:46 | |
| *** avishayb has joined #openstack-infra | 10:48 | |
| *** markmcclain has joined #openstack-infra | 10:49 | |
| *** lcestari has joined #openstack-infra | 10:51 | |
| *** Ryan_Lane has joined #openstack-infra | 10:53 | |
| *** sarob has quit IRC | 10:55 | |
| *** zoresvit has quit IRC | 10:56 | |
| *** weshay has joined #openstack-infra | 10:58 | |
| *** Ryan_Lane has quit IRC | 10:58 | |
| *** markmcclain has quit IRC | 11:00 | |
| *** markmcclain has joined #openstack-infra | 11:00 | |
| *** rongze has joined #openstack-infra | 11:03 | |
| *** jcooley_ has joined #openstack-infra | 11:05 | |
| *** jcooley_ has quit IRC | 11:10 | |
| *** DinaBelova has quit IRC | 11:11 | |
| *** weshay has quit IRC | 11:11 | |
| *** wenlock has quit IRC | 11:14 | |
| *** weshay has joined #openstack-infra | 11:16 | |
| *** ryanpetrello has joined #openstack-infra | 11:17 | |
| *** kui has joined #openstack-infra | 11:17 | |
| *** weshay has quit IRC | 11:17 | |
| *** ryanpetrello has quit IRC | 11:22 | |
| *** sarob has joined #openstack-infra | 11:22 | |
| *** Ryan_Lane has joined #openstack-infra | 11:24 | |
| matel | Hi, anyone fancy a review? This change has been in the queue since september, and is needed to get devstack passing: https://review.openstack.org/#/c/48092/5 | 11:25 |
| matel | A change that depends on it has been approved. | 11:26 |
| *** boris-42 has joined #openstack-infra | 11:27 | |
| *** dizquierdo has joined #openstack-infra | 11:28 | |
| *** Ryan_Lane has quit IRC | 11:29 | |
| *** jcoufal has joined #openstack-infra | 11:31 | |
| *** rongze has quit IRC | 11:39 | |
| *** fifieldt has quit IRC | 11:40 | |
| openstackgerrit | Sergey Lukjanov proposed a change to openstack-infra/config: Setup devstack-gate tests for Savanna https://review.openstack.org/57317 | 11:50 |
| *** jcooley_ has joined #openstack-infra | 11:51 | |
| *** adalbas has joined #openstack-infra | 11:53 | |
| *** rongze_ has joined #openstack-infra | 11:56 | |
| *** sarob has quit IRC | 11:56 | |
| *** dims has joined #openstack-infra | 11:58 | |
| *** DinaBelova has joined #openstack-infra | 12:02 | |
| *** yaguang has quit IRC | 12:04 | |
| *** boris-42_ has joined #openstack-infra | 12:05 | |
| *** DinaBelova has quit IRC | 12:05 | |
| *** boris-42 has quit IRC | 12:08 | |
| *** CaptTofu has quit IRC | 12:08 | |
| *** CaptTofu has joined #openstack-infra | 12:08 | |
| *** bingbu has joined #openstack-infra | 12:13 | |
| *** bingbu has quit IRC | 12:19 | |
| *** sarob has joined #openstack-infra | 12:22 | |
| *** crank has quit IRC | 12:24 | |
| *** jcooley_ has quit IRC | 12:25 | |
| *** boris-42_ has quit IRC | 12:27 | |
| *** crank has joined #openstack-infra | 12:30 | |
| *** markmcclain has quit IRC | 12:30 | |
| *** beagles has joined #openstack-infra | 12:31 | |
| *** markmcclain has joined #openstack-infra | 12:32 | |
| *** boris-42 has joined #openstack-infra | 12:33 | |
| *** crank has quit IRC | 12:35 | |
| *** crank has joined #openstack-infra | 12:36 | |
| *** markmcclain has quit IRC | 12:36 | |
| *** markmcclain has joined #openstack-infra | 12:40 | |
| *** crank has quit IRC | 12:41 | |
| *** crank has joined #openstack-infra | 12:48 | |
| openstackgerrit | sebastian marcet proposed a change to openstack-infra/config: Review and create Puppet configuration for dev server https://review.openstack.org/53644 | 12:51 |
| *** smarcet has joined #openstack-infra | 12:52 | |
| *** adalbas has quit IRC | 12:52 | |
| sdague | lifeless: when you get up, I'm trying to understand the config_drive vs. metadata issue, and want to walk through the code on what's not hit in nova to understand what we should be doing here | 12:54 |
| *** sarob has quit IRC | 12:55 | |
| *** wenlock has joined #openstack-infra | 12:55 | |
| *** wenlock has quit IRC | 12:56 | |
| *** fbo_away is now known as fbo | 12:56 | |
| *** wenlock has joined #openstack-infra | 12:57 | |
| *** dkranz has joined #openstack-infra | 12:58 | |
| smarcet | hi , please review https://review.openstack.org/#/c/53644, thx | 12:58 |
| *** yaguang has joined #openstack-infra | 12:58 | |
| *** SpamapS has quit IRC | 12:59 | |
| *** bingbu has joined #openstack-infra | 12:59 | |
| *** ryanpetrello has joined #openstack-infra | 13:02 | |
| *** jcooley_ has joined #openstack-infra | 13:08 | |
| *** yamahata_ has joined #openstack-infra | 13:08 | |
| *** dstanek has joined #openstack-infra | 13:08 | |
| *** adalbas has joined #openstack-infra | 13:09 | |
| openstackgerrit | Nikita Konovalov proposed a change to openstack-infra/storyboard: Added basic error handling https://review.openstack.org/59706 | 13:10 |
| *** dolphm has joined #openstack-infra | 13:11 | |
| *** jergerber has joined #openstack-infra | 13:11 | |
| *** jcooley_ has quit IRC | 13:13 | |
| *** CaptTofu has quit IRC | 13:15 | |
| *** dolphm has quit IRC | 13:16 | |
| *** CaptTofu has joined #openstack-infra | 13:16 | |
| *** weshay has joined #openstack-infra | 13:18 | |
| *** dims has quit IRC | 13:21 | |
| *** sarob has joined #openstack-infra | 13:22 | |
| *** jcooley_ has joined #openstack-infra | 13:23 | |
| *** sandywalsh has quit IRC | 13:24 | |
| *** dims has joined #openstack-infra | 13:25 | |
| *** Ryan_Lane has joined #openstack-infra | 13:26 | |
| *** dolphm has joined #openstack-infra | 13:28 | |
| *** jcooley_ has quit IRC | 13:28 | |
| openstackgerrit | Dzmitry Horbach proposed a change to openstack-infra/jenkins-job-builder: Corrected typo in example for pipeline plugin https://review.openstack.org/59709 | 13:28 |
| *** zul has quit IRC | 13:29 | |
| *** zul has joined #openstack-infra | 13:31 | |
| *** Ryan_Lane has quit IRC | 13:31 | |
| *** cyril__ has quit IRC | 13:34 | |
| *** cyril__ has joined #openstack-infra | 13:35 | |
| *** sandywalsh has joined #openstack-infra | 13:36 | |
| *** kui has quit IRC | 13:37 | |
| *** dkliban has joined #openstack-infra | 13:37 | |
| *** matsuhashi has quit IRC | 13:40 | |
| *** thomasem has joined #openstack-infra | 13:40 | |
| *** inkerra has joined #openstack-infra | 13:41 | |
| *** aguzikov1 has joined #openstack-infra | 13:44 | |
| ekarlso | can zuul be used to fire off say tests towards diff versions of a compiler in parallell in different slaves ? | 13:45 |
| *** bingbu has quit IRC | 13:46 | |
| *** DinaBelova has joined #openstack-infra | 13:55 | |
| *** sarob has quit IRC | 13:55 | |
| *** rcleere has joined #openstack-infra | 13:58 | |
| *** aguzikova has joined #openstack-infra | 13:58 | |
| *** dprince has joined #openstack-infra | 13:59 | |
| *** SergeyLukjanov has quit IRC | 14:00 | |
| *** aguzikova has quit IRC | 14:00 | |
| *** aguzikova has joined #openstack-infra | 14:00 | |
| *** pblaho has quit IRC | 14:01 | |
| *** julim has joined #openstack-infra | 14:01 | |
| mordred | ekarlso: yes | 14:01 |
| *** inkerra__ has joined #openstack-infra | 14:02 | |
| *** avishayb has quit IRC | 14:03 | |
| *** inkerra__ has quit IRC | 14:03 | |
| *** inkerra_ has joined #openstack-infra | 14:03 | |
| *** DinaBelova has quit IRC | 14:03 | |
| mordred | morning sdague and fungi | 14:04 |
| *** inkerra_ has quit IRC | 14:04 | |
| *** aguzikova has quit IRC | 14:04 | |
| *** reaper has joined #openstack-infra | 14:04 | |
| *** aguzikov1 has quit IRC | 14:04 | |
| *** inkerra has quit IRC | 14:05 | |
| *** bpokorny1 has joined #openstack-infra | 14:05 | |
| *** inkerra has joined #openstack-infra | 14:05 | |
| *** inkerra has quit IRC | 14:08 | |
| sdague | mordred: morning | 14:08 |
| *** inkerra has joined #openstack-infra | 14:09 | |
| *** mriedem has joined #openstack-infra | 14:10 | |
| *** pabelanger_ has joined #openstack-infra | 14:10 | |
| mordred | sdague: I miss anyhting exciting? | 14:10 |
| *** avishayb has joined #openstack-infra | 14:10 | |
| *** markmcclain has quit IRC | 14:10 | |
| sdague | not that I know of, but I was writing sieve rules yesterday afternoon, so not really paying attension | 14:11 |
| *** pabelanger_ has quit IRC | 14:12 | |
| *** pabelanger_ has joined #openstack-infra | 14:12 | |
| *** pabelanger has quit IRC | 14:12 | |
| *** pabelanger_ is now known as pabelanger | 14:12 | |
| *** pabelanger_ has joined #openstack-infra | 14:13 | |
| *** dizquierdo has quit IRC | 14:13 | |
| *** inkerra_ has joined #openstack-infra | 14:14 | |
| openstackgerrit | Doug Hellmann proposed a change to openstack-infra/config: Move stevedore to stackforge https://review.openstack.org/58389 | 14:14 |
| *** inkerra_ has quit IRC | 14:15 | |
| *** dstanek has quit IRC | 14:15 | |
| *** krtaylor has quit IRC | 14:15 | |
| *** pabelanger has quit IRC | 14:16 | |
| *** pabelanger has joined #openstack-infra | 14:16 | |
| *** jcooley_ has joined #openstack-infra | 14:17 | |
| *** dkranz has quit IRC | 14:18 | |
| *** nosnos has quit IRC | 14:20 | |
| *** jcooley_ has quit IRC | 14:21 | |
| *** sarob has joined #openstack-infra | 14:22 | |
| *** dizquierdo has joined #openstack-infra | 14:25 | |
| fungi | howdy | 14:27 |
| *** rongze_ has quit IRC | 14:27 | |
| * fungi works back through the numerous scrollback questions first | 14:28 | |
| *** reaper has quit IRC | 14:28 | |
| fungi | katyafervent: by chat logs i assume you mean the irc log for this channel. if so, http://eavesdrop.openstack.org/irclogs/%23openstack-infra/ | 14:29 |
| fungi | giulivo: where do you need tempest's tox.ini file explained? i can point you to the jenkins jobs corresponding to tox commands calling different sections (environments), but anything tempest-specific is better covered in #openstack-qa | 14:30 |
| giulivo | fungi! thanks for helping | 14:31 |
| *** CaptTofu has quit IRC | 14:31 | |
| *** mfer has joined #openstack-infra | 14:31 | |
| giulivo | I think I'm okay with the tempest side but I'm trying to figure by which jenkins jobs get spawned the different testenv | 14:31 |
| *** CaptTofu has joined #openstack-infra | 14:31 | |
| fungi | matel: if you're trying to find core reviewers for devstack, you'll get better results asking in #openstack-qa | 14:31 |
| *** krtaylor has joined #openstack-infra | 14:33 | |
| fungi | giulivo: i can point you to where that configuration is broken out. just a sec | 14:33 |
| *** salv-orlando has quit IRC | 14:33 | |
| fungi | smarcet: i'll retest your latest patchset on 53644 shortly and see if that hopefully gets us closer to being able to puppet a server with it (at which point you'll be able to iterate a little more rapidly once you can get to it in our environment) | 14:36 |
| smarcet | fungi: thx a lot :) | 14:36 |
| *** rongze has joined #openstack-infra | 14:38 | |
| fungi | giulivo: the main switch (well, elif series which would probably work better as a switch) you want is at http://git.openstack.org/cgit/openstack-infra/devstack-gate/tree/devstack-vm-gate.sh#n294 | 14:38 |
| mordred | dhellmann: ooh, neat (re stevedore change) | 14:38 |
| giulivo | fungièè | 14:38 |
| giulivo | doh fungi++ | 14:38 |
| fungi | giulivo: that correlates different vars from the calling environment to the various tox environments (sections) | 14:38 |
| fungi | giulivo: then you can see here where the jenkins job templates are defined, and what envvars get passed to the script from each... http://git.openstack.org/cgit/openstack-infra/config/tree/modules/openstack_project/files/jenkins_job_builder/config/devstack-gate.yaml | 14:39 |
| mordred | fungi: can I be helpful there? | 14:39 |
| fungi | mordred: by all means! | 14:39 |
| mordred | fungi: (on the smarcet change) | 14:40 |
| fungi | mordred: please. review it, test it, anythinog | 14:40 |
| mordred | fungi: what if I just approve it without looking at it? is that helpful? | 14:40 |
| fungi | mordred: mmm... probably slightly less so | 14:40 |
| *** alcabrera has joined #openstack-infra | 14:40 | |
| mordred | ok. good to know where the line is | 14:40 |
| *** amotoki has quit IRC | 14:41 | |
| mordred | fungi: you know - when I look at this change, it makes me think that perhaps we should really be writing more facter things | 14:41 |
| fungi | mordred: though i think it's close to being able to spin up the dev openid server (smarcet has done great). just working through testing the redis module he wrote | 14:41 |
| mordred | like, I feel like our collective dislike of ruby may have led us to do some things not in facter that could probably be done there more sensibly | 14:42 |
| fungi | i would buy that assertion | 14:42 |
| giulivo | fungi, I see there is a "structured" naming scheme there as jobs suffixed with -full run testenv:full | 14:44 |
| giulivo | thanks | 14:44 |
| mordred | smarcet: I have questions... | 14:44 |
| mordred | smarcet: mainly owing to not doing much with facter myself | 14:44 |
| mordred | smarcet: you have a facter thing, which looks like an excellent thing and something I should use more | 14:44 |
| smarcet | mordred: sure tell me | 14:44 |
| mordred | which tells the version of the server | 14:44 |
| mordred | but then the redis class takes version as an input param and switches on that | 14:44 |
| smarcet | mordred: yes but for some reason its not working on your servers | 14:44 |
| mordred | how does redis_version get used? | 14:44 |
| smarcet | i already tested on my local env and works | 14:45 |
| mordred | ah. so the facter thing doesn't get used, because of $reasons - which is why we have explicit version | 14:45 |
| smarcet | however, i would remove that custom fact | 14:45 |
| mordred | ? | 14:45 |
| smarcet | yes | 14:45 |
| mordred | awesome | 14:45 |
| *** sarob has quit IRC | 14:45 | |
| mordred | well, I'm sad that it doesn't work on our servers | 14:45 |
| mordred | because it looks really slick | 14:45 |
| mordred | but that makes sense | 14:45 |
| *** johnthetubaguy has joined #openstack-infra | 14:45 | |
| fungi | i suspect it's a bootstrapping problem... it may be either getting called while there is no redis installed? | 14:45 |
| smarcet | i put already put a variable there just to explicit the redis version | 14:45 |
| mordred | I'm going to blame bodepd | 14:46 |
| smarcet | fungi: not sure | 14:46 |
| fungi | (or at some other point where the version can't be determined) | 14:46 |
| smarcet | tested on a raw ubuntu server | 14:46 |
| smarcet | and worked like a charm | 14:46 |
| mordred | fungi: I know that I have had issues trying to deal with ruby plugins in our setup | 14:46 |
| mordred | fungi: I tihn kthere is something weird with caching | 14:46 |
| mordred | have we tried this since clarkb landed the change to touch files on the puppetmaster? | 14:46 |
| fungi | smarcet: just for some background, i'm hitting that error when trying to apply the puppet manifest to a completely unconfigured/fresh vm, so there could be all sorts of stuff not installed yet if the dependencies aren't properly defined | 14:47 |
| mordred | there was a thing bodepd pointed out about how puppetmaster caches things | 14:47 |
| fungi | mordred: i think so--i last tested it yesterday | 14:47 |
| *** dizquierdo has quit IRC | 14:47 | |
| mordred | ok. | 14:47 |
| mordred | then yes | 14:47 |
| *** nprivalova has quit IRC | 14:47 | |
| *** dolphm has quit IRC | 14:48 | |
| smarcet | fungi: i tested on a fresh ubuntu server | 14:48 |
| *** ruhe has joined #openstack-infra | 14:49 | |
| fungi | smarcet: okay, perfect. i'll go ahead and try the latest patchset now | 14:50 |
| smarcet | fungi: ok cool :) | 14:51 |
| openstackgerrit | Shantanu Tushar proposed a change to openstack-infra/devstack-gate: Adds retry and timeout to git_fetch_at_ref Closes-Bug: 1252137 https://review.openstack.org/59742 | 14:54 |
| *** jamesmcarthur has joined #openstack-infra | 14:54 | |
| *** esker has joined #openstack-infra | 14:55 | |
| *** sarob has joined #openstack-infra | 14:55 | |
| fungi | smarcet: out of curiosity, what amount of memory should we be targeting with this server? will a 2gb vm be sufficient for dev or does it need more/less? | 14:56 |
| *** dizquierdo has joined #openstack-infra | 14:57 | |
| smarcet | fungi:well it depends, if server will host also db server , i think that 4gb would me more suitable, if not, 2gb will be enough i think | 14:57 |
| *** dolphm has joined #openstack-infra | 14:58 | |
| fungi | smarcet: awesome--thanks! 2gb then, since the db is in a separate 2gb ram trove instance anyway | 14:58 |
| *** dolphm has quit IRC | 14:58 | |
| smarcet | cool | 14:58 |
| *** krtaylor has quit IRC | 14:58 | |
| *** dolphm has joined #openstack-infra | 14:59 | |
| *** sarob has quit IRC | 15:00 | |
| *** afazekas has joined #openstack-infra | 15:01 | |
| *** dstanek has joined #openstack-infra | 15:03 | |
| *** Loquacity has quit IRC | 15:04 | |
| *** Loquacity has joined #openstack-infra | 15:04 | |
| fungi | smarcet: looks like fixing that problem exposed an unresolved dependency elsewhere in openid_dev.pp. i'll mark it in the review so you have a bit more context to go on | 15:09 |
| smarcet | fungi: ok tnx | 15:09 |
| openstackgerrit | Shantanu Tushar proposed a change to openstack-infra/devstack-gate: Adds retry and timeout to git_fetch_at_ref https://review.openstack.org/59742 | 15:10 |
| *** prad has joined #openstack-infra | 15:11 | |
| fungi | smarcet: i'll make that change on my local checkout and see if i can spot anything else breaking | 15:11 |
| *** rfolco has quit IRC | 15:12 | |
| smarcet | fungi: roger, and thx again | 15:14 |
| *** nprivalova has joined #openstack-infra | 15:16 | |
| *** jcooley_ has joined #openstack-infra | 15:17 | |
| *** jcoufal has quit IRC | 15:17 | |
| *** dkranz has joined #openstack-infra | 15:18 | |
| *** avishayb has quit IRC | 15:19 | |
| *** rcleere has quit IRC | 15:19 | |
| *** jcooley_ has quit IRC | 15:21 | |
| *** sarob has joined #openstack-infra | 15:22 | |
| fungi | smarcet: i'd hold off on updating it yet, because i think that require may need to be replaced with a different one (hit another unsatisfied requirement deeper inside the mysql::db class manifest) | 15:22 |
| fungi | i'll have it identified here in a bit | 15:22 |
| smarcet | fungi : roger that | 15:24 |
| *** Ryan_Lane has joined #openstack-infra | 15:24 | |
| *** Ryan_Lane has quit IRC | 15:29 | |
| *** mfer has quit IRC | 15:32 | |
| *** rfolco has joined #openstack-infra | 15:33 | |
| *** krtaylor has joined #openstack-infra | 15:35 | |
| *** datsun180b has joined #openstack-infra | 15:35 | |
| *** mfer has joined #openstack-infra | 15:35 | |
| *** ruhe has quit IRC | 15:36 | |
| *** dolphm has quit IRC | 15:37 | |
| *** ruhe has joined #openstack-infra | 15:38 | |
| *** dcramer__ has joined #openstack-infra | 15:38 | |
| *** rcleere has joined #openstack-infra | 15:40 | |
| *** dolphm has joined #openstack-infra | 15:41 | |
| jeblair | fungi, mordred, clarkb: please review https://review.openstack.org/#/c/58841/ we're waiting on it to remove reverify. | 15:43 |
| mordred | looking | 15:43 |
| *** vkozhukalov has quit IRC | 15:45 | |
| fungi | smarcet: got that mysql class dependency figured out and uncovered what i think is a service typo in the redis module now. testing with my proposed correction to see if it works | 15:47 |
| *** datsun180b is now known as datsun_F40PH | 15:47 | |
| smarcet | fungi: roger, thx | 15:47 |
| fungi | smarcet: but i'm about to head into a meeting, so will be somewhat afk over the next hour | 15:47 |
| dolphm | fun question... the global requirements update job is actually closing a bug in keystone, so a revised the commit message on the automated patch to include a Closes-Bug... | 15:48 |
| dolphm | A) if jenkins uploads a new patchset, will the current commit message be preserved? | 15:48 |
| *** datsun_F40PH is now known as datsun180b | 15:48 | |
| smarcet | fungi: sure np , and thx again | 15:48 |
| dolphm | B) if A, will the universe explode? | 15:48 |
| *** yaguang has quit IRC | 15:49 | |
| dolphm | s/so a revised/so I revised/ | 15:49 |
| *** mrodden has quit IRC | 15:50 | |
| dolphm | i assume hudson-openstack will reassign the bug to itself if that happens | 15:50 |
| *** SpamapS has joined #openstack-infra | 15:51 | |
| ekarlso | is there a way to see the test logs when using testr ? | 15:51 |
| *** sarob has quit IRC | 15:55 | |
| *** reed has joined #openstack-infra | 15:57 | |
| *** reaper has joined #openstack-infra | 16:01 | |
| openstackgerrit | sebastian marcet proposed a change to openstack-infra/config: Review and create Puppet configuration for dev server https://review.openstack.org/53644 | 16:02 |
| *** dkranz has quit IRC | 16:02 | |
| *** dcramer__ has quit IRC | 16:02 | |
| *** primemin1sterp is now known as primeministerp | 16:02 | |
| *** jcooley_ has joined #openstack-infra | 16:04 | |
| *** ^d has joined #openstack-infra | 16:05 | |
| *** jamesmcarthur has quit IRC | 16:05 | |
| *** sdake_ has joined #openstack-infra | 16:06 | |
| *** sdake_ has quit IRC | 16:07 | |
| *** sdake_ has joined #openstack-infra | 16:07 | |
| *** sdake_ has quit IRC | 16:07 | |
| *** sdake_ has joined #openstack-infra | 16:07 | |
| *** jamesmcarthur has joined #openstack-infra | 16:08 | |
| openstackgerrit | sebastian marcet proposed a change to openstack-infra/config: Review and create Puppet configuration for dev server https://review.openstack.org/53644 | 16:08 |
| *** rnirmal has joined #openstack-infra | 16:08 | |
| *** dkranz has joined #openstack-infra | 16:17 | |
| *** mrodden has joined #openstack-infra | 16:18 | |
| *** jcooley_ has quit IRC | 16:19 | |
| openstackgerrit | Roman Prykhodchenko proposed a change to openstack-infra/config: Adds devstack-gate tests for Ironic https://review.openstack.org/53917 | 16:19 |
| *** jcooley_ has joined #openstack-infra | 16:19 | |
| *** jcooley_ has quit IRC | 16:20 | |
| *** sarob has joined #openstack-infra | 16:22 | |
| *** nicedice has joined #openstack-infra | 16:23 | |
| pleia2 | good morning | 16:23 |
| *** hashar has quit IRC | 16:23 | |
| *** pete5 has joined #openstack-infra | 16:24 | |
| *** Ryan_Lane has joined #openstack-infra | 16:24 | |
| *** dprince has quit IRC | 16:25 | |
| *** romcheg has joined #openstack-infra | 16:26 | |
| romcheg | clarkb: Hi, are you around? | 16:26 |
| *** nprivalova has quit IRC | 16:28 | |
| *** Ryan_Lane has quit IRC | 16:29 | |
| jaypipes | jeblair, clarkb, fungi: question for you guys... so the chef+openstack community would like to add a post-merge hook that updates cookbook version strings (in metadata.rb files) and adds a Changelog entry automatically (either by git commit -a --amend'ing the commit or creating a separate "version commit"). Would the appropriate place to put code like that be the "post" Zuul pipeline, or would you recommend we go wi | 16:29 |
| jaypipes | th some other approach? | 16:29 |
| openstackgerrit | Ben Nemec proposed a change to openstack-dev/hacking: Enforce grouping like imports together https://review.openstack.org/54402 | 16:29 |
| openstackgerrit | Ben Nemec proposed a change to openstack-dev/hacking: Enforce import group ordering https://review.openstack.org/54403 | 16:29 |
| romcheg | Or maybe there is someone else who can help me with infra config | 16:31 |
| jeblair | jaypipes: i'd highly recommend working out a tag-based release system, like we use for client libs, etc. :) | 16:31 |
| *** atiwari has joined #openstack-infra | 16:31 | |
| jeblair | jaypipes: but if not, yeah, post is "after every commit lands" | 16:31 |
| jaypipes | jeblair: I totally agree :) unfortunately, in Chef-land, the version string in the metadata.rb file is what is used... :( | 16:32 |
| jeblair | jaypipes: (so such a job would have to ignore commits that it generated, otherwise you'd have a cycle) | 16:32 |
| *** danger_fo_away is now known as dangers | 16:32 | |
| jeblair | jaypipes: zaro has done some interesting work in that area... | 16:32 |
| jeblair | jaypipes: he has created tag-based release jobs for java libraries, where versions are in a pom.xml file | 16:32 |
| jaypipes | jeblair: I'd actually prefer to "inline amend" the commit if that is possible in one of the other pipelines... | 16:33 |
| jeblair | jaypipes: it's possible that it's similar enough you could adapt zaro's work. | 16:33 |
| mordred | jaypipes: if it's in metadata.rb ... couldn't you write ruby code in there which knows how to pull from tags? | 16:33 |
| mordred | yeah - what jeblair said | 16:33 |
| jaypipes | mordred: ah, that might actually work... hmmm... | 16:34 |
| romcheg | mordred: Do you have a minute? | 16:34 |
| mordred | romcheg: I do not - I'm on the phone :( | 16:34 |
| *** sparkycollier has joined #openstack-infra | 16:34 | |
| romcheg | mordred: Oh, sorry then | 16:34 |
| jaypipes | mordred: so if we were able to do that, then all we'd need to do is copy one of the existing git tag jobs and adapt to our needs, yeah? | 16:34 |
| mordred | jaypipes: yup | 16:34 |
| mordred | romcheg: I mean, feel free to ask me thing - | 16:35 |
| *** jcooley_ has joined #openstack-infra | 16:35 | |
| mordred | romcheg: I can respond to things a bit | 16:35 |
| *** sarob has quit IRC | 16:35 | |
| *** sarob has joined #openstack-infra | 16:35 | |
| *** marun has quit IRC | 16:37 | |
| *** pcrews has joined #openstack-infra | 16:37 | |
| romcheg | mordred: I will try then. I have made some changes to add two new jobs for Ironic: https://review.openstack.org/#/c/53917/7. I tried running Jenkins Job Builder and I found that it generates the appropriate jobs. However config-compare-xml says that Jenkins XML output is unchanged. | 16:38 |
| *** hashar has joined #openstack-infra | 16:38 | |
| romcheg | mordred: Is that a problem? | 16:38 |
| jeblair | romcheg: not a problem | 16:38 |
| romcheg | jeblair: Oh, thanks | 16:38 |
| jeblair | romcheg: there's a defect in the config compare job: it does not notice new jobs | 16:38 |
| romcheg | jeblair: Ah, that explains the things. Thank you | 16:39 |
| jaypipes | mordred: bah, that ain't gonna work... tooling like Berkshelf can't handle "code in metadata.rb that sets version string based on git tag". And, Matt Ray points out that once a cookbook is uploaded to a Chef server, it no longer is in a Git repository and all git tags are lost... | 16:39 |
| jaypipes | jeblair: got a link to zaro's work I could check out? :) | 16:40 |
| jeblair | jaypipes: just amin | 16:41 |
| mordred | jaypipes: heh. so, basically they have a git-based workflow that has an sdist-like step but that doesn't actually have the benefit of sdist artifact generation | 16:41 |
| jaypipes | mordred: yup :( | 16:41 |
| mordred | jaypipes: wow. what a great design </endsnark> | 16:42 |
| jaypipes | mordred: welcome to chef. :) | 16:42 |
| mordred | jaypipes: so what you'd want is a knife plugin I think | 16:42 |
| *** mattray has joined #openstack-infra | 16:42 | |
| mordred | jaypipes: that generates metadata.rb perhaps | 16:42 |
| mordred | ? | 16:42 |
| jaypipes | mordred: no, I don't think so... | 16:42 |
| mordred | ok | 16:42 |
| * mordred just trying to say chef words | 16:42 | |
| jaypipes | lol | 16:42 |
| openstackgerrit | Roman Prykhodchenko proposed a change to openstack-infra/config: Adds devstack-gate tests for Ironic https://review.openstack.org/53917 | 16:43 |
| jeblair | http://git.openstack.org/cgit/openstack-infra/config/tree/modules/openstack_project/files/jenkins_job_builder/config/jenkins-plugin-jobs.yaml | 16:45 |
| jeblair | http://git.openstack.org/cgit/openstack-infra/config/tree/modules/jenkins/files/slave_scripts/maven-properties.sh | 16:45 |
| jeblair | http://git.openstack.org/cgit/openstack-infra/gearman-plugin/tree/pom.xml | 16:45 |
| jeblair | jaypipes: ^ | 16:45 |
| katyafervent | fungi, Yes, thanks! I asked question about branch creation and missed the answer :) | 16:45 |
| jeblair | jaypipes: the rough overview is that the maven-properties.sh script generates a version number, and then it's passed to "mvn package" as a command line parameter, and that is used when generating the .war file | 16:46 |
| fungi | katyafervent: right, which project was it for? i can check the acl | 16:46 |
| *** jcooley_ has quit IRC | 16:47 | |
| jaypipes | jeblair: rock on. much appreciated. I suspect we'll need to do something similar, calling a ThorSCM ruby script instead of the maven-properties.sh script. Thanks again! | 16:47 |
| jeblair | jaypipes: no problem, good luck! | 16:48 |
| *** herndon has joined #openstack-infra | 16:49 | |
| *** yassine has quit IRC | 16:49 | |
| *** mestery has quit IRC | 16:49 | |
| *** mestery has joined #openstack-infra | 16:50 | |
| openstackgerrit | Matthew Treinish proposed a change to openstack-infra/config: Remove tempest coverage job https://review.openstack.org/59775 | 16:50 |
| openstackgerrit | Matthew Treinish proposed a change to openstack-infra/devstack-gate: Rip out the tempest coverage runs from devstack-gate https://review.openstack.org/59777 | 16:51 |
| *** gyee has joined #openstack-infra | 16:52 | |
| *** SpamapS has quit IRC | 16:52 | |
| *** DennyZhang has joined #openstack-infra | 16:52 | |
| romcheg | Guys, does that UNSTABLE in a lot of *-docs jobs mean that there is a problem? | 16:52 |
| romcheg | http://status.openstack.org/zuul/ | 16:52 |
| jeblair | romcheg: yes | 16:53 |
| *** moted has quit IRC | 16:53 | |
| jeblair | fungi: so how's the volume move on static coming? | 16:53 |
| *** ruhe has quit IRC | 16:53 | |
| fungi | jeblair: slowly. i've got two of the .5t pvs moved | 16:54 |
| *** hashar has quit IRC | 16:54 | |
| katyafervent | fungi, murano-docs. oh, I just found person who can see the create branch button. But I forget where this config file is located | 16:54 |
| fungi | pvmove takes hours to complete, so i keep nudging it one pv at a time | 16:54 |
| jeblair | fungi: any space we can allocate to docs-draft right now? | 16:54 |
| fungi | jeblair: definitely. i'll do that now | 16:54 |
| *** openstackstatus has joined #openstack-infra | 16:54 | |
| fungi | ahh, yeah. filled up it did | 16:55 |
| *** hashar has joined #openstack-infra | 16:55 | |
| *** moted has joined #openstack-infra | 16:55 | |
| EmilienM | someone from infra-core could approve a patch to add ironic puppet module into stackforge ? > https://review.openstack.org/#/c/59247/ | 16:56 |
| jeblair | #status alert docs jobs are failing due to a full filesystem; fix eta 1750 UTC | 16:56 |
| openstackstatus | NOTICE: docs jobs are failing due to a full filesystem; fix eta 1750 UTC | 16:56 |
| *** ChanServ changes topic to "docs jobs are failing due to a full filesystem; fix eta 1750 UTC" | 16:57 | |
| jeblair | fungi: logs needs some immediate help too. | 16:57 |
| *** marun has joined #openstack-infra | 16:57 | |
| fungi | jeblair: there's available space on docs-draft again now | 16:57 |
| fungi | doing logs | 16:58 |
| *** Ryan_Lane has joined #openstack-infra | 16:59 | |
| fungi | resize2fs sure takes a while when you edge up into the multi-tb range | 17:03 |
| *** guitarzan is now known as primemin1sterp | 17:03 | |
| *** primemin1sterp is now known as guitarzan | 17:04 | |
| mordred | fungi: maybe we should use reiser | 17:04 |
| pleia2 | mordred: yeah, with mongo | 17:04 |
| fungi | mordred: i did, for years. too bad it stagnated | 17:04 |
| fungi | mordred: itym "maybe we should use btrfs" | 17:04 |
| *** sarob has quit IRC | 17:04 | |
| *** SpamapS has joined #openstack-infra | 17:05 | |
| *** SpamapS has quit IRC | 17:05 | |
| *** SpamapS has joined #openstack-infra | 17:05 | |
| *** sarob has joined #openstack-infra | 17:05 | |
| *** hashar has quit IRC | 17:07 | |
| fungi | but yeah, back when reiserfs was pretty much the only option for online resizing, it was so useful | 17:08 |
| zaro | good morning | 17:09 |
| zaro | jaypipes: you interested in something about java? | 17:09 |
| *** dolphm has quit IRC | 17:10 | |
| jeblair | zaro: he was looking at doing tag-based versioning for ruby, so i pointed him at your java work because it may be similar enough it could be helpful | 17:10 |
| fungi | okay, breathing room added to docs-draft and logs filesystems on static.o.o, snmpd restarted | 17:12 |
| *** syerrapragada has quit IRC | 17:12 | |
| zaro | ok. help if i can, but i know nothing about ruby. | 17:12 |
| * fungi resumes shuffling physical volumes there | 17:12 | |
| *** jcooley_ has joined #openstack-infra | 17:12 | |
| *** UtahDave has joined #openstack-infra | 17:13 | |
| *** syerrapragada has joined #openstack-infra | 17:13 | |
| zaro | fungi: did you see the change to https://review.openstack.org/#/c/54639/ | 17:13 |
| *** marun has quit IRC | 17:13 | |
| jeblair | #status ok | 17:14 |
| *** ChanServ changes topic to "Discussion of OpenStack Project Infrastructure | Docs http://ci.openstack.org/ | Bugs https://launchpad.net/openstack-ci | Code https://git.openstack.org/cgit/openstack-infra/" | 17:14 | |
| fungi | zaro: yeah, saw you updated it. haven't circled back around to re-review since last night | 17:14 |
| fungi | thanks@ | 17:14 |
| *** jcooley_ has quit IRC | 17:15 | |
| *** Ryan_Lane has quit IRC | 17:15 | |
| zaro | fungi: opps, i meant the nodepool about the jenkins_dev_credentials_id https://review.openstack.org/#/c/57333/ | 17:15 |
| fungi | zaro: right that one | 17:16 |
| *** jcooley_ has joined #openstack-infra | 17:16 | |
| *** reed has quit IRC | 17:16 | |
| *** dstanek has quit IRC | 17:16 | |
| fungi | zaro: i'll try to get the hiera entries generated and added for the other two items here shortly | 17:16 |
| zaro | fungi: no wait. i made a mistake. that var is used. | 17:17 |
| fungi | okay | 17:18 |
| zaro | fungi: i'll have to revert. | 17:18 |
| fungi | i assume there's a way to generate it (an api credential for jenkins-dev, right?) | 17:18 |
| fungi | either from an api call or via the webui | 17:18 |
| zaro | i'm not sure. let me take a look to see if i can figure it out. | 17:20 |
| jeblair | fungi, clarkb, mordred: i'm going to single-core approve some stackforge changes. any objections? | 17:20 |
| *** jcooley_ has quit IRC | 17:20 | |
| fungi | jeblair: no objection from me | 17:20 |
| mordred | jeblair: do it | 17:20 |
| *** jcooley_ has joined #openstack-infra | 17:23 | |
| zaro | mgagne: got a minute? | 17:23 |
| mgagne | zaro: go | 17:23 |
| zaro | mgagne: i think either i'm wrong or i'm not getting the point across, https://review.openstack.org/#/c/57469 | 17:24 |
| openstackgerrit | A change was merged to openstack-infra/config: Remove rtfd-id from jenkins_jobs_builder https://review.openstack.org/58913 | 17:24 |
| *** markmc has quit IRC | 17:24 | |
| openstackgerrit | A change was merged to openstack-infra/config: Add docs and readthedocs.org support to solum https://review.openstack.org/58912 | 17:24 |
| mgagne | zaro: batch_task in a subentry in tasks | 17:24 |
| zaro | mgagne: does it need to be inside the for loop? | 17:25 |
| mgagne | zaro: just like this one: https://github.com/openstack-infra/jenkins-job-builder/blob/master/jenkins_jobs/modules/publishers.py#L978 | 17:25 |
| mgagne | zaro: from what I understand, batch_task is some kind of reference to the newly created element. You can't reuse it for a 2nd element. | 17:26 |
| zaro | mgagne: yes, that will work. but will it not work if it's outside the for loop? | 17:26 |
| mgagne | zaro: it looks to be a reference, updating the SAME element would update the previous one | 17:27 |
| openstackgerrit | yolanda.robla proposed a change to openstack-infra/jenkins-job-builder: Enable ignore_cache flag on jenkins_jobs.ini https://review.openstack.org/59515 | 17:27 |
| zaro | mgagne: ohh, i see. i got it now. | 17:28 |
| zaro | mgagne: thnx. | 17:28 |
| mgagne | zaro: np | 17:28 |
| *** krtaylor has quit IRC | 17:28 | |
| *** dcramer__ has joined #openstack-infra | 17:32 | |
| *** dcramer__ has quit IRC | 17:33 | |
| *** dcramer__ has joined #openstack-infra | 17:33 | |
| *** senk has joined #openstack-infra | 17:35 | |
| fungi | grr... i'm stuck on troubleshooting behavior for the mysql::db class in puppet. want to have it contain a copy of the account credentials but not actually attempt to create the account | 17:35 |
| jeblair | fungi: why do you want it to have a copy then? | 17:36 |
| fungi | jeblair: mmm, that's a very good point... presumably the variables can just get stuffed directly into the apps that need them | 17:36 |
| fungi | though the reason why this arises is... interesting. apparently it wants to create hostname-specific credentials in the db, and the hostnames rackspace hands out for trove databases are too long (i think according to mysql, though i guess it might be a puppet mysql module limitation instead) | 17:37 |
| *** boris-42 has quit IRC | 17:38 | |
| *** fbo is now known as fbo_away | 17:38 | |
| jeblair | fungi: isn't a hostname specifier in a mysql auth record indicating the _connecting_ hostname? | 17:38 |
| fungi | returned 1: ERROR 1470 (HY000) at line 1: String 'blahblahblah.rackspaceclouddb.com' is too long for host name (should be no longer than 60) | 17:38 |
| jeblair | fungi: (so, not the hostname of the trove db?) | 17:38 |
| fungi | oh, that's a good point | 17:39 |
| fungi | i wonder why it's trying to do that | 17:39 |
| jeblair | mordred: ^ (double check me?) | 17:39 |
| mordred | me look | 17:39 |
| fungi | err: /Stage[main]/Openstack_project::Openid_dev/Mysql::Db[openstackid_openid_dev | 17:40 |
| fungi | er, i'll use paste.o.o. stray newlines in this buffer | 17:40 |
| mordred | that's correct | 17:40 |
| jaypipes | zaro: have a moment to chat? | 17:40 |
| mordred | hostnames in auth db are for connecting from hosts | 17:40 |
| *** reed has joined #openstack-infra | 17:41 | |
| mordred | and we don't want puppet to _create_ databases for us - we just want to configure what creds it should use to connect to the one that's there | 17:41 |
| zaro | jaypipes: yup | 17:41 |
| *** derekh has quit IRC | 17:41 | |
| ekarlso | has the jeepyb stuff been landed ? | 17:42 |
| jaypipes | zaro: so, the chef+openstack community is trying to get something like this executed when a patch is merged into one of the repositories: https://gist.github.com/mattray/7683491 | 17:42 |
| fungi | mordred: jeblair: slightly more context... http://paste.openstack.org/show/54372/ | 17:42 |
| jaypipes | zaro: I know it's Ruby, but the gist of that is that a file (metadata.rb) has a version string replaced and a Changelog entry is generated. | 17:42 |
| jaypipes | zaro: and then git commit'd to the repository | 17:43 |
| mattray | zaro: it reads the git tag, increments it based on the commit message and writes it out to the metadata.rb and changelog | 17:43 |
| jaypipes | zaro: what we'd like to do is follow your maven example, only what I can't figure out is how you commit the change to the pom.xml file back to the openstack-infra/gearman-plugin project... | 17:43 |
| zaro | want to do this for every build? | 17:43 |
| jeblair | mattray, jaypipes: so one of the fundamental things about tag-based versioning is that we don't have commits for version increments | 17:43 |
| jaypipes | zaro: every commit to every cookbook in stackforge, yeah. | 17:44 |
| jaypipes | jeblair: yes, I don't want them either. | 17:44 |
| jaypipes | jeblair: I want to amend the existing commit if possible. | 17:44 |
| jeblair | mattray, jaypipes: let me amend my statement then -- the versioning is not in the code at all | 17:44 |
| jaypipes | jeblair: the only problem with that is ensuring that multiple patches queued for merging get the correct version. | 17:44 |
| jeblair | mattray, jaypipes: it's in either the tag history of the git repo, or in the release artifact | 17:45 |
| jaypipes | jeblair: yes, I understand that, but I can't change a deficiency in Chef unfortunatyely :( | 17:45 |
| jaypipes | jeblair: is pom.xml considered a release artifact? | 17:45 |
| zaro | typically yes. | 17:46 |
| jeblair | jaypipes: it's used to generate the metadata that are stored in the archive that is built upon release. | 17:46 |
| *** cyril__ has quit IRC | 17:46 | |
| jaypipes | jeblair: understood. | 17:46 |
| jaypipes | mattray: does Chef have anything like that? basically, the problem is that the metadata.rb is not a release artifact (generated from a release/sdist procedure). it's in the repository as code itself | 17:47 |
| jaypipes | jeblair: ^^ correct me if I am wrong in above statement... I think that sums up the issue, yes? | 17:47 |
| jeblair | jaypipes: i think so. it would be easy enough to re-write it upon building a release, but you'd still need something to handle the case that you are running or installing directly from a git repo. | 17:48 |
| mattray | the metadata.rb gets converted to a metadata.json file when uploaded to the server. Same difference as a pom.xml, it defines dependencies and versions | 17:48 |
| jaypipes | mattray: the difference is that other projects don't store version information in a file in the repo itself. they use git tags. | 17:50 |
| jeblair | mattray, jaypipes: oh, that sounds promising. you could look into opportunities to alter how metadata.json gets built. | 17:50 |
| jaypipes | jeblair: problem is that it is built from metadata.rb, not git tags ;) | 17:50 |
| jeblair | mattray, jaypipes: either hook into that, or modify it post-facto. | 17:50 |
| mattray | jaypipes: so require client-side tooling specific to StackForge :/ | 17:50 |
| jaypipes | mattray: specific to git, but yeah.... | 17:51 |
| reed | should https://wiki.openstack.org/wiki/Programs page list Telemetry instead of Metering? | 17:51 |
| mordred | mattray, jaypipes: but the things that buld metadata.json from metadata.rb - are those extensible? | 17:51 |
| jaypipes | mordred: I don't think so. | 17:51 |
| jaypipes | mattray: ? | 17:51 |
| reed | ttx, should https://wiki.openstack.org/wiki/Programs page list Telemetry instead of Metering? | 17:51 |
| fungi | i know this is probably crazy ugly, but what if a release artifact of the tagged commit were a second git repo, for clients who want to consume directly from git? | 17:51 |
| mattray | jaypipes: yeah, that's knife | 17:52 |
| mordred | jaypipes: I was right earlier! | 17:52 |
| jeblair | score 1 point for mordred! | 17:52 |
| mordred | mattray: so, in theory, a knife plugin that knew how to do different things with metadata.json | 17:52 |
| mordred | mattray: would be what we'd be looking for here, right? | 17:52 |
| jaypipes | mordred: the problem is all the Chef tooling like Berkshelf that reads the metadata.rb file as-is... | 17:52 |
| mattray | mordred: yeah, we've preferred to keep the tooling standard | 17:52 |
| mordred | ok. well, I agree with keeping tooling standard | 17:53 |
| mattray | I mean, we can do all sorts of crazy stuff on the client side | 17:53 |
| mordred | I'd love it if it were possible to declare a particular extension and have that be executed | 17:53 |
| mattray | but some users may not even use git | 17:53 |
| *** alcabrera is now known as alcabrera|afk | 17:53 | |
| mordred | right. | 17:53 |
| jaypipes | mattray: it's not about git, though... it's about storing the release artifact in the source tree. | 17:53 |
| mordred | so what you're saying is that chef really requires the versoin to be in metadata.rb and having it be computed is really hard | 17:54 |
| clarkb | morning | 17:54 |
| jaypipes | mordred: precvisely. | 17:54 |
| mordred | are we expecting peiple to consume chef-openstack repos not from git? | 17:54 |
| *** dkranz has quit IRC | 17:55 | |
| mordred | I know people can use chef without git | 17:55 |
| mordred | but do we expect to support people using chef-nova from not-git? | 17:55 |
| mattray | mordred: that is one of the goals | 17:55 |
| mordred | ok | 17:55 |
| *** vkozhukalov has joined #openstack-infra | 17:55 | |
| *** alcabrera|afk is now known as alcabrera | 17:55 | |
| mordred | in that case, I do not believe creating tag-based-versioning for chef is possible | 17:55 |
| mordred | and you're going to be stuck with making additional commits to the repo | 17:56 |
| *** jamesmcarthur has quit IRC | 17:56 | |
| mordred | for python- we have two different supported ways you can work | 17:56 |
| mattray | yeah, I'd kinda already accepted that. But it gets us past the conflicting merges on versions faster | 17:56 |
| mordred | either you use git | 17:56 |
| mordred | or you use a tarball which is built | 17:57 |
| jaypipes | mattray: so we expect folks to be able to git clone git@github.com:stackforge/cookbook-openstack-compute and then not use git? :( | 17:57 |
| mordred | the tarball can contain cached info from builtfrom git | 17:57 |
| *** jamesmcarthur has joined #openstack-infra | 17:57 | |
| mattray | jaypipes: if we publish the cookbooks on the community site, there's no use of git | 17:57 |
| mattray | jaypipes: of course, that could require the client-side change before the upload | 17:58 |
| jaypipes | mattray: explain... | 17:58 |
| *** BobBall is now known as BobBallAway | 17:58 | |
| *** rongze has quit IRC | 17:58 | |
| mattray | jaypipes: if we use git tags without modifying metadata.rb, we could require users to have a knife plugin or berksfile plugin that automatically modifies the metadata.rb locally. | 17:59 |
| mattray | jaypipes: Berkshelf understands git tags, so that's fine | 17:59 |
| mattray | jaypipes: and if we upload to the community site, whoever does that has to manually update the metadata.rb | 17:59 |
| jaypipes | mattray: and how would we then queue versions up properly? | 18:00 |
| mattray | jaypipes: ignore versioning completely in reviews? | 18:00 |
| mattray | and let the workstation-side update it? | 18:00 |
| mattray | seems potentially error prone and requires more out of developers | 18:00 |
| *** ArxCruz has quit IRC | 18:00 | |
| jaypipes | mattray: and how does the workstation know about other proposed commits, though... | 18:01 |
| jaypipes | mattray: that is what Zuul and the infrastructure here is for :) | 18:01 |
| mattray | jaypipes: it just reads the git log? | 18:01 |
| jaypipes | mattray: but then you'll have multiple people submitting patches with the exact same version in metadata.rb... | 18:01 |
| *** ArxCruz has joined #openstack-infra | 18:01 | |
| mattray | jaypipes: so what if we take metadata.rb and changelog out of git? | 18:01 |
| jaypipes | mattray: ++! | 18:02 |
| mattray | and make those generated? | 18:02 |
| jaypipes | yay! | 18:02 |
| mattray | changelog we could drop, tell people to read git log | 18:02 |
| jaypipes | they are build artifacts, after all... | 18:02 |
| mattray | just requires writing more client-side tooling | 18:02 |
| mattray | s/client/workstation/ | 18:03 |
| jaypipes | mattray: the only problem I see is that various Chef tooling like Berkshelf and ridley rely on metadata.rb existing and having the version identifier (and proper cookbook name, etc) | 18:03 |
| *** markmcclain has joined #openstack-infra | 18:03 | |
| mattray | jaypipes: berkshelf will pull a git tag, but yeah we may need a berkshelf plugin too :( | 18:03 |
| clarkb | mordred: still around? I am going to activate hpcloud region-a on the jenkins account too and put up a nodepool change to use that region as well | 18:05 |
| mattray | jaypipes: I'm going to have to think how we get this to all work | 18:05 |
| jaypipes | mattray: I think it would be best to move to a git tag-based versioning system, with metadata.rb possibly just containing some placeholder for the version. | 18:05 |
| mattray | jaypipes: I agree with that, can we get the git tag automatically incremented major/minor/patch? | 18:05 |
| mattray | read the git tag, push it into metadata.rb, upload to server | 18:06 |
| jaypipes | mattray: yup, already prior art for that. | 18:06 |
| clarkb | mordred: shouldn't be any opposition to that right? | 18:06 |
| mordred | clarkb: do it | 18:06 |
| mattray | jaypipes: parsing commit messages for #major, #minor? cool | 18:07 |
| jaypipes | mattray: sure, totally. | 18:07 |
| mattray | ok, I'll look at the problem again. | 18:07 |
| *** dkranz has joined #openstack-infra | 18:07 | |
| *** UtahDave has quit IRC | 18:08 | |
| openstackgerrit | Clark Boylan proposed a change to openstack-infra/config: Add HPCloud region-a to nodepool. https://review.openstack.org/59801 | 18:09 |
| clarkb | mordred: jeblair: fungi: I think we can probably just push that through since region-b seems to be mostly ok | 18:10 |
| clarkb | I can babysit | 18:10 |
| jeblair | clarkb: any chance you want to review and approve https://review.openstack.org/#/c/59572/ first? | 18:11 |
| clarkb | jeblair: sure looking | 18:11 |
| *** sandywalsh_ has joined #openstack-infra | 18:11 | |
| clarkb | jeblair: is the idea behind that to have the bare minimum root capable slave? | 18:12 |
| *** krtaylor has joined #openstack-infra | 18:12 | |
| clarkb | jeblair: were you going to babysit that change? | 18:13 |
| jeblair | clarkb: no, jenkins does not have root on a bare_slave. | 18:13 |
| clarkb | gotcha | 18:13 |
| clarkb | I see the commit message now (I really should read that first) | 18:13 |
| clarkb | approved | 18:13 |
| *** sandywalsh has quit IRC | 18:14 | |
| *** wabat has joined #openstack-infra | 18:14 | |
| *** nati_ueno has joined #openstack-infra | 18:15 | |
| *** vipul is now known as vipul-away | 18:15 | |
| jeblair | clarkb: likewise | 18:15 |
| openstackgerrit | A change was merged to openstack-infra/config: Add a bare precise node to nodepool https://review.openstack.org/59572 | 18:15 |
| openstackgerrit | A change was merged to openstack-infra/config: Add HPCloud region-a to nodepool. https://review.openstack.org/59801 | 18:16 |
| clarkb | woot thanks | 18:16 |
| openstackgerrit | Clark Boylan proposed a change to openstack-infra/jeepyb: Always sleep after group creation https://review.openstack.org/59583 | 18:17 |
| *** DennyZhang has quit IRC | 18:17 | |
| clarkb | so ^ fixes H7803 | 18:17 |
| clarkb | *H803 | 18:17 |
| openstackgerrit | A change was merged to openstack-infra/config: Add zuul-packaging to #openstack-infra https://review.openstack.org/59595 | 18:18 |
| clarkb | mordred: fungi: after the meeting I can babysit https://review.openstack.org/#/c/58653/1 if that change doesn't look too horrible to you guys | 18:18 |
| *** sdake has quit IRC | 18:19 | |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/jeepyb: Ignore E125 and H flake8 violations. https://review.openstack.org/59802 | 18:19 |
| *** sdake has joined #openstack-infra | 18:19 | |
| *** sdake has quit IRC | 18:19 | |
| *** sdake has joined #openstack-infra | 18:19 | |
| clarkb | romcheg: pong | 18:19 |
| jeblair | this ^ also fixes H803. | 18:19 |
| *** sdake has quit IRC | 18:20 | |
| *** sdake has joined #openstack-infra | 18:20 | |
| *** sdake has quit IRC | 18:20 | |
| *** sdake has joined #openstack-infra | 18:20 | |
| openstackgerrit | Khai Do proposed a change to openstack-infra/config: add nodepool to jenkins-dev server https://review.openstack.org/57333 | 18:22 |
| openstackgerrit | A change was merged to openstack-infra/config: Use purge option where appropriate. https://review.openstack.org/59553 | 18:24 |
| *** datsun180b is now known as datsun_F40PH | 18:25 | |
| clarkb | jeblair: double checking, but won't https://review.openstack.org/#/c/57098/4/modules/openstack_project/files/jenkins_job_builder/config/solum.yaml fail to match the single use slave regex? | 18:25 |
| pleia2 | jeblair: is there a file that has both violations I can run this tox.ini again to test? | 18:25 |
| clarkb | jeblair: the job should use devstack-vm of dsvm | 18:25 |
| *** ryanpetrello has quit IRC | 18:26 | |
| openstackgerrit | sebastian marcet proposed a change to openstack-infra/config: Review and create Puppet configuration for dev server https://review.openstack.org/53644 | 18:26 |
| *** markmcclain has quit IRC | 18:26 | |
| clarkb | pleia2: H803 is period at the end of commit message, E125 is well I thought I would have an example but paste.o.i is slow | 18:27 |
| clarkb | pleia2: finally http://paste.openstack.org/show/54374/ | 18:27 |
| pleia2 | clarkb: cool, thanks | 18:27 |
| *** jerryz has joined #openstack-infra | 18:27 | |
| jeblair | clarkb: you're right; it should be called 'gate-solum-devstack-dsvm' | 18:27 |
| clarkb | jeblair: thanks, leaving a comment now | 18:28 |
| jeblair | pleia2: i intentionally violated H803 in my commit for that change. | 18:28 |
| jeblair | commit msg | 18:29 |
| *** DinaBelova has joined #openstack-infra | 18:29 | |
| clarkb | jeblair: gate-solum-devstack-dsvm or gate-solum-dsvm ? | 18:29 |
| jeblair | clarkb: i don't feel strongly, but if i were naming it, i would go with gate-solum-devstack-dsvm because the test itself is "devstack" (eg, the exercises) | 18:30 |
| mordred | ++ | 18:30 |
| clarkb | ok | 18:30 |
| *** dizquierdo has quit IRC | 18:30 | |
| *** senk has quit IRC | 18:30 | |
| openstackgerrit | sebastian marcet proposed a change to openstack-infra/config: Review and create Puppet configuration for dev server https://review.openstack.org/53644 | 18:30 |
| fungi | smarcet: i'll give that one ^ a try. sorry about the churn on the earlier patchset--i hadn't finished trying fixes when i posted the first few comments | 18:31 |
| *** krotscheck has joined #openstack-infra | 18:31 | |
| *** vipul-away is now known as vipul | 18:32 | |
| *** SergeyLukjanov has joined #openstack-infra | 18:32 | |
| smarcet | fungi: np, althought i think that we would need to use https://forge.puppetlabs.com/puppetlabs/apt bc the php version ( need php 5.4) to add ppa ppa:ondrej/php5-oldstable | 18:32 |
| sdague | it's going to take me a while for dsvm to mean something to me :) | 18:33 |
| *** osanchez has quit IRC | 18:33 | |
| *** openstackgerrit has quit IRC | 18:33 | |
| *** openstackgerrit has joined #openstack-infra | 18:33 | |
| fungi | smarcet: i think we can probably accept using a ppa for a backport from a later ubuntu release for something like this if it's needed | 18:34 |
| jeblair | sdague: i wanted to get rid of it entirely, and hope too once all our slaves are dynamic. | 18:34 |
| openstackgerrit | Russell Bryant proposed a change to openstack-infra/config: Add gate-solum-devstack job https://review.openstack.org/57098 | 18:34 |
| fungi | sdague: my eyes still see it as "dsm-iv" (damn you, college psych class!) | 18:35 |
| sdague | heh | 18:35 |
| jeblair | sdague: er, so configdrive/metadata.... where did we get with that? | 18:36 |
| jeblair | sdague: do you want to test both? or only one? | 18:36 |
| sdague | so... actually, I want to wait for lifeless to get back on | 18:36 |
| sdague | because I was reading the relevant nova code, and now I feel like I want to understand more about exactly what we missed testing | 18:37 |
| fungi | zomg red hat marketing just spammed my personal e-mail address out of be blue. who do i hunt down and kill? | 18:37 |
| fungi | out of the blue | 18:37 |
| *** ruhe has joined #openstack-infra | 18:38 | |
| jeblair | sdague: ok. i'm going to WIP that change and indicate that i would like to know which things we plan to use, and also, you still have some unanswered questions. | 18:38 |
| sdague | sounds good | 18:38 |
| *** dprince has joined #openstack-infra | 18:39 | |
| openstackgerrit | A change was merged to openstack-infra/config: Add jenkins jobs for puppet-savanna https://review.openstack.org/59304 | 18:40 |
| clarkb | jeblair: russellb addressed the solum devstack gate change -1 | 18:40 |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/elastic-recheck: add support for installing the web dashboard https://review.openstack.org/59289 | 18:41 |
| *** sarob has quit IRC | 18:41 | |
| *** sarob has joined #openstack-infra | 18:41 | |
| fungi | we've got another split queue showing on the gate status page, fyi | 18:43 |
| *** herndon has quit IRC | 18:43 | |
| fungi | i expect it'll clear up once 54626,4 reports though | 18:44 |
| jeblair | sdague: i want to make sure to draw your attention to my comment on https://review.openstack.org/#/c/59289/ | 18:44 |
| fungi | and there it goes | 18:44 |
| jeblair | fungi: er, it's not fixed, it's just hidden. | 18:44 |
| fungi | ahh | 18:44 |
| clarkb | I think I approved the savanna devstack related cahnges in the wrong order, but those jobs aren't voting so hopefully not a problem | 18:44 |
| clarkb | jeblair: fungi: is that related to the rendering of data or is there actually a split gate? | 18:45 |
| SergeyLukjanov | oops, I was offline, am I missed something? | 18:45 |
| jeblair | clarkb: i believe there is actually a split gate. | 18:45 |
| russellb | clarkb: appreciate the help | 18:45 |
| *** sarob has quit IRC | 18:46 | |
| SergeyLukjanov | clarkb, jeblair, thanks for reviewing and merging https://review.openstack.org/#/c/57317/ | 18:46 |
| fungi | the last time this appeared, zuul logged during a config reload that it configured two gates with those separate lists of projects, and when we did another reload (restart?) it cleared up | 18:47 |
| jeblair | clarkb, fungi: I have requested a graceful zuul restart. | 18:47 |
| sdague | jeblair: yep, good point on config | 18:47 |
| fungi | sounds good | 18:47 |
| sdague | actually, is mordred active? | 18:47 |
| sdague | I had a pbr feature question | 18:47 |
| sdague | it would be really nice to have config_files= | 18:47 |
| sdague | like data_files= | 18:47 |
| sdague | except it does not overwrite | 18:47 |
| jeblair | sdague: one more step on the road to being a complete package manager... | 18:48 |
| sdague | so that we could seed a good approximation of /etc | 18:48 |
| sdague | yep | 18:48 |
| *** herndon_ has joined #openstack-infra | 18:48 | |
| sdague | well, it would be super handy | 18:48 |
| *** rongze has joined #openstack-infra | 18:49 | |
| jeblair | clarkb, fungi: i'll see if i can replicate the zuul bug; however, it's going to be _very_ difficult. | 18:50 |
| yolanda | hi jeblair, i saw your comment, do you mean create a new section in .ini file, or just list this arg as optional in documentation? | 18:51 |
| jeblair | yolanda: just documentation | 18:51 |
| yolanda | ok | 18:51 |
| mordred | sdague: I do not think that's a good feature | 18:52 |
| openstackgerrit | A change was merged to openstack-infra/config: Setup devstack-gate tests for Savanna https://review.openstack.org/57317 | 18:52 |
| mordred | sdague: I think that trying to do config file management is way out of scope | 18:52 |
| mordred | actually, I want debs and rpms that stop trying to manage those things | 18:52 |
| lifeless | sdague: hi | 18:53 |
| openstackgerrit | Sean Dague proposed a change to openstack-infra/elastic-recheck: make ResultSet actually inherit from list https://review.openstack.org/59805 | 18:53 |
| mordred | because you're really going ot have to have config management to deal with a real deployment anyway | 18:53 |
| clarkb | SergeyLukjanov: np, you may notice some weirdness until the d-g change gets in (I didn't realize that wasn't already approved) but the jobs are non voting so it shouldn't hurt anything you are doing | 18:53 |
| *** dkranz has quit IRC | 18:53 | |
| sdague | mordred: that's fine, but providing stubs for installation that you can edit are actually nice | 18:53 |
| lifeless | sdague: config drive is generated once from the hypervisor; metadata is generated on demand from the API server | 18:53 |
| mordred | sdague: so - pip can't deal with upgrades in a sensible way anyway | 18:54 |
| sdague | lifeless: ok, one sec, let me argue with mordred first :) | 18:54 |
| SergeyLukjanov | clarkb, that's why I'd like to see them working as non-voting first ;) | 18:54 |
| mordred | sdague: but - I could be argued down potentially | 18:54 |
| sdague | mordred: ok, so here's the thing. Take something like elastic-recheck | 18:54 |
| *** rongze has quit IRC | 18:54 | |
| SergeyLukjanov | clarkb, but it's not clear for me - what kind of weirdness could happen? | 18:54 |
| sdague | it actually requires a reasonable amount of stuff in /etc to work | 18:54 |
| *** mrodden has quit IRC | 18:55 | |
| lifeless | sdague: I would never get in the way of that | 18:55 |
| sdague | and it would be really nice if in a blank environment it generated a full set of stubs | 18:55 |
| openstackgerrit | yolanda.robla proposed a change to openstack-infra/jenkins-job-builder: Enable ignore_cache flag on jenkins_jobs.ini https://review.openstack.org/59515 | 18:55 |
| clarkb | SergeyLukjanov: the savanna stuff won't actually run until the devstack-gate change goes in | 18:55 |
| lifeless | sdague: BTW my 4 hours of hell starts in 5m, so you'll have some bw issues from me later :) | 18:55 |
| clarkb | SergeyLukjanov: but the job will still run | 18:55 |
| sdague | because otherwise you are hunting a lot to get the 4 config files in the right place | 18:55 |
| sdague | (note: er has other issues about silent failures.... but one thing at a time) | 18:56 |
| sdague | lifeless: I can hit you up in 4 hrs | 18:56 |
| SergeyLukjanov | clarkb, oh, that's ok, I think that you mean some really bad behavior ;) | 18:56 |
| sdague | or even 5 | 18:56 |
| lifeless | sdague: hit me up whenever just explaining why I'll be high latency | 18:56 |
| pleia2 | woo #MeetingTuesday | 18:57 |
| zaro | fungi: ok. figured out how to create jenkins credentials. | 18:57 |
| jeblair | sdague, mordred: i would be really upset if 'python setup.py install' started putting files in system locations. i am much happier just 'cp samples/etc/* /etc/foo'. | 18:57 |
| fungi | zaro: awesome--thanks. i'll try to get that knocked out after the meeting | 18:57 |
| mordred | jeblair: ++ | 18:58 |
| sdague | lifeless: ok, the root of it is the variable that dprince flipped to force config drive on always, the chunk of nova code that that enables is really narrow | 18:58 |
| *** pblaho has joined #openstack-infra | 18:58 | |
| sdague | and mostly just turns off file injection | 18:58 |
| lifeless | sdague: if you ignore the API server entirely, sure. | 18:59 |
| *** alcabrera is now known as alcabrera|afk | 18:59 | |
| *** pblaho has left #openstack-infra | 18:59 | |
| sdague | lifeless: as far as I can tell that variable only effects a really narrow amount of the virt driver | 19:00 |
| notmyname | I've not seen a jenkins gate fail with "unstable" before now. what does that mean? | 19:00 |
| sdague | so I'm second guessing whether reverting that change, or making it flip in d-g is actually going to address your issue | 19:00 |
| lifeless | sdague: right, its nto a virt driver problem. | 19:00 |
| jeblair | notmyname: if it's a docs build, it was due to an error earlier since corrected | 19:00 |
| notmyname | jeblair: ah, ok. it was. and what should I reference to reverify? | 19:01 |
| lifeless | sdague: examining as a virt driver problem misses the bug, which wasn't a virt driver bug. | 19:01 |
| pleia2 | meeting time | 19:01 |
| sdague | lifeless: right, but that means that my d-g patch is probably not the right one | 19:01 |
| jeblair | notmyname: not sure someone filed a bug; 'no bug' should be fine; the filesystem filled up. | 19:01 |
| notmyname | jeblair: I thought reverify no bug was turned off | 19:01 |
| jeblair | notmyname: almost, not yet. | 19:02 |
| lifeless | sdague: why not? | 19:02 |
| fungi | jeblair: the related change for that lgtm though | 19:02 |
| notmyname | jeblair: FWIW, I'm reiterating that I'd prefer not turning it off, for times like this :-) | 19:02 |
| jeblair | notmyname: we'll file a bug in the future. :) | 19:02 |
| clarkb | notmyname: can you reapprove? | 19:02 |
| clarkb | I don't see a change so don't know what we are talking about | 19:03 |
| notmyname | clarkb: https://review.openstack.org/#/c/59481/ | 19:03 |
| clarkb | notmyname: right so you can just reapprove that change | 19:03 |
| clarkb | which is why I don't think we need reverify no bug | 19:03 |
| notmyname | clarkb: does it require me reapproving? just a reverify would work | 19:03 |
| *** ryanpetrello has joined #openstack-infra | 19:04 | |
| fungi | basically taking out "reverify no bug" simply increases the burden for documentation on non-core reviewers | 19:04 |
| fungi | core reviewers can just use the approve button | 19:04 |
| clarkb | today reverify works fine, but the way to deal with reverify no bug going away in my opinion is to have core reviewers reapprove | 19:04 |
| sdague | lifeless: ok, so is the part that I'm missing that cloud-init looks for config-drive, if it doesn't find it, it fails back to md server? | 19:04 |
| lifeless | sdague: yes | 19:05 |
| sdague | ok, fair | 19:05 |
| *** che-arne has quit IRC | 19:05 | |
| sdague | so... in the nova drivers, we do ugly file injection if we are not down the config-drive path | 19:05 |
| sdague | anyway, I actually think the right patch is probably just to revert dprince's blanket devstack patch | 19:06 |
| SpamapS | soren: could you have uvirtbot join #openstack-meeting-alt ? kthxbai | 19:07 |
| dprince | sdague: Ideally we would continue to test config drive somewhere though... | 19:07 |
| *** jamesmcarthur_ has joined #openstack-infra | 19:08 | |
| mtreinish | dprince: there are config drive tests in tempest | 19:09 |
| dprince | mtreinish: Those didn't stop the regression a month or so back... | 19:10 |
| *** jamesmcarthur has quit IRC | 19:10 | |
| *** jamesmcarthur_ is now known as jamesmcarthur | 19:10 | |
| mtreinish | dprince: then we need more tests :) | 19:10 |
| sdague | dprince: so the problem is now, we're not testing metadata server | 19:10 |
| *** dkranz has joined #openstack-infra | 19:11 | |
| dprince | mtreinish: I also like config drive by default because it avoids using injection which in my experience is quite slow. With config drive my experience was it was much faster. | 19:11 |
| mtreinish | I think the config drive tests there now are pretty limited (I'm not sure though I haven't really looked in detail) | 19:11 |
| sdague | which let it get broken in neutron, and totally busted tripleo, as they can't use config drive for the undercloud | 19:11 |
| mtreinish | dprince: I use config drive by default too, but sdague brings up the good point that we don't test the metadata server that way | 19:11 |
| ttx | reed: indeed | 19:12 |
| dprince | sdague: My only concern is that we have config_drive=True for at least one test configuration. At the time everyone seemed to think using it as the default was the way to go. | 19:12 |
| sdague | dprince: yeh, well I didn't think about the undercloud problem | 19:12 |
| dprince | sdague: Disabling it for a single run works too... | 19:12 |
| ttx | can't wait until we get those pages in sync with governance repo | 19:12 |
| reed | ttx, thanks, fixing it now | 19:12 |
| sdague | I +2ed your patch, so I did agree previously | 19:12 |
| sdague | but lifeless brought up this very specific break we caused in tripleo | 19:13 |
| ttx | reed: I fixed it | 19:13 |
| reed | ttx, you were faster than me :) | 19:13 |
| lifeless | dprince: I want config drive in one config and metadata + neutron in another. | 19:13 |
| lifeless | we need to test both, since both will be used | 19:14 |
| lifeless | as baremetal really can't do config drive, not for the indefinite future | 19:14 |
| dprince | lifeless: exactly, me too (we agree for once!) | 19:14 |
| jog0 | notmyname: misspellings doesn't do grammer, and no I don't plan on gating | 19:14 |
| sdague | dprince: no fair, I agreed with lifeless today as well | 19:14 |
| notmyname | jog0: good to know, but why me? | 19:15 |
| sdague | ok, so I guess my d-g change is probably the right one | 19:15 |
| notmyname | jog0: I was staying out of that thread. maybe you meant jgriffith? | 19:15 |
| *** chandankumar has quit IRC | 19:15 | |
| * jgriffith runs for the door | 19:16 | |
| sdague | jej | 19:16 |
| jog0 | notmyname: woops wrong john | 19:16 |
| sdague | heh | 19:16 |
| clarkb | jog0: the way I feel about a lot of these things is we should just shame people into using a proper editor. but then we would never get any work done... | 19:16 |
| jog0 | jgriffith: ^ | 19:17 |
| jgriffith | jog0: :) | 19:17 |
| sdague | jgriffith: if it's any consolation, I tried to talk him out of it in #openstack-qa this morning :) | 19:17 |
| jgriffith | sdague: well thanks for that at any rate | 19:17 |
| jog0 | clarkb: shaming is always fun, I use spell check in mine | 19:17 |
| *** fbo_away is now known as fbo | 19:17 | |
| jgriffith | afraid that posting just came up on a bad day | 19:17 |
| jgriffith | normally I just ignore | 19:17 |
| sdague | :) | 19:18 |
| jgriffith | jog0: clarkb isn't that supposed to be "shameing" | 19:18 |
| jog0 | jgriffith: someone -1 you for grammar? | 19:18 |
| jgriffith | :) | 19:18 |
| jgriffith | jog0: I've seen it happen to other people multiple times and it's happened to me in the past | 19:18 |
| jgriffith | but I'm just a grumpy old man today | 19:18 |
| jgriffith | it's cold and gray and I'm a grouch | 19:18 |
| sdague | jeblair: https://review.openstack.org/#/c/59303/1 - so we actually agree on that change now | 19:19 |
| jog0 | jgriffith: yeah the grammer thing is a whole nother can of worms | 19:19 |
| jog0 | that you just opened, so have fun! | 19:19 |
| sdague | -1 nother not a word | 19:19 |
| jgriffith | darn me!! | 19:19 |
| sdague | dprince: ok, so that review of mine will give us the infrastructure to test both, then we can just pick which jobs are on which | 19:19 |
| *** shardy is now known as shardy_afk | 19:20 | |
| jeblair | sdague: so you plan on dividing it up? say, running postgres with metadata or something? | 19:20 |
| lifeless | jeblair: isn't postgresql flakey already? | 19:20 |
| lifeless | jeblair: If I have any influence here I'd want to see mysql + neutron + metadata. | 19:20 |
| sdague | yeh, I'd say we'd put the neutron job on it | 19:21 |
| jeblair | lifeless: my example was arbitrary | 19:21 |
| lifeless | We have to pair neutron + metadata. | 19:21 |
| sdague | the big concern is neutron is doing extra cleverness here | 19:21 |
| lifeless | thats the combo that we need coverage on, which we won't get anywhere else. | 19:21 |
| lifeless | the namespace escaping code, the proxying of requests to the api server on behalf of the user. | 19:22 |
| jeblair | is non-neutron metadata interesting? since neutron runs so few tests | 19:22 |
| sdague | jeblair: and, yes, I'll propose it | 19:22 |
| sdague | jeblair: it does boot some guests | 19:22 |
| sdague | honestly, I'd actually probably turn this on for some of the neutron jobs | 19:22 |
| sdague | and we can pick which nova job as well | 19:22 |
| *** sarob has joined #openstack-infra | 19:23 | |
| *** sdake has quit IRC | 19:24 | |
| *** sdake has joined #openstack-infra | 19:24 | |
| *** reed has quit IRC | 19:25 | |
| *** reed has joined #openstack-infra | 19:28 | |
| *** dkranz has quit IRC | 19:28 | |
| *** jerryz has quit IRC | 19:28 | |
| openstackgerrit | A change was merged to openstack-infra/devstack-gate: Add Savanna testing support https://review.openstack.org/57325 | 19:29 |
| *** melwitt has joined #openstack-infra | 19:29 | |
| *** melwitt has quit IRC | 19:29 | |
| *** melwitt has joined #openstack-infra | 19:29 | |
| *** ewindisch has joined #openstack-infra | 19:31 | |
| *** vipul is now known as vipul-away | 19:33 | |
| *** vipul-away is now known as vipul | 19:33 | |
| *** gyee has quit IRC | 19:35 | |
| *** jcooley_ has quit IRC | 19:37 | |
| *** jcooley_ has joined #openstack-infra | 19:37 | |
| *** denis_makogon has joined #openstack-infra | 19:39 | |
| *** dkranz has joined #openstack-infra | 19:41 | |
| *** jcooley_ has quit IRC | 19:41 | |
| *** Hunner_irssi has quit IRC | 19:42 | |
| *** jcooley_ has joined #openstack-infra | 19:42 | |
| *** cederb02 has joined #openstack-infra | 19:44 | |
| *** ewindisch has quit IRC | 19:46 | |
| *** cederb02 has quit IRC | 19:46 | |
| *** jcooley_ has quit IRC | 19:47 | |
| *** jcooley_ has joined #openstack-infra | 19:47 | |
| jog0 | jeblair: https://review.openstack.org/#/c/59289/ | 19:50 |
| jog0 | I am confused at your +2 | 19:50 |
| *** rongze has joined #openstack-infra | 19:51 | |
| jaypipes | jeblair, mordred: are we allowed to git tag inside a script block in a jenkins/zuul job? | 19:51 |
| *** jcooley_ has quit IRC | 19:51 | |
| jeblair | jog0: it means it looks good to me. what's confusing about it? | 19:51 |
| jaypipes | also, IIRC, only zuul/jenkins and gerrit admins are allowed to push tags to projects... is that correct? we will likely need help from your team to do initial git tags on the cookbooks. | 19:52 |
| *** jcooley_ has joined #openstack-infra | 19:52 | |
| jeblair | jaypipes: no, small groups of core folks can push tags. jenkins is not allowed to do that. | 19:52 |
| openstackgerrit | sebastian marcet proposed a change to openstack-infra/config: Review and create Puppet configuration for dev server https://review.openstack.org/53644 | 19:53 |
| *** vipul is now known as vipul-away | 19:53 | |
| jog0 | jeblair: the name stuff in that repo | 19:53 |
| jaypipes | jeblair: ok. so by default, can the openstack-chef-core team (or whatever it's called, can't remember) push tags to the cookbook repos? | 19:53 |
| jog0 | I guess we will jsut remember to change it | 19:53 |
| mordred | jaypipes: yeah | 19:54 |
| jaypipes | mordred: cheers. | 19:54 |
| jeblair | jog0: oh, you mean the comment before my +2. i'm saying that job names should not be in e-r, they should be in config. | 19:54 |
| mordred | jaypipes: well, however you have set up in your acls files | 19:54 |
| jeblair | jog0: so in the long run, e-r should grow the ability to have config files that list things like job names | 19:54 |
| jeblair | jog0: (eg, for things like that graph) | 19:54 |
| jog0 | jeblair: ahh | 19:54 |
| jaypipes | mordred: if jenkins/zuul is not able to push tags, how exactly is a pipeline job supposed to automatically call git tag with an incremented semver? | 19:54 |
| mordred | what? | 19:55 |
| jog0 | I ask because I want to add a line to that graph | 19:55 |
| *** sparkycollier has quit IRC | 19:55 | |
| mordred | oh - I have no idea, it's one of the reasons that we have done things in the opposite direction | 19:55 |
| jog0 | will just do a dependandant change | 19:55 |
| jaypipes | mordred: is that what for me? :) | 19:55 |
| mordred | jaypipes: yeah - sorry, I'm splitting my attention in 4 different directions right now - so I might be scattered | 19:55 |
| jaypipes | no worries :) | 19:55 |
| lifeless | no, really? | 19:55 |
| jeblair | jog0: well, right now i think you'd just add it to the e-r repo | 19:56 |
| *** rongze has quit IRC | 19:56 | |
| jeblair | jog0: but that's what i'm saying won't work in the long run, because we can, and will, change job names and you'll be forever catching up if they aren't in the infra/config repo | 19:57 |
| fungi | mordred: jaypipes: the bits i gleaned suggest they want autoversioning akin to how pbr makes magic version numbers for arbitrary git commits which don't correspond to a tag | 19:57 |
| jaypipes | mordred: so we are trying to go in the *same* direction :) we are removing what are essentially build artifacts from the cookbook trees, but we still need to increment the cookbook semver on commits. we were planning to calculate the version based on git tags, but in order to have the branch tag'd with a semver, we need to be able to apply the git tag during some Zuul (post-merge?) pipeline | 19:57 |
| jaypipes | fungi: precisely. | 19:58 |
| fungi | which would, by definition, not be tags | 19:58 |
| mordred | jaypipes: ah - so, we do the git tag ourselves | 19:58 |
| jaypipes | fungi: well... | 19:58 |
| mordred | jaypipes: as in, we drive the whole process by a human doing a git tag on the repo and pushing the tag | 19:58 |
| pleia2 | clarkb: can I borrow you for a call+etherpad to chat about the process for tagging our historical publications? | 19:59 |
| fungi | which i can imagine is potentially burdensome if someone is going to need to tag every commit by hand, even when they're not release-worthy commits (minor tyop correction) | 19:59 |
| pleia2 | clarkb: doesn't need to be now, just some time this week if you have time | 19:59 |
| jog0 | jeblair: right, so long run its config. thanks for the clarification | 20:00 |
| *** mrodden has joined #openstack-infra | 20:00 | |
| mordred | jaypipes: sorry - this is too complex for split attention - let me find you when I can give you good attention | 20:00 |
| clarkb | pleia2: sure I can do that after lunch | 20:00 |
| jaypipes | mordred: we are currently doign this versioning crap in reviews for the cookbooks, and it is a giant pain in the butt (having to constantly rebase just to change the semver increment). We're trying to automate the process of tagging each commit with an appropriately-incremented semantic version that we can then use to drive further Chef-specific automation | 20:00 |
| jaypipes | mordred: I understand you're busy, no probs. | 20:01 |
| clarkb | pleia2: ~1:30 pm PST work? if not I can do times later this week | 20:01 |
| pleia2 | clarkb: perfect | 20:01 |
| SergeyLukjanov | are there any docs about nodepool project? | 20:02 |
| fungi | jaypipes: the short answer is that tag-based versioning alone may not be suitable for that specific use case. we use tags for human-determined release and prerelease/milestone points in the code, but commits in between those get autogenerated dev versions which do not have corresponding tags | 20:03 |
| jeblair | fungi, clarkb, mordred: the devstack gate shared queue looks more normal now. | 20:03 |
| jaypipes | fungi: I understand. | 20:03 |
| clarkb | jeblair: woot | 20:03 |
| clarkb | I am going to grab lunch now | 20:03 |
| *** dhellmann has quit IRC | 20:04 | |
| *** vipul-away is now known as vipul | 20:04 | |
| fungi | jaypipes: if chef itself expects every single commit in a git repo to have non-repeated/ever-increasing version numbers while at the same time providing no mechanism to generate those (for example looking at git describe), that sounds mildly broken | 20:04 |
| jaypipes | fungi: preaching to the choir :) | 20:05 |
| fungi | jaypipes: it just sounds like they haven't thought through their consume-directly-from-git tooling | 20:05 |
| *** thingee has quit IRC | 20:06 | |
| jaypipes | fungi: no, it's because of a whole bunch of dependent tooling in Chef land... | 20:06 |
| *** jcooley_ has quit IRC | 20:06 | |
| fungi | because people don't usually increase the version number of a project for every single commit | 20:06 |
| jaypipes | fungi: well, semver actually does do that... | 20:06 |
| *** rockyg has joined #openstack-infra | 20:06 | |
| fungi | semver a thing (not just the spec?) | 20:06 |
| openstackgerrit | Khai Do proposed a change to openstack-dev/pbr: show diff between pip installed packages and requirements https://review.openstack.org/54639 | 20:07 |
| openstackgerrit | Joe Gordon proposed a change to openstack-infra/elastic-recheck: Remove queries for bugs 1225024 and 1225664 https://review.openstack.org/59816 | 20:07 |
| uvirtbot | Launchpad bug 1225024 in tempest "tempest.api.compute.admin.test_hosts.HostsAdminTestXML.test_list_hosts_with_zone[gate] unexpected conductor service" [Undecided,Incomplete] https://launchpad.net/bugs/1225024 | 20:07 |
| uvirtbot | Launchpad bug 1225664 in devstack "Swift: No space left on device" [Undecided,Fix committed] https://launchpad.net/bugs/1225664 | 20:07 |
| fungi | jaypipes: you mean there's some tool named semver which chef uses to generate intermediate version numbers for non-tagged commits in a git repository? | 20:08 |
| jaypipes | fungi: berkshelf and other Chef tooling uses semver to differentiate between cookbook versions, so if we change pretty much *anything* in the cookbook, we need to increment the semver. | 20:08 |
| fungi | or you just mean that the idea of semver suggests that there should be a way to make up a version number for situations like that | 20:08 |
| jaypipes | fungi: no, I mean the spec... | 20:08 |
| *** elasticio has joined #openstack-infra | 20:08 | |
| *** elasticio is now known as Guest50750 | 20:09 | |
| *** dhellmann has joined #openstack-infra | 20:09 | |
| fungi | it really sounds like they don't want people consuming non-release versions of cookbooks then | 20:09 |
| *** Guest50750 has quit IRC | 20:09 | |
| jaypipes | fungi: which is kinda antithetical to continuously-deployed systems, don't ya think? :) | 20:09 |
| fungi | i guess their assumption is that you queue up your edits, and then squash them all together and commit that as your "next version" | 20:10 |
| jaypipes | fungi: a lot of this is because the way that Chef and Berkshelf determine conflicting dependencies between cookbooks is solely by the use of semver. | 20:10 |
| *** datsun180b has joined #openstack-infra | 20:11 | |
| *** dolphm has joined #openstack-infra | 20:11 | |
| *** elasticio__ has joined #openstack-infra | 20:11 | |
| *** vipul is now known as vipul-away | 20:11 | |
| *** vipul-away is now known as vipul | 20:12 | |
| fungi | right, so presumably they expect very-slow-moving/glacial cookbook development where commits fall far enough apart so as to correspond to actual releases, or the idea of consuming these from a version control system directly suggests the need for some artificial filter to batch up those changes into versions | 20:12 |
| *** thingee has joined #openstack-infra | 20:12 | |
| openstackgerrit | Ryan Petrello proposed a change to openstack-infra/config: Run pecan-ceilometer gate tests on centos6 https://review.openstack.org/59818 | 20:12 |
| *** datsun_F40PH has quit IRC | 20:13 | |
| mattray | fungi: sounds right. Unfortunately we're trying to have it both ways | 20:13 |
| *** ^d is now known as ^demon|lunch | 20:13 | |
| fungi | which takes me back to the (not entirely serious) suggestion that one of the release artifacts could be an artificially-constructed separate git repository conforming to those questionable assumptions | 20:14 |
| *** datsun180b has quit IRC | 20:14 | |
| *** datsun180b has joined #openstack-infra | 20:14 | |
| fungi | "git the way chef expects it, not the way devs are using it) | 20:14 |
| fungi | or perhaps a branch which gets merge commits from git tags in development | 20:15 |
| *** DinaBelova has quit IRC | 20:15 | |
| fungi | so that at any point in time the latest commit on that branch corresponds to an actual release tag | 20:15 |
| mgagne | jaypipes: if zuul (or whatever) autotags every commits with a version based on symver spec. How are major version cut? | 20:16 |
| jaypipes | fungi: right... we're not saying it's pretty. :) | 20:16 |
| fungi | and downstream consumers simply don't ever see a state where what they're using has intermediate between-tag commits | 20:16 |
| jaypipes | mgagne: zuul does not autotag anything. | 20:16 |
| mgagne | jaypipes: I was referring to an hypothetical situation mentioned above. | 20:16 |
| *** elasticio__ has left #openstack-infra | 20:16 | |
| fungi | right, in our workflow tags are something humans deliberate, cryptographically sign to assert a particular set of conditions, push to the repository and then automation triggers from those | 20:17 |
| jaypipes | mgagne: oh... looking for #major in the commit message was a proposed solution. | 20:17 |
| *** salv-orlando has joined #openstack-infra | 20:17 | |
| jaypipes | fungi: we're not arguing with you about the merit of that approach :) | 20:17 |
| mgagne | jaypipes: I'm interested in what you guys have to set. just asking questions for clarification =) | 20:17 |
| fungi | yeah, got it | 20:17 |
| mgagne | say* | 20:17 |
| jaypipes | mgagne: oh, no worries... mainly just brainstorming here... | 20:18 |
| mordred | this hurts my head | 20:18 |
| mordred | how does this work in a non-gerrit workflow? | 20:18 |
| mgagne | jaypipes: we have a similar internal system here. | 20:18 |
| mordred | does a person edit metadata.rb on every commit? | 20:18 |
| mgagne | jaypipes: but it's kind of cumbersome | 20:18 |
| * mordred feels there is something he's missing on how the system is intended to work | 20:19 | |
| mgagne | jaypipes: version is stored in pom.xml and incremented accordingly which kind of results in 2n commits | 20:19 |
| *** vkozhukalov has quit IRC | 20:19 | |
| jaypipes | mgagne: yes, it's very cumbersome for us right now... manually having to keep in our mind the next patch number for semver.. sounds like we're going to give up the ghost on a patch number increment and just focus on minor increments and tag manually. | 20:19 |
| fungi | so i think having the automation add tags to every commit is missing the point... tail wagging the dog as it were. instead tooling could just trigger off the commits themselves, regardless of the presence of tags (that is, tags lose their meaning if they're autogenerated and apply to every single commit) | 20:19 |
| jaypipes | fungi: agreed | 20:20 |
| *** _david_ has joined #openstack-infra | 20:20 | |
| *** dstanek has joined #openstack-infra | 20:20 | |
| mattray | fungi: yeah, that tooling exists. Again the problem with trying to solve everyone's use cases simultaneously | 20:22 |
| *** alcabrera|afk is now known as alcabrera | 20:22 | |
| *** elasticio__ has joined #openstack-infra | 20:22 | |
| openstackgerrit | Sean Dague proposed a change to openstack-infra/elastic-recheck: implementation of FacetSet for client side nested facets https://review.openstack.org/59822 | 20:24 |
| mgagne | jaypipes: on a side note, it still looks to be the perfect solution (tag everything) for people pedantic (or very concerned) about versioning and control. I still haven't found an alternate solution for people with this kind of needs. People aiming for continuous delivery have this issue. (IMO/AFAIK) | 20:24 |
| jaypipes | mgagne: yes, but fungi's point about if you tag every commit, you lose the meaning of the tag -- why not just use the SHA1 at that point? :) | 20:25 |
| mordred | well, sha or even commit count since last tag | 20:26 |
| jaypipes | indeed. | 20:26 |
| mordred | that's what we do on the things between tags | 20:26 |
| mgagne | jaypipes: is it correct to say chef/cookbook only understands symver ? | 20:26 |
| mikal | So... In zuul, if I have a success layout entry which says gerrit verified 0, will that just put a comment on the review with no review score? | 20:26 |
| mordred | $tag.$revcount (althoug we do pep 440 - so it's $tag.a$revcount) | 20:26 |
| jaypipes | mgagne: pretty much, yes... at least, the dependency and cookbook conflict resolution only works on semver. | 20:26 |
| *** vipul is now known as vipul-away | 20:27 | |
| mgagne | jaypipes: My conclusion is that we have common needs (although for different systems/softwares). | 20:27 |
| * mgagne listens | 20:27 | |
| jaypipes | :) | 20:27 |
| *** jasond has joined #openstack-infra | 20:27 | |
| jasond | how can i add a new group in gerrit? | 20:28 |
| mgagne | jaypipes: Does it have to be from git? Or can artifacts (for each commit) be published somewhere else? Is the version stored somewhere in the artifact itself? | 20:28 |
| *** salv-orlando_ has joined #openstack-infra | 20:28 | |
| mgagne | damn, it's semver btw, right? =) | 20:29 |
| mgagne | http://semver.org/ | 20:29 |
| mgagne | basically the same versioning system as every respectful libraries | 20:30 |
| *** salv-orlando has quit IRC | 20:30 | |
| *** salv-orlando has joined #openstack-infra | 20:31 | |
| jaypipes | mgagne: right. | 20:31 |
| fungi | mikal: i believe so | 20:32 |
| fungi | jasond: which gerrit. the review.openstack.org gerrit or any random gerrit? | 20:32 |
| mikal | fungi: thanks | 20:32 |
| jasond | fungi: in review.openstack.org | 20:33 |
| fungi | jasond: we autogenerate groups from project acls | 20:33 |
| jaypipes | mgagne: our (the OpenStack+Chef) community seems to be moving towards using git tags for major and minor semver increments, with core reviewers manually pushing those tags themselves, removing the manual Changelog entries in the cookbooks (and generating that information from git logs), and essentially stopping the enforcement of the patch semver increment entirely. | 20:33 |
| *** salv-orlando_ has quit IRC | 20:33 | |
| jasond | fungi: oh ok, thanks | 20:34 |
| fungi | jasond: so any group mentioned in an acl will be created automatically. we simply add accounts to them (for most groups, an initial member only, and let that member add more users to a self-managed group) | 20:34 |
| *** hartsocks has joined #openstack-infra | 20:34 | |
| openstackgerrit | sebastian marcet proposed a change to openstack-infra/config: Review and create Puppet configuration for dev server https://review.openstack.org/53644 | 20:34 |
| mikal | fungi: where does the zuul merger get its git repo from? I'm getting ssh permission denieds not that I've moved out zuul to using that new service account you setup yesterday. | 20:34 |
| mikal | fungi: I'm not sure if that's a problem on my side or yours | 20:35 |
| fungi | mikal: zuul constructs a git reference in a checkout of the project or projects in question and serves that to be accessed by the jenkins slaves (acting in the role of git clients to it) | 20:35 |
| mikal | fungi: ugh, found it. The origin in the git repo has jhesketh's account name hard coded. | 20:36 |
| fungi | or rather, zuul doesn't serve it, so much as it passes along the reference name in a parameter, and an apache vhost on the zuul server actually serves the repo | 20:36 |
| fungi | mikal: that'd do it. is flogging mandated in that situation? | 20:37 |
| mikal | Heh | 20:37 |
| *** yolanda has quit IRC | 20:37 | |
| mikal | It doesn't help that jhesketh is in a training course | 20:37 |
| mikal | I'm jsut trying to cover for him (badly) | 20:37 |
| fungi | oh, then he's already being punished enough from the sound of it | 20:37 |
| *** elasticio__ has quit IRC | 20:38 | |
| *** elasticio__ has joined #openstack-infra | 20:38 | |
| *** kui has joined #openstack-infra | 20:39 | |
| mikal | fungi: sorry to be a pain. What does "no connected gearman servers" from zuul mean? That all my workers are down? Or is there a central gearman process? | 20:40 |
| *** dstanek has quit IRC | 20:41 | |
| *** rockyg has quit IRC | 20:42 | |
| fungi | mikal: zuul and the jenkins master(s) connect to a gearman server, which could be (often is) the built-in lightweight gearman server provided in zuul | 20:43 |
| mikal | fungi: ok, I will try restarting zuul then to see if it fixes the server | 20:44 |
| mikal | fungi: noting that will dump the queue, but that I don't care at the moment | 20:44 |
| fungi | mikal: i'd have to look through the source to confirm the circumstances under which it will log that error, and whether it is emitted by the zuul server thread or zuul's gearman server | 20:44 |
| notmyname | what is "dvsm"? | 20:45 |
| fungi | mikal: also, graceful restart of zuul should resume the previous queue contents, though you have to wait for running jobs to report first | 20:45 |
| fungi | notmyname: dsvm? DevStack Virtual Machine | 20:45 |
| *** vipul-away is now known as vipul | 20:45 | |
| notmyname | fungi: thanks. I just started seeing it today in the job names | 20:45 |
| fungi | notmyname: job names got way too long as our combinatorials exploded | 20:46 |
| fungi | notmyname: so they got renamed/abbrev'd yesterday | 20:46 |
| mordred | fungi: we sohuld maybe send out a dev list email about that | 20:46 |
| mikal | fungi: is there a way to ask zuul how many workers it thinks are attached? | 20:47 |
| openstackgerrit | Joe Gordon proposed a change to openstack-infra/elastic-recheck: Cleanup graphite graph https://review.openstack.org/59829 | 20:47 |
| openstackgerrit | Joe Gordon proposed a change to openstack-infra/elastic-recheck: Remove check jobs from graphite graph https://review.openstack.org/59830 | 20:47 |
| openstackgerrit | Joe Gordon proposed a change to openstack-infra/elastic-recheck: Add more gate jobs to graphite graph https://review.openstack.org/59831 | 20:47 |
| openstackgerrit | Sean Dague proposed a change to openstack-infra/elastic-recheck: port check_success to facets https://review.openstack.org/59832 | 20:48 |
| fungi | mikal: mmm, not sure. if it's not in the status.json (which i think it's probably not) then you'd have to figure it out from the log file or trigger a thread dump and count the worker threads maybe | 20:48 |
| mordred | mikal: buy it a drink | 20:48 |
| mikal | Heh | 20:49 |
| fungi | though would make a great wishlist bug for new feature in the zuul client tool up for review currently | 20:49 |
| fungi | either that or maybe a way to expose it through the server status details | 20:49 |
| *** johnthetubaguy has quit IRC | 20:50 | |
| *** salv-orlando has quit IRC | 20:51 | |
| fungi | mikal: you might be able to determine it by looking at netstat -nt (is gearman tcp?) and doing a grep -c for the gearman port number | 20:51 |
| * fungi tries that | 20:51 | |
| mikal | fungi: looking at a worker, I think they were disconnected by the grceless zuul restart | 20:51 |
| *** jcooley_ has joined #openstack-infra | 20:51 | |
| *** jasond has left #openstack-infra | 20:52 | |
| *** jcooley_ has quit IRC | 20:53 | |
| *** alcabrera has quit IRC | 20:53 | |
| *** rongze has joined #openstack-infra | 20:53 | |
| mikal | fungi: so to provide some context, because we're not ready to expose the test run logs we want to run tests but not provide scores (just a comment). We'll then work on fixing the log issue. We're also only going to run tests on changesets which have already passed jenkins to reduce our false positive chatter. | 20:53 |
| fungi | mmm, when i 'netstat -nt|grep -c :4730' on zuul.o.o i get 288 | 20:53 |
| fungi | which looks like it might roughly correspond to the number of active jenkins slaves we have at the moment | 20:54 |
| mikal | fungi: wow, I do that on my zuul and get a number roughly double the number of workers I expect | 20:55 |
| fungi | mikal: maybe some of those are in timewait or something other than established | 20:56 |
| *** CaptTofu has quit IRC | 20:56 | |
| fungi | netstat -nt|grep -c ':4730.*ESTABLISHED' | 20:57 |
| mikal | fungi: oh, possible | 20:57 |
| *** CaptTofu has joined #openstack-infra | 20:57 | |
| mikal | fungi: except they're all established | 20:57 |
| fungi | huh. neat | 20:57 |
| *** hashar has joined #openstack-infra | 20:58 | |
| *** dhellmann has quit IRC | 20:58 | |
| *** rongze has quit IRC | 20:58 | |
| mikal | fungi: yeah, I get two entries per connecting worker | 20:58 |
| mikal | fungi: but there is only one worker process on this here random worker I am looking at | 20:59 |
| *** dhellmann has joined #openstack-infra | 20:59 | |
| *** krotscheck has quit IRC | 21:00 | |
| *** datsun_F40PH has joined #openstack-infra | 21:00 | |
| *** vipul is now known as vipul-away | 21:00 | |
| *** krotscheck has joined #openstack-infra | 21:01 | |
| fungi | mikal: perhaps half of those are stale connections? tcpdump them and see if some aren't chatty at all? | 21:02 |
| *** rockyg has joined #openstack-infra | 21:03 | |
| *** datsun180b has quit IRC | 21:03 | |
| *** datsun_F40PH is now known as datsun180b | 21:03 | |
| *** hartsocks has left #openstack-infra | 21:04 | |
| *** che-arne has joined #openstack-infra | 21:04 | |
| *** hogepodge has joined #openstack-infra | 21:04 | |
| openstackgerrit | Yathiraj Udupi proposed a change to openstack/requirements: Added a requirement to COIN PULP LP Modeler module https://review.openstack.org/53446 | 21:05 |
| *** boris-42 has joined #openstack-infra | 21:07 | |
| *** shardy_afk is now known as shardy | 21:07 | |
| jog0 | we are seeing nova unit tests fail and its a timeout, was hoping a few infra can look at it | 21:08 |
| jog0 | we are talking about it in infra | 21:08 |
| jeblair | mikal: yes, 1 sec | 21:08 |
| jeblair | mikal: gearman has a text-based admin protocol on the same port | 21:08 |
| jeblair | mikal: telnet HOST 4730 | 21:09 |
| jeblair | mikal: and issue commands like "status" and "workers" | 21:09 |
| jeblair | mikal: http://gearman.org/protocol/ | 21:10 |
| jeblair | mikal: (search for "Administrative Protocol" on that page, i don't think it has an anchor) | 21:10 |
| *** sandywalsh_ has quit IRC | 21:10 | |
| fungi | oh, right--i totally forgot about the admin proto | 21:11 |
| fungi | jog0: have a link (review, log, something)? | 21:11 |
| jog0 | fungi: in -qa yeah | 21:12 |
| fungi | jog0: oh, you said in -infra ;) | 21:12 |
| jog0 | doh, FWIW its 11PM here | 21:13 |
| *** dhellmann has quit IRC | 21:13 | |
| *** weshay has quit IRC | 21:13 | |
| *** dhellmann has joined #openstack-infra | 21:14 | |
| mikal | jeblair: thanks | 21:15 |
| jeblair | mikal, fungi: and totally not opposed to including some of that it status.json | 21:15 |
| jeblair | _in_ status.json | 21:15 |
| jeblair | i'm going to crawl into a hole and try very hard to reproduce this zuul bug | 21:16 |
| mikal | jeblair: good luck! | 21:17 |
| clarkb | pleia2: I am back from lunch, ping me when you are ready | 21:19 |
| *** vipul-away is now known as vipul | 21:19 | |
| *** mfer has quit IRC | 21:19 | |
| *** dhellmann has quit IRC | 21:19 | |
| openstackgerrit | A change was merged to openstack-infra/jenkins-job-builder: Corrected typo in example for pipeline plugin https://review.openstack.org/59709 | 21:19 |
| pleia2 | clarkb: hey, now is fine, want to give our asterisk server a spin? | 21:20 |
| *** hogepodge has quit IRC | 21:20 | |
| clarkb | pleia2: sure (I will be using a normal phone though so anything works) | 21:20 |
| clarkb | pleia2: let me hide in a quiet room | 21:20 |
| pleia2 | clarkb: ok :) | 21:20 |
| *** mfer has joined #openstack-infra | 21:21 | |
| *** dhellmann has joined #openstack-infra | 21:21 | |
| clarkb | of course xrandr hates me and won't let me undock | 21:22 |
| clarkb | pleia2: I may be a few minutes | 21:22 |
| *** CaptTofu has quit IRC | 21:22 | |
| pleia2 | clarkb: no worries | 21:22 |
| *** jcooley_ has joined #openstack-infra | 21:24 | |
| *** sandywalsh_ has joined #openstack-infra | 21:27 | |
| *** dhellmann has quit IRC | 21:27 | |
| *** elasticio__ has quit IRC | 21:27 | |
| *** gyee has joined #openstack-infra | 21:28 | |
| zaro | pleia2: i created a patch for python-jenkins project in LP. i can't tell whether it's active. do you know? | 21:28 |
| pleia2 | zaro: I'll have a look | 21:28 |
| zaro | pleia2: https://code.launchpad.net/~zaro0508/python-jenkins/plugin-info-method | 21:29 |
| jeblair | zaro: i think i have access there, but haven't had time to look in a while | 21:29 |
| *** CaptTofu has joined #openstack-infra | 21:29 | |
| pleia2 | zaro: yeah, looks like https://code.launchpad.net/python-jenkins/+activereviews | 21:30 |
| pleia2 | zaro: technically it looks like you did everything right :) | 21:30 |
| *** ruhe has quit IRC | 21:31 | |
| pleia2 | socially I'm not sure about their nudging procedure for looking at MPs | 21:31 |
| jeblair | nudge me because i don't look at launchpad | 21:31 |
| zaro | jeblair, pleia2 : good to know. | 21:32 |
| clarkb | pleia2: ok I am going to dial into the asterisk server in just a minute | 21:33 |
| pleia2 | clarkb: cool, let's use 6001 | 21:34 |
| clarkb | ok | 21:34 |
| pleia2 | hehe, it hangs up on me | 21:34 |
| clarkb | me too | 21:35 |
| jeblair | dialing in via sip wfm | 21:37 |
| *** dstanek has joined #openstack-infra | 21:37 | |
| SergeyLukjanov | looks like https://review.openstack.org/#/c/58702/ was missed by zuul, am I right? | 21:37 |
| *** vipul is now known as vipul-away | 21:39 | |
| *** pabelanger has quit IRC | 21:39 | |
| *** vipul-away is now known as vipul | 21:39 | |
| jeblair | clarkb, pleia2: weird; after a few hangups, it worked | 21:39 |
| jeblair | (using pstn) | 21:39 |
| fungi | SergeyLukjanov: no, it's waiting for someone to approve https://review.openstack.org/57098 (and will potentially need rebasing when that happens) | 21:41 |
| jeblair | clarkb, pleia2: try again? | 21:41 |
| fungi | SergeyLukjanov: you can see that in the "dependencies" for the change you linked | 21:42 |
| SergeyLukjanov | fungi, several page refreshes helps me... thank you | 21:42 |
| *** mrodden1 has joined #openstack-infra | 21:42 | |
| *** mrodden has quit IRC | 21:43 | |
| jeblair | clarkb, pleia2: i restarted asterisk; it seems more reliable now :/ | 21:46 |
| *** fbo is now known as fbo_away | 21:47 | |
| *** Ryan_Lane has joined #openstack-infra | 21:50 | |
| *** MarkAtwood has joined #openstack-infra | 21:50 | |
| *** luisg has quit IRC | 21:51 | |
| *** rockyg has quit IRC | 21:51 | |
| *** markvan has joined #openstack-infra | 21:53 | |
| *** rongze has joined #openstack-infra | 21:55 | |
| *** Ryan_Lane has quit IRC | 21:57 | |
| ekarlso | Can this be approved ? https://review.openstack.org/#/c/56927/ | 21:57 |
| *** rongze has quit IRC | 22:00 | |
| *** jcoufal has joined #openstack-infra | 22:02 | |
| *** dkliban has quit IRC | 22:03 | |
| *** Ryan_Lane has joined #openstack-infra | 22:03 | |
| *** krtaylor has quit IRC | 22:05 | |
| *** reed has quit IRC | 22:06 | |
| *** che-arne_ has joined #openstack-infra | 22:06 | |
| *** che-arne has quit IRC | 22:07 | |
| *** reed has joined #openstack-infra | 22:07 | |
| *** Ryan_Lane has quit IRC | 22:08 | |
| bnemec | Hey, quick question. I'm looking at https://review.openstack.org/#/c/59270/2 and wondering why the requirements change didn't get synced automatically when it merged in global requirements. | 22:08 |
| clarkb | jeblair: thanks, sorry missed it, we ended up just doing a normal phone call | 22:08 |
| bnemec | Did that job get shut off because of https://bugs.launchpad.net/openstack-ci/+bug/1246249 ? | 22:08 |
| uvirtbot | Launchpad bug 1246249 in openstack-ci "Auto update requirement review proposal run on wrong branches" [Undecided,New] | 22:08 |
| clarkb | bnemec: yes | 22:08 |
| clarkb | bnemec: gerrit doesn't provide branch info after things have merged so there isn't a good easy way to make that change work | 22:09 |
| jeblair | clarkb: it doesn't? | 22:09 |
| bnemec | clarkb: Okay, thanks. It came up on the review and I wanted to find out what was going on before approving any requirements changes. | 22:09 |
| jeblair | clarkb: pretty sure ref-updated tells you what ref (eg master, stable/foo) was updated... | 22:10 |
| *** reed has quit IRC | 22:10 | |
| *** sdake_ has quit IRC | 22:10 | |
| jeblair | clarkb: and there's always the 'change-merged' event. | 22:10 |
| *** reed has joined #openstack-infra | 22:10 | |
| *** sdake_ has joined #openstack-infra | 22:10 | |
| clarkb | jeblair: hmm there is no ZUUL_BRANCH should we look at ZUUL_REF in post? | 22:11 |
| jeblair | clarkb: correct | 22:12 |
| *** mrodden1 has quit IRC | 22:12 | |
| clarkb | I will need to look at that again | 22:12 |
| *** krotscheck has quit IRC | 22:14 | |
| Alex_Gaynor | Hmm, are teh failures on https://review.openstack.org/#/c/59533/ because of the changes in CI? | 22:15 |
| *** pabelanger_ has quit IRC | 22:15 | |
| *** pabelanger has joined #openstack-infra | 22:15 | |
| clarkb | Alex_Gaynor: looks like dkranz's log filtering failed that job | 22:17 |
| *** krotscheck has joined #openstack-infra | 22:17 | |
| clarkb | jeblair: is ZUUL_REF or ZUUL_REFNAME better for getting master or stable/foo? | 22:17 |
| *** torgomatic has joined #openstack-infra | 22:18 | |
| jeblair | clarkb: equivalent; i'd go with ref | 22:18 |
| clarkb | jeblair: thanks | 22:18 |
| jeblair | clarkb: look for existing usage in slave_scripts; you may need to do some regex matching to avoid doing things on tags. | 22:19 |
| jeblair | clarkb: oh, nm, zuul handles that | 22:19 |
| Alex_Gaynor | clarkb: meaning everything is going to fail right now? | 22:19 |
| clarkb | Alex_Gaynor: meaning everything with ERRORs in the service logs is going to fail (with the exception of some whitelisted failures) | 22:19 |
| dkranz | clarkb: This seems to be a very infrequent but persistent log error http://logstash.openstack.org/#eyJzZWFyY2giOiJcIjQwNCBOb3QgRm91bmQgIFtmaXJzdCA2MCBjaGFycyBvZiByZXNwb25zZV1cIiIsImZpZWxkcyI6W10sIm9mZnNldCI6MCwidGltZWZyYW1lIjoiNjA0ODAwIiwiZ3JhcGhtb2RlIjoiY291bnQiLCJ0aW1lIjp7InVzZXJfaW50ZXJ2YWwiOjB9LCJzdGFtcCI6MTM4NjEwOTExNTkzN30= | 22:20 |
| clarkb | jeblair: where does zuul do that? | 22:20 |
| fungi | Alex_Gaynor: ERROR swiftclient [-] Object GET failed: http://127.0.0.1:8080/v1/AUTH_4d22a858761e4b90b536f489ccff34ca/glance/a6c33fc7-4871-45f7-8b3c-fd0a7452cea0 404 Not Found [first 60 chars of response] <html><h1>Not Found</h1><p>The resource could not be found.< | 22:20 |
| dkranz | Alex_Gaynor: This is a flaky issue happening 19 times in the past 7 days | 22:20 |
| jeblair | clarkb: in the layout, post jobs only run on commits | 22:21 |
| Alex_Gaynor | Ok, so not my fault, it's killing everyone :/ | 22:21 |
| clarkb | jeblair: is that what the negative lookahead regex is doing in the pipeline manager for post? | 22:21 |
| jeblair | clarkb: yeah | 22:21 |
| openstackgerrit | Tom Fifield proposed a change to openstack-infra/jeepyb: Add script to post a gerrit welcome message https://review.openstack.org/58900 | 22:21 |
| dkranz | Alex_Gaynor: I'm not sure 19 times in a week out of all runs is killing. | 22:21 |
| Alex_Gaynor | dkranz: sorry s/killing/effecting/ is all I meant | 22:21 |
| *** ryanpetrello has quit IRC | 22:22 | |
| dkranz | Alex_Gaynor: Well, other ones have been killing :) | 22:22 |
| *** dprince has quit IRC | 22:22 | |
| dkranz | I will file a bug | 22:23 |
| openstackgerrit | Clark Boylan proposed a change to openstack-infra/config: Enable requirement update proposal jobs https://review.openstack.org/59855 | 22:23 |
| clarkb | jeblair: something like that? I can't believe it is so simple (I think I convinced myself that this is hard for some reason) | 22:23 |
| dkranz | clarkb: Is there an easy way to go from some value reported in a logstash result to the patch whose build caused it? | 22:23 |
| clarkb | isn't there another set of jobs that have trouble with this? | 22:23 |
| jeblair | clarkb: can it wait for a bit? (zuul hole) | 22:23 |
| clarkb | jeblair: yup no rush | 22:24 |
| clarkb | dkranz: yes, the change and patchset numbers are recorded in the events | 22:24 |
| clarkb | dkranz: in kibana just click on one of them to expand the whole thing | 22:24 |
| dkranz | clarkb: ok, great. | 22:24 |
| *** thomasem has quit IRC | 22:27 | |
| *** smarcet has left #openstack-infra | 22:28 | |
| *** ArxCruz has quit IRC | 22:29 | |
| *** ArxCruz has joined #openstack-infra | 22:30 | |
| *** luisg has joined #openstack-infra | 22:30 | |
| *** salv-orlando has joined #openstack-infra | 22:33 | |
| *** denis_makogon has quit IRC | 22:34 | |
| dkranz | Alex_Gaynor: I filed https://bugs.launchpad.net/glance/+bug/1257509 and did a reverify on that patch | 22:34 |
| uvirtbot | Launchpad bug 1257509 in glance "swiftclient ERROR in g-api after successful tempest run" [Undecided,New] | 22:34 |
| *** fifieldt has joined #openstack-infra | 22:35 | |
| Alex_Gaynor | dkranz: thanks | 22:37 |
| *** mriedem has quit IRC | 22:39 | |
| clarkb | fungi: mordred: https://review.openstack.org/#/c/54639/ seems to be ready and gives us a nice extra requirements list | 22:39 |
| clarkb | zaro: ^ the recent change appears to have fixed it | 22:39 |
| reed | I think we need to re-establish the reply-to in openstack-dev to its sane configuration | 22:40 |
| * reed ranting again | 22:40 | |
| jeblair | clarkb, fungi: i have reproduced the zuul bug! | 22:40 |
| jeblair | clarkb, fungi: moreover, i understand it, and can write a test/fix! | 22:40 |
| fungi | jeblair: that *is* exciting | 22:41 |
| clarkb | jeblair: hurray! | 22:41 |
| clarkb | I think we should all drink a shot now | 22:41 |
| fungi | as long as we don't turn elastic recheck into a drinking game. i don't think my liver would survive that | 22:41 |
| jeblair | fungi: haha! | 22:41 |
| mordred | jeblair: wow! | 22:41 |
| clarkb | I will be very interested in readnig the fix | 22:42 |
| mordred | jeblair: I must tell you I was fully expecting that bug to be un understandable | 22:42 |
| mordred | so I'm impressed | 22:42 |
| * reed cheers at jeblair | 22:43 | |
| fungi | it seemed infrequent and likely hard to recreate the exact conditions | 22:43 |
| jeblair | mordred: i wrote the bug, but i'm pretty sure the reason it's cropped up now is your fault. :) | 22:43 |
| mordred | jeblair: my god. how did I break ZUUL ? | 22:43 |
| reed | mordred, not surprising, you break a lot of stuff | 22:43 |
| reed | :) | 22:43 |
| reed | uhm ... a dude from adobe.com sent a hard-to-parse message to the general list | 22:44 |
| fungi | was it in pdf? (couldn't resist) | 22:46 |
| fungi | was it a list of password hashes? | 22:46 |
| bnemec | Heh, the elastic recheck drinking game. | 22:46 |
| fungi | okay, i'll stop now | 22:46 |
| * bnemec has a new suggestion for his New Year's Eve party... | 22:46 | |
| openstackgerrit | David Ostrovsky proposed a change to openstack-infra/config: Delete melody https://review.openstack.org/59859 | 22:47 |
| reed | fungi, LOL | 22:48 |
| reed | moments of pure entertainment | 22:48 |
| *** krtaylor has joined #openstack-infra | 22:50 | |
| *** Ryan_Lane has joined #openstack-infra | 22:52 | |
| *** fifieldt_ has joined #openstack-infra | 22:52 | |
| *** dkranz has quit IRC | 22:53 | |
| *** michchap has quit IRC | 22:56 | |
| *** Ryan_Lane has quit IRC | 22:56 | |
| *** bpokorny1 has quit IRC | 22:56 | |
| *** CaptTofu has quit IRC | 22:57 | |
| *** rongze has joined #openstack-infra | 22:57 | |
| *** jamesmcarthur has quit IRC | 22:57 | |
| *** jamesmcarthur has joined #openstack-infra | 22:59 | |
| *** jpich has quit IRC | 23:01 | |
| *** mfer has quit IRC | 23:01 | |
| *** michchap has joined #openstack-infra | 23:02 | |
| *** thingee has quit IRC | 23:02 | |
| *** luisg has quit IRC | 23:02 | |
| *** rongze has quit IRC | 23:02 | |
| *** jcoufal has quit IRC | 23:02 | |
| *** thingee has joined #openstack-infra | 23:03 | |
| *** CaptTofu has joined #openstack-infra | 23:03 | |
| *** _david_ has quit IRC | 23:04 | |
| *** fifieldt_ has quit IRC | 23:05 | |
| *** sarob has quit IRC | 23:07 | |
| *** fifieldt_ has joined #openstack-infra | 23:07 | |
| *** sarob has joined #openstack-infra | 23:07 | |
| *** mrodden has joined #openstack-infra | 23:10 | |
| *** sarob has quit IRC | 23:11 | |
| *** hashar has quit IRC | 23:12 | |
| zaro | fungi: i just added instrcutions on how to create maven central account: https://review.openstack.org/#/c/46099 | 23:12 |
| *** markvan has quit IRC | 23:14 | |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/zuul: Fix merging transitively-connected change queues. https://review.openstack.org/59865 | 23:14 |
| fungi | zaro: great--adding to my list of things i'll hopefully look at this evening | 23:16 |
| *** jcooley_ has quit IRC | 23:18 | |
| *** jamesmcarthur has quit IRC | 23:19 | |
| EmilienM | jeblair: for your comment on https://review.openstack.org/#/c/59039/1 - I'm wondering if i have to add the grenade jobs in devstack-gate-periodic-jobs.yaml (there is no job for grenade yet in periodics) | 23:19 |
| jeblair | clarkb: do you want to answer that? ^ | 23:20 |
| EmilienM | fyi, current periodics jobs don't support grenade. I'm adding neutron support to grenade jobs and just wonder if i also add jobs in periodics | 23:21 |
| *** esker has quit IRC | 23:24 | |
| *** giulivo has quit IRC | 23:24 | |
| *** dolphm has quit IRC | 23:25 | |
| *** SergeyLukjanov has quit IRC | 23:26 | |
| sdague | dstufft: you about? just chatting with the heat guys who are blocked on some gating things waiting on pip 1.4.2 to release | 23:26 |
| sdague | curious what the update is on when that's going to happen | 23:26 |
| dstufft | sdague: pip 1.4.2 was killed in favor of 1.5 | 23:26 |
| dstufft | which there is a RC available for | 23:26 |
| sdague | ok, so when does that go to release? | 23:26 |
| dstufft | next week or two probably | 23:26 |
| dstufft | Dec 29th is the deadline for a 1.5 final | 23:27 |
| dstufft | for inclusion in Python 3.4 | 23:27 |
| sdague | hmph | 23:27 |
| dstufft | what problem are they having? | 23:27 |
| sdague | 1.4.1 can't install disk image builder | 23:27 |
| sdague | which they need in the gate | 23:27 |
| dstufft | can 1.5? | 23:28 |
| *** nati_uen_ has joined #openstack-infra | 23:28 | |
| sdague | stevebaker: comments? on above? | 23:28 |
| stevebaker | dstufft: I would say so. What I needed was already fixed in master | 23:29 |
| *** nati_uen_ has quit IRC | 23:29 | |
| *** nati_uen_ has joined #openstack-infra | 23:30 | |
| dstufft | so even if we moved to release a 1.4.2 it'd probably come out after 1.5 at this point | 23:30 |
| *** nati_ueno has quit IRC | 23:30 | |
| dstufft | I'm not sure how to help here besides try to get 1.5 out sooner :) | 23:30 |
| *** dolphm has joined #openstack-infra | 23:32 | |
| dstufft | unless you all are willing to run a pre release version of pip, but I'm not sure what the policy would be on that | 23:32 |
| openstackgerrit | A change was merged to openstack-infra/config: Add gerrit notifications to #storyboard https://review.openstack.org/59143 | 23:32 |
| stevebaker | A workaround could be to install dib in devstack by unpacking the tarball and setup.py install, but I had assumed there would be objections to that | 23:33 |
| fungi | reed: i took pity on vague adobe guy and responded too | 23:34 |
| *** yamahata_ has quit IRC | 23:34 | |
| stevebaker | sdague: but waiting for fedora 20 might be a reasonable workaround | 23:34 |
| sdague | stevebaker: ok | 23:35 |
| sdague | that's next week? | 23:35 |
| *** paul-- has joined #openstack-infra | 23:35 | |
| jeblair | fungi, clarkb, mordred: i'm single-core approving week-old changes. | 23:35 |
| mordred | ++ | 23:35 |
| stevebaker | sdague: next week, assuming the cloud image lands at the same time as the main release | 23:36 |
| fungi | jeblair: fine with me. i'm still trying to work through some that old, but in cases where it needs testing/vetting it's taking time | 23:36 |
| openstackgerrit | A change was merged to openstack-infra/config: Install graphite-web and carbon with pip. https://review.openstack.org/58653 | 23:37 |
| sdague | stevebaker: ok, lets do that. | 23:37 |
| *** mrodden has quit IRC | 23:38 | |
| openstackgerrit | A change was merged to openstack-infra/config: Sort subscriber map alphabetically https://review.openstack.org/58633 | 23:38 |
| openstackgerrit | A change was merged to openstack-infra/config: Add some documentation to the start of subscriber_map https://review.openstack.org/58637 | 23:38 |
| openstackgerrit | A change was merged to openstack-infra/config: Adding in Matt Oliver to rcbau team https://review.openstack.org/58674 | 23:38 |
| dstufft | sdague: stevebaker let me know if there's something pip can do to help out besides trying to get 1.5 out as fast as we can :) | 23:38 |
| openstackgerrit | Yuuichi Fujioka proposed a change to openstack-dev/hacking: checking for metaclass to be Python 3.x compatible https://review.openstack.org/56890 | 23:40 |
| openstackgerrit | A change was merged to openstack-infra/config: Add new openstack-doc-tools project https://review.openstack.org/58623 | 23:40 |
| stevebaker | dstufft: ok, thanks | 23:41 |
| stevebaker | sdague: so just to confirm, is this a halt on any changes to heat-slow tests until they are runnable in gate? | 23:41 |
| openstackgerrit | A change was merged to openstack-infra/config: Add Climate-nova sub-project https://review.openstack.org/58560 | 23:41 |
| openstackgerrit | A change was merged to openstack-infra/config: Climate project: add gerrit tags ACL https://review.openstack.org/58688 | 23:42 |
| openstackgerrit | A change was merged to openstack-infra/config: Move stevedore to stackforge https://review.openstack.org/58389 | 23:42 |
| openstackgerrit | A change was merged to openstack-infra/config: Readd object-storage-manual for Grizzly https://review.openstack.org/58313 | 23:43 |
| *** CaptTofu has quit IRC | 23:44 | |
| openstackgerrit | Jeremy Stanley proposed a change to openstack-infra/git-review: Ignore content of merge commits in reporting https://review.openstack.org/50401 | 23:44 |
| fungi | clarkb: does that updated commit message jove? if so, approving | 23:44 |
| fungi | s/jove/jive/ by jove | 23:45 |
| sdague | stevebaker: we really shouldn't be landing code we can't execute, or have something able to execute them. | 23:45 |
| sdague | so yeh, we should hold until we can get some job running some of this | 23:45 |
| openstackgerrit | A change was merged to openstack-infra/config: Start running regular jobs on reviewstats. https://review.openstack.org/58210 | 23:45 |
| *** rnirmal has quit IRC | 23:45 | |
| sdague | if it's a smaller job, that's cool, or if we can piece together that job so it can work with an rc pip (outside the integrated gate), that's cool as well | 23:46 |
| *** sarob has joined #openstack-infra | 23:46 | |
| stevebaker | sdague: ok, that will be motivating ;) | 23:46 |
| *** slong has joined #openstack-infra | 23:47 | |
| *** dolphm has quit IRC | 23:47 | |
| *** slong-gone has quit IRC | 23:48 | |
| sdague | stevebaker: awesome :) | 23:48 |
| sdague | I look forward to it | 23:48 |
| *** atiwari has quit IRC | 23:49 | |
| *** dstanek_afk has quit IRC | 23:50 | |
| *** mattray has quit IRC | 23:51 | |
| *** loq_mac has joined #openstack-infra | 23:52 | |
| fungi | mordred: did you want to answer lifeless's question on https://review.openstack.org/58106 before i approve it? also, does that additionally address bugs 1251462 and 1257079? | 23:52 |
| uvirtbot | Launchpad bug 1251462 in openstack-ci "Don't require pypi.python.org to install pip" [Undecided,New] https://launchpad.net/bugs/1251462 | 23:52 |
| uvirtbot | Launchpad bug 1257079 in devstack "SyntaxError occured with the retieval of devstack/files/get-pip.py in gate" [Undecided,New] https://launchpad.net/bugs/1257079 | 23:52 |
| fungi | if so, i'll mark them as dupes of bug 1254275 | 23:52 |
| uvirtbot | Launchpad bug 1254275 in openstack-ci "check-swift-devstack-vm-functional fails fetching requirements from pypi.python.org" [Undecided,In progress] https://launchpad.net/bugs/1254275 | 23:52 |
| *** Ryan_Lane has joined #openstack-infra | 23:52 | |
| mordred | lifeless, fungi: uhm. it's cached in the devstack location ... | 23:53 |
| fungi | yeah, i can dig it out of the devstack source code if it's an important detail | 23:53 |
| mordred | fungi: and yes, I believe they are dupes of that | 23:53 |
| mordred | fungi: ALTHOUGH | 23:54 |
| mordred | the devstack bug should probably also be fixed if it's pulling the wrong location | 23:54 |
| fungi | the two components there seem to be that the script gets pulled from github and then retrieves a package from pypi.p.o | 23:54 |
| *** CaptTofu has joined #openstack-infra | 23:55 | |
| lifeless | mordred: ok, so we cache a copy of the get-pip upstream script? | 23:55 |
| lifeless | or the pip that that script gets? | 23:55 |
| mordred | we cache a copy of the git-pip upstream script in d-g | 23:55 |
| lifeless | Cause the thing worrying me is that caching get-pip doesn't get you freedom from internet downloads | 23:55 |
| mordred | lifeless: it can | 23:55 |
| lifeless | you need to cache a copy of the result | 23:55 |
| lifeless | don't you? | 23:55 |
| mordred | nope | 23:55 |
| mordred | well, kinda | 23:56 |
| mordred | you can cache get-pip | 23:56 |
| mordred | and then you can tell get-pip to use the mirror | 23:56 |
| mordred | so get-pip itself can grab pip from the mirror | 23:56 |
| mordred | OR | 23:56 |
| *** openstackgerrit has quit IRC | 23:56 | |
| mordred | if you get really clever, you can tell git-pip to install the pip it finds in itself (which is I believe how I fixed devsack) | 23:56 |
| *** openstackgerrit has joined #openstack-infra | 23:56 | |
| *** kui has quit IRC | 23:56 | |
| lifeless | oh, am blind - read https://raw.github.com/pypa/pip/master/contrib/get-pip.py | 23:56 |
| mordred | so if devstack finds a get-pip script in its cache location | 23:56 |
| openstackgerrit | Emilien Macchi proposed a change to openstack-infra/config: Neutron support in Grenade jobs https://review.openstack.org/59039 | 23:57 |
| *** Ryan_Lane has quit IRC | 23:57 | |
| *** luisg has joined #openstack-infra | 23:57 | |
| lifeless | so | 23:57 |
| lifeless | since get-pip is unversioned | 23:57 |
| *** ^demon|lunch has quit IRC | 23:57 | |
| *** mrodden has joined #openstack-infra | 23:57 | |
| lifeless | when a new pip is released | 23:57 |
| lifeless | we'll install newer pip, not whatever our requirements say we want | 23:58 |
| *** dangers is now known as danger_fo_away | 23:58 | |
| lifeless | ? | 23:58 |
| *** jerryz has joined #openstack-infra | 23:58 | |
| EmilienM | jeblair: I've applied your comment for neutron / grenade patch | 23:58 |
| *** rongze has joined #openstack-infra | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!