| jeblair | mikal: mordred surmised the actual units are 1000s of 512k blocks. which may be crazy or perfectly sane depending on how you think about storage. | 00:00 |
|---|---|---|
| *** sacharya has quit IRC | 00:00 | |
| mikal | this is more johnthetubaguy's thing than mine, but I'll happy to poke somene if that would help | 00:00 |
| clarkb | jeblair: ++ to keepings lots of dumps | 00:01 |
| clarkb | jeblair: just throwing the idea out there that we can reduce it if we need to | 00:01 |
| jeblair | mikal, johnthetubaguy: i just checked iad,ord,dfw and our performance1-8 nodes all have 40G root partitions | 00:02 |
| jeblair | mikal, johnthetubaguy: it seems it's just the new performance2-30 node i just spun up in dfw that has 20G / | 00:03 |
| *** gokrokve has quit IRC | 00:03 | |
| jeblair | clarkb: checking the rax web page, which has units, we are supposed to have a 40G / | 00:04 |
| jeblair | i think this might be a good thing to work out before monday; 20 vs 40G / will make a big difference for us | 00:04 |
| clarkb | ++ | 00:04 |
| jeblair | mikal, johnthetubaguy: how about i open a support ticket? | 00:05 |
| *** nati_ueno has quit IRC | 00:05 | |
| jeblair | mikal: (for context, this is on the new gerrit server we're moving to on monday) | 00:06 |
| *** UtahDave has joined #openstack-infra | 00:06 | |
| clarkb | jeblair: let me know if you need help debugging anything | 00:06 |
| jeblair | huh, i think the device/partition are 40G, but the filesystem is not | 00:08 |
| jeblair | this is from a pvhvm image; i wonder if that has something to do with it | 00:08 |
| jeblair | clarkb: shall i "resize2fs /" ? | 00:09 |
| clarkb | oh | 00:09 |
| clarkb | well we aren't using the node so nothing terrible should happen :) | 00:09 |
| mikal | jeblair: a support ticket sounds like a good idea | 00:09 |
| clarkb | why not, then hopefully this will be corrected on future nodes that we boot | 00:09 |
| mikal | jeblair: but I will fanatically try to help as much as you want | 00:09 |
| mikal | jeblair: bearing in mind I have no access to those systems | 00:09 |
| jeblair | The filesystem on /dev/xvda1 is now 10484164 blocks long. | 00:10 |
| jeblair | poof it's 40G now | 00:10 |
| clarkb | magic | 00:11 |
| *** tjones has left #openstack-infra | 00:12 | |
| *** marcoemorais has quit IRC | 00:13 | |
| jeblair | okay, ticket 140424-00007-3 filed | 00:14 |
| jeblair | since the resize worked and we have 40G now, I think we can proceed and just file that away for the next time we spin up servers (we may want to add that to install_puppet_and_destroy_your_disks.sh) | 00:15 |
| *** marcoemorais has joined #openstack-infra | 00:15 | |
| davidlenwell | way off topic .. but why doesn't http://paste.openstack.org/ have a remove paste button ? | 00:15 |
| mordred | davidlenwell: what would that do? | 00:16 |
| mordred | jeblair: WEIRD | 00:17 |
| clarkb | its unauthenticated, not sure you can do destructive actions | 00:17 |
| davidlenwell | a coworker posted something with a production password and ip in it and he's all crap .. where is the delete button | 00:18 |
| davidlenwell | I advised him to change the password | 00:18 |
| jeblair | clarkb, mordred: i'm thinking maybe we shouldn't have /var/lib/apt on our cinder volume... (based on the idea that we should put only the application data we want to keep around when we, say, replace the server on the cinder volume) | 00:19 |
| clarkb | good point | 00:19 |
| jeblair | so that's pretty much /home/gerrit2/review_site /var/lib/jeepyb and /var/lib/git | 00:19 |
| jeblair | we could try to move all 3 of those to something like /srv | 00:20 |
| clarkb | jeblair: or go the symlink route | 00:20 |
| clarkb | I am not a huge fan of symlinks for that but it is simple | 00:20 |
| *** sweston has quit IRC | 00:21 | |
| jeblair | i bet moving /var/lib/{jeepyb,git} to /srv will be easy. and i bet review_site wouldn't be too hard. | 00:21 |
| clarkb | ya shouldn't be terrible on the new host | 00:21 |
| JayF | Bind mounts can be an interesting path for that sort of thing. Behaves more like actual filesystems, and codified in /etc/fstab they'll be up every boot | 00:21 |
| clarkb | though harder on the existing one if we want to keep them in sync to monday | 00:21 |
| JayF | I actually prefer it to symlinks for things like "move dir X to place Y" when properly changing pointers isn't possible | 00:22 |
| jeblair | JayF: yeah, i like that idea. that might be a good migration strategy too | 00:22 |
| clarkb | wfm | 00:22 |
| clarkb | my only concern with it is again cfg managemtn | 00:22 |
| clarkb | it doesn't play nice with fstab | 00:22 |
| jeblair | bind mount them to /srv, then later move them to actually use /srv | 00:22 |
| JayF | only thing that a bind mount doesn't work on | 00:23 |
| JayF | is if you have things calling rename() | 00:23 |
| JayF | but that doesn't always work for symlinks either | 00:23 |
| davidlenwell | It would be less of an issue if the past ids were uuids instead of sequence numbers, so it;s impossible to guess or iterate over the pastes | 00:23 |
| * JayF had trouble with this setting up an HA chef server, it assumed same root dir == same physical partition and used rename() in that case | 00:23 | |
| davidlenwell | i.e. http://paste.openstack.org/show/76862/ | 00:23 |
| clarkb | davidlenwell: feel free to patch lodgeit or implement a deployment of the thing notmyname showed | 00:24 |
| davidlenwell | which brings up the issue of how much of the paste database is this kind of spam | 00:24 |
| clarkb | davidlenwell: but, I don't think we can do unauthenticated deletions | 00:24 |
| clarkb | and everyone hates the idea of authenticated pate | 00:24 |
| davidlenwell | clarkb: I do as well | 00:24 |
| davidlenwell | like I said .. delete wouldn't be an issue if the id's weren't sequential numbers | 00:25 |
| clarkb | it would be though | 00:25 |
| clarkb | anyone could delete any publicly known uuid | 00:25 |
| jeblair | you could do session cookies so you can at least delete the thing you just pasted | 00:25 |
| clarkb | they aren't secret just because they are uuids | 00:25 |
| davidlenwell | yeah .. thats not fun | 00:25 |
| davidlenwell | jeblair: yeah .. thats a great idea | 00:25 |
| *** UtahDave has quit IRC | 00:25 | |
| jeblair | okay, tomorrow i'll do cinder and look into moving those things to /srv in puppet or bind mounting them | 00:26 |
| clarkb | jeblair: sounds good | 00:26 |
| *** praneshp has quit IRC | 00:26 | |
| clarkb | davidlenwell: fwiw I don't want to run paste.o.o anymore but am in the minority :) | 00:26 |
| davidlenwell | clarkb: running a paste server is a pain in the ass | 00:27 |
| *** fungi has quit IRC | 00:30 | |
| *** gokrokve has joined #openstack-infra | 00:30 | |
| *** melwitt has quit IRC | 00:31 | |
| *** amcrn has joined #openstack-infra | 00:34 | |
| *** gokrokve has quit IRC | 00:34 | |
| *** nati_ueno has joined #openstack-infra | 00:36 | |
| *** fungi has joined #openstack-infra | 00:36 | |
| *** zhiyan_ is now known as zhiyan | 00:37 | |
| *** nati_ueno has quit IRC | 00:40 | |
| *** hogepodge has quit IRC | 00:41 | |
| *** mwagner_lap has joined #openstack-infra | 00:42 | |
| *** talluri has joined #openstack-infra | 00:43 | |
| *** talluri has quit IRC | 00:47 | |
| *** mfer has joined #openstack-infra | 00:48 | |
| *** UtahDave has joined #openstack-infra | 00:48 | |
| openstackgerrit | Steve Kowalik proposed a change to openstack-infra/devstack-gate: Add os-cloud-config into PROJECTS https://review.openstack.org/89978 | 00:51 |
| *** zhiyan has quit IRC | 00:58 | |
| *** kiall has quit IRC | 00:58 | |
| *** proffalken has quit IRC | 00:58 | |
| *** Shrews has quit IRC | 00:58 | |
| *** UtahDave has quit IRC | 00:59 | |
| *** asselin2 has joined #openstack-infra | 00:59 | |
| *** gokrokve has joined #openstack-infra | 00:59 | |
| *** Shrews has joined #openstack-infra | 01:00 | |
| *** zhiyan has joined #openstack-infra | 01:00 | |
| *** fungi has quit IRC | 01:00 | |
| *** proffalken has joined #openstack-infra | 01:00 | |
| *** harlowja has joined #openstack-infra | 01:00 | |
| *** rlandy has quit IRC | 01:01 | |
| *** fungi has joined #openstack-infra | 01:01 | |
| *** kiall has joined #openstack-infra | 01:02 | |
| *** marcoemorais has quit IRC | 01:02 | |
| *** annegentle has quit IRC | 01:02 | |
| *** khyati_ has quit IRC | 01:03 | |
| *** marcoemorais has joined #openstack-infra | 01:03 | |
| *** gokrokve has quit IRC | 01:04 | |
| *** nati_ueno has joined #openstack-infra | 01:10 | |
| clarkb | jhesketh: https://review.openstack.org/#/c/76057/ has actual testing now | 01:11 |
| clarkb | jhesketh: feedback would be great | 01:11 |
| jhesketh | awesome | 01:12 |
| jhesketh | clarkb: I'll take a look but need to read from scratch as it left my mind | 01:12 |
| *** mfer has quit IRC | 01:12 | |
| clarkb | yeah I had to refill caches my self | 01:12 |
| clarkb | jhesketh: the tests are still a bit rudimentary but do exercise the polling | 01:13 |
| *** yaguang has joined #openstack-infra | 01:13 | |
| jhesketh | great | 01:13 |
| openstackgerrit | Monty Taylor proposed a change to openstack-infra/nodepool: Build images using diskimage-builder https://review.openstack.org/46482 | 01:13 |
| openstackgerrit | Monty Taylor proposed a change to openstack-infra/config: WIP: Use ansible instead of direct ssh calls https://review.openstack.org/87821 | 01:15 |
| *** lakshmi has quit IRC | 01:18 | |
| *** yamahata_ has joined #openstack-infra | 01:20 | |
| *** yamahata has joined #openstack-infra | 01:20 | |
| *** mburned is now known as mburned_out | 01:22 | |
| *** changbl has joined #openstack-infra | 01:22 | |
| *** homeless has quit IRC | 01:24 | |
| *** marcoemorais has quit IRC | 01:25 | |
| *** oomichi has joined #openstack-infra | 01:26 | |
| *** nati_ueno has quit IRC | 01:28 | |
| *** sweston has joined #openstack-infra | 01:32 | |
| openstackgerrit | Monty Taylor proposed a change to openstack-infra/config: Add support for disk-image-builder in nodepool https://review.openstack.org/88479 | 01:32 |
| mordred | clarkb: ^^ just used cache_devstack.py directly | 01:33 |
| clarkb | also does that change cover pvhwm stuff? | 01:33 |
| clarkb | extra drivers? | 01:33 |
| *** wchrisj has quit IRC | 01:33 | |
| clarkb | my xen is bad | 01:34 |
| mordred | oh - I didn't touch the xen scripts yet | 01:34 |
| clarkb | we need it for slaves on rax | 01:34 |
| clarkb | not testing xen | 01:34 |
| mordred | gotcha | 01:34 |
| mordred | I thought lifeless said we didn't need to do anything special - just needed to not strip out normal drives from the normal kernel | 01:35 |
| mordred | lifeless: ? | 01:35 |
| *** saschpe has quit IRC | 01:35 | |
| lifeless | mordred: dib? include the baremetal element probably is all. | 01:36 |
| *** nosnos has joined #openstack-infra | 01:36 | |
| *** changbl has quit IRC | 01:36 | |
| mordred | lifeless: but these are images to be loaded into clouds? shoudl we still inlude the bare metal element? | 01:37 |
| *** saschpe has joined #openstack-infra | 01:37 | |
| openstackgerrit | A change was merged to openstack/requirements: Zake 0.0.15 released https://review.openstack.org/86928 | 01:39 |
| *** yjiang has joined #openstack-infra | 01:39 | |
| *** talluri has joined #openstack-infra | 01:44 | |
| *** Ryan_Lane1 has quit IRC | 01:47 | |
| *** talluri has quit IRC | 01:48 | |
| *** changbl has joined #openstack-infra | 01:49 | |
| mestery | sdague: Looks like your "bc" fix had issues, can you reverify this one? It failed on a cinder volume bug: https://review.openstack.org/#/c/89920/ | 01:58 |
| clarkb | mestery: you can reverify it | 01:58 |
| mestery | clarkb: Ah, yes. Doing it now, thanks! | 01:58 |
| *** arosen2 has joined #openstack-infra | 01:59 | |
| clarkb | thanks | 01:59 |
| *** thomasem has joined #openstack-infra | 01:59 | |
| mestery | clarkb: I've seen that issue trip up a few neutron patches today, waiting for it to land before rechecking/verfiying some of those. | 01:59 |
| *** gokrokve has joined #openstack-infra | 01:59 | |
| *** mburned_out is now known as mburned | 01:59 | |
| *** arosen1 has quit IRC | 02:00 | |
| *** gokrokve has quit IRC | 02:00 | |
| mordred | clarkb, jeblair: I'm testing the dib patch now - the part where we have the cache repos script actually just create a dib source-repositories file and let the dib machinery put the source repos in place for us while keeping local caches of the thigns on the dev workstation is pretty sweet | 02:01 |
| mordred | lifeless: ^^ thanks | 02:01 |
| *** gokrokve has joined #openstack-infra | 02:01 | |
| clarkb | mordred: speed up the builds? | 02:01 |
| mordred | yah | 02:01 |
| clarkb | mordred: dib is going to cut out so many cpu hours | 02:01 |
| clarkb | it will be great | 02:02 |
| *** weshay has quit IRC | 02:02 | |
| mordred | clarkb: ultimately, we want to take cache_devstack and re-write it into a few parts I think | 02:02 |
| clarkb | mordred: oh also does your dib stuff cover the local dns resolver and the grub changes? | 02:02 |
| clarkb | I didn't see them but havne't looked too hard yet | 02:02 |
| mordred | clarkb: grub changes yes - dns no | 02:02 |
| mordred | thanks for reminding me - I had put that on my stack and forgot to pop it | 02:02 |
| mordred | clarkb: if we split cache_devstack into two - we can have the part which scans for images to cache create a source-repositories- file like the cache_repos does and let the dib caching machinery do its thing | 02:03 |
| *** thomasem has quit IRC | 02:03 | |
| mordred | which means that the build host will only download most of those images once and then it'll just keep them | 02:03 |
| mordred | then, we let the other part run in-image and do the package pre-warming | 02:04 |
| mordred | it's a bit too many moving pieces for this step | 02:04 |
| clarkb | that sounds good | 02:04 |
| clarkb | will the dib caching download new cirros if say they update the image upstream though? | 02:04 |
| clarkb | or is it only checking the local file is present? | 02:05 |
| mordred | yup | 02:05 |
| mordred | in fact, I want to steal their url caching script for places where we do the same - because it does all of the thing we want it to do | 02:05 |
| *** gokrokve has quit IRC | 02:05 | |
| mordred | clarkb: # Save the nameservers configured by our provider. | 02:05 |
| *** coolsvap has quit IRC | 02:05 | |
| mordred | this does not seem to do what the comment says | 02:06 |
| clarkb | yeah its wrong now because we gave up on provider resolvers | 02:06 |
| mordred | so really we could put the unbound setup into puppet now yeah? | 02:06 |
| *** sweston has quit IRC | 02:08 | |
| *** sacharya1 has quit IRC | 02:08 | |
| clarkb | maybe? I think the static nodes use the provider DNS | 02:08 |
| clarkb | provider DNS is only a problem when grabbing lots of IPs | 02:09 |
| *** yaguang has quit IRC | 02:09 | |
| clarkb | because they don't remove their blacklists afte rnodes are deleted | 02:09 |
| mordred | well, I mean, we have "openstack_project::single_use_slave" | 02:09 |
| *** coolsvap has joined #openstack-infra | 02:10 | |
| mordred | clarkb: (I'm going to do it in dib for now - just pointing it out for whatever) | 02:10 |
| *** sweston has joined #openstack-infra | 02:13 | |
| openstackgerrit | Monty Taylor proposed a change to openstack-infra/config: Add support for disk-image-builder in nodepool https://review.openstack.org/88479 | 02:17 |
| *** sweston has quit IRC | 02:20 | |
| *** signed8bit has quit IRC | 02:21 | |
| *** signed8bit has joined #openstack-infra | 02:22 | |
| mordred | nibalizer: when you are around ... | 02:23 |
| *** zehicle_at_dell has joined #openstack-infra | 02:24 | |
| *** changbl has quit IRC | 02:26 | |
| *** nati_ueno has joined #openstack-infra | 02:26 | |
| jhesketh | clarkb: the zuul stuff looks good... what do you think about making the poll timeout configurable? | 02:31 |
| clarkb | jhesketh: I figured this would come up :) I can certainly do that and make that one test shorter | 02:32 |
| clarkb | jhesketh: do you think 60 seconds is a good default? | 02:32 |
| jhesketh | heh | 02:32 |
| jhesketh | I don't have any data on how gerrit replication works | 02:32 |
| jhesketh | ie, how slow it potentially is | 02:32 |
| jhesketh | we should also consider making the poll interval configurable too | 02:33 |
| clarkb | yeah | 02:33 |
| jhesketh | what if a mirror only syncs every 30min, would it be terrible to poll every 5min and just have zuul hold the jobs in the queue for half an hour | 02:33 |
| clarkb | I think I will go with the current values as defaults and make it configurable | 02:33 |
| clarkb | jhesketh: well gerrit does pushes whenver things change to do replication | 02:34 |
| clarkb | so we shouldn't see delays that large | 02:34 |
| jhesketh | yeah, I'm just thinking outloud about supporting somebody who wants to do really slow replication | 02:34 |
| clarkb | but with really large objects and distances between gerrit and replicas we may see things take more than 60 minutes | 02:34 |
| jhesketh | but I think that's silly | 02:35 |
| clarkb | so making it configurable would be good | 02:35 |
| clarkb | er 60 seconds | 02:35 |
| jhesketh | it certainly can't hurt to add it, but zuul is becoming increasingly complex | 02:35 |
| clarkb | victim of its own success | 02:35 |
| jhesketh | clarkb: so as for having a mixin.. you decided against it? | 02:39 |
| clarkb | jhesketh: yeah, I needed to modify the tests anyways to do the replication | 02:40 |
| clarkb | jhesketh: so I am borrowing now more than copy pasta | 02:40 |
| jhesketh | yep | 02:40 |
| jhesketh | I'm fine with that | 02:40 |
| lifeless | jhesketh: o/ we should put some prose together for pyconau | 02:40 |
| jhesketh | I don't think it's a terrible duplication, they are testing different things but they happen to be similar for weight | 02:40 |
| clarkb | jhesketh: yup | 02:40 |
| jhesketh | lifeless, +1 | 02:40 |
| jhesketh | lifeless: did you see my email to Chris yesterday? | 02:41 |
| lifeless | I think so | 02:41 |
| lifeless | I was up till 3am, so not entirely sure about anything that happened yesterday | 02:41 |
| jhesketh | heh | 02:41 |
| *** amcrn has quit IRC | 02:41 | |
| jhesketh | lifeless: so I figure as the first step(s) to gain some more information around miniconfs to know what we should be doing | 02:41 |
| lifeless | oh right | 02:42 |
| lifeless | yes | 02:42 |
| lifeless | I asked chrisjrn a little in #pyconau | 02:42 |
| lifeless | earlier today | 02:42 |
| *** amotoki has quit IRC | 02:43 | |
| jhesketh | ah cool, will read | 02:44 |
| *** talluri has joined #openstack-infra | 02:45 | |
| jhesketh | lifeless: just sent a reminder to openstack-dev re main conf CFP's | 02:45 |
| jhesketh | didn't mention the miniconf as I want people to propose to the main conference first and we'll do an announce re miniconf later | 02:45 |
| *** NobodyCam has quit IRC | 02:46 | |
| *** morganfainberg is now known as morganfainberg_Z | 02:47 | |
| *** alexpilotti has quit IRC | 02:49 | |
| *** talluri has quit IRC | 02:49 | |
| nibalizer | mordred: hi | 02:52 |
| nibalizer | greghaynes: and I did a pair-review of 88479 so my thoughts are all up there | 02:53 |
| *** yamahata has quit IRC | 02:57 | |
| greghaynes | woo pair reviewing | 02:57 |
| lifeless | reviewing pears. | 02:58 |
| lifeless | mmmm | 02:58 |
| mordred | nibalizer: I put up a patch earlier which kinda is the same as one of your patches except it's different | 02:58 |
| mordred | nibalizer: which makes me think we should, you know, merge them | 02:58 |
| nibalizer | mmmm yes | 02:58 |
| mordred | nibalizer: https://review.openstack.org/#/c/88477/ | 02:58 |
| mordred | nibalizer: although - perhaps I should skip the puppet3 parts of mine and leave that to yours | 02:59 |
| mordred | and just make mine the "What shoudl you do if you find ourself on trusty" patch | 02:59 |
| *** gokrokve has joined #openstack-infra | 02:59 | |
| nibalizer | mordred: yea either way more complexity goes into install_puppet.sh | 02:59 |
| nibalizer | and yea merging my changes and your changes might be a bit hairy | 03:00 |
| nibalizer | i think we should just land your changes and i'll rebase against it | 03:00 |
| nibalizer | its not that big a deal and when I see the final state of what lands to ifcase trusty I can build on that with the puppet3 logic | 03:00 |
| nibalizer | but yea there will be a combination of ... oh wait | 03:00 |
| tchaypo | dstufft: I've just had the (in hindsight not very) startling realisation that in order to test what happens if the --index-url uses file:///, I'm going to have to write an index.html | 03:00 |
| *** praneshp has joined #openstack-infra | 03:01 | |
| nibalizer | wait no | 03:01 |
| dstufft | tchaypo: ok | 03:01 |
| dstufft | tchaypo: just use a py.test tmpdir to do it in | 03:01 |
| nibalizer | yea there are 4 possibilities: precise and puppet2.7, precise and puppet3, trusty and 2.7, trusty and 3 | 03:01 |
| dstufft | or I think there might be some tests/data | 03:01 |
| nibalizer | so it will get gnarly but noprobs | 03:01 |
| mordred | nibalizer: trusty and 2.7 does not seem to be a possibliity | 03:02 |
| tchaypo | dstufft: it probably needs 3 entries in it, so I was thinking of just hand-coding and dropping it into tests/data/packages | 03:02 |
| *** dkranz has joined #openstack-infra | 03:02 | |
| nibalizer | mordred: dangit are there technical limitations? | 03:02 |
| tchaypo | dstufft: unless there's a nice tool that can generate it on the fly.. but then I'd need to add tests for that as well | 03:03 |
| dstufft | tchaypo: dropping it in works for me | 03:03 |
| mordred | nibalizer: there are no repos with puppet 2 for trusty | 03:03 |
| nibalizer | gah! | 03:04 |
| nibalizer | well if we need to there is always gem install | 03:04 |
| nibalizer | but lets hope we dont need that | 03:04 |
| mordred | nah. I think we're going to work the puppet 3 upgrade with thigns still on precise | 03:05 |
| mordred | then do the trusty upgrade | 03:05 |
| nibalizer | if I pull in update_slave from your patch into mine? | 03:05 |
| mordred | except - we WILL do a little puppet3 probably on the single use slaves since they do puppet apply and it all mostly works | 03:05 |
| *** rfolco has quit IRC | 03:07 | |
| *** sandywalsh has quit IRC | 03:08 | |
| *** harlowja is now known as harlowja_away | 03:09 | |
| *** dkranz has quit IRC | 03:09 | |
| *** sandywalsh has joined #openstack-infra | 03:10 | |
| *** Ryan_Lane has joined #openstack-infra | 03:12 | |
| *** asselin2 has quit IRC | 03:16 | |
| nibalizer | mordred: so if i bring update_slave into my puppet3master change do you think we could land it? | 03:16 |
| *** yaguang has joined #openstack-infra | 03:17 | |
| mordred | mayhap? | 03:17 |
| nibalizer | okay ill do that | 03:18 |
| *** prad has quit IRC | 03:19 | |
| *** UtahDave has joined #openstack-infra | 03:29 | |
| *** changbl has joined #openstack-infra | 03:30 | |
| *** HenryG has quit IRC | 03:36 | |
| mordred | clarkb: wait. why wouldn't we manage the cinder volume mounting in puppet | 03:38 |
| clarkb | mordred: because it is too late by that point | 03:39 |
| clarkb | mordred: it needs to happen prior to or during boot | 03:39 |
| clarkb | also uuids make it a pita with puppet | 03:39 |
| mordred | instead of trying to figure out how to deal with seuqncing | 03:39 |
| mordred | we HAVE puppet that is tied to specific hosts - | 03:39 |
| mordred | we could do some mounting things with onlyoif and whatnot | 03:40 |
| clarkb | you could but that doesn't fix the races | 03:40 |
| mordred | clarkb: not for /var/log | 03:40 |
| clarkb | so its mostly useless | 03:40 |
| mordred | for other places | 03:40 |
| clarkb | mordred: and not for elasticsearch | 03:40 |
| mordred | liek, /home/gerrit2 | 03:40 |
| clarkb | gerrit is an exception not rule | 03:40 |
| mordred | since /home/gerrit2 is a thing that puppet creates | 03:40 |
| clarkb | because we isntall gerrit ourselves not from pacakges | 03:40 |
| mordred | so, what you're saying is that the problem is packages and config management :) | 03:41 |
| mordred | ssince they each overlap each other's problem domain | 03:41 |
| clarkb | partially | 03:41 |
| mordred | so they fight | 03:41 |
| clarkb | but the underlying problem is that my infrastructure as a service doesn't know how to communicate to each other so that I can boot a cohesive piece of infrastructure | 03:42 |
| openstackgerrit | Monty Taylor proposed a change to openstack-infra/config: Add support for disk-image-builder in nodepool https://review.openstack.org/88479 | 03:42 |
| mordred | clarkb: well yeah | 03:42 |
| mordred | I'm just still brainstorming on making it work | 03:42 |
| mordred | but I think that it all comes back to AFS | 03:42 |
| clarkb | :) | 03:42 |
| clarkb | I think sdague's idea is not completely terrible | 03:42 |
| clarkb | as we start to use DIB more we could bake the stuff into our images | 03:42 |
| mordred | ++ | 03:43 |
| mordred | I think that's a GREAT idea | 03:44 |
| clarkb | itdoesn't fix the make me a volume and partition format it | 03:44 |
| clarkb | but it does fix the first boot race | 03:44 |
| *** sweston has joined #openstack-infra | 03:47 | |
| mordred | clarkb: AFS fixes the other thing | 03:48 |
| mordred | and I know I'm annoying about tht | 03:48 |
| mordred | but seriously - I think it actually does | 03:48 |
| *** mfer has joined #openstack-infra | 03:50 | |
| *** UtahDave has quit IRC | 03:52 | |
| *** unicell has joined #openstack-infra | 03:54 | |
| *** nati_ueno has quit IRC | 03:55 | |
| *** mfer has quit IRC | 03:55 | |
| *** sweston has quit IRC | 04:01 | |
| *** yfried has quit IRC | 04:01 | |
| *** sweston has joined #openstack-infra | 04:03 | |
| *** coolsvap has quit IRC | 04:05 | |
| openstackgerrit | Monty Taylor proposed a change to openstack-infra/config: Add support for disk-image-builder in nodepool https://review.openstack.org/88479 | 04:06 |
| *** amcrn has joined #openstack-infra | 04:06 | |
| *** coolsvap has joined #openstack-infra | 04:07 | |
| *** gokrokve has quit IRC | 04:08 | |
| *** Longgeek has joined #openstack-infra | 04:10 | |
| *** yamahata has joined #openstack-infra | 04:17 | |
| *** harlowja_away is now known as harlowja | 04:18 | |
| *** marcoemorais has joined #openstack-infra | 04:20 | |
| YorikSar | jeblair: Still around? | 04:20 |
| *** UtahDave has joined #openstack-infra | 04:20 | |
| clarkb | YorikSar: I think he AFK'd a while back. whats up? | 04:20 |
| YorikSar | clarkb: I'm just whaitng when there will be a good moment to upgrade Zuul. | 04:22 |
| clarkb | YorikSar: I think the priority this week is prep for gerrit upgrade. I wouldn't expect zuul upgrade to happen before then | 04:22 |
| YorikSar | clarkb: Ah... Ok. | 04:22 |
| YorikSar | clarkb: I guess we'll have to wait 40-50s for each UI update this week then. | 04:23 |
| clarkb | YorikSar: ya, its possible that gerrit will get done soon and zuul can happen but we are working to make sure gerrit upgrade happens on monday | 04:24 |
| *** yfried has joined #openstack-infra | 04:24 | |
| YorikSar | clarkb: btw, what about those stuck patchsets in check queue? | 04:24 |
| clarkb | YorikSar: there was a gearman blip last night apparently. If you push a new patchset it will move them along | 04:25 |
| YorikSar | clarkb: On Monday? Already? Yay! :) I'll be able to write some bot that will convert PEP8 output to Gerrit inline comments :) | 04:25 |
| clarkb | yup on monday | 04:26 |
| YorikSar | clarkb: REST API will be acessible to everyone? | 04:26 |
| *** talluri has joined #openstack-infra | 04:28 | |
| clarkb | I think so | 04:29 |
| *** yfried has quit IRC | 04:29 | |
| *** talluri has quit IRC | 04:32 | |
| *** yamahata has quit IRC | 04:48 | |
| *** yamahata has joined #openstack-infra | 04:49 | |
| mordred | YorikSar: dude. if you write a bot that converts pep8 output to gerrit inline comments, I will hand you a giant ham | 04:49 |
| YorikSar | mordred: Hey, Monty :) Which part of the world are you in this time of year? | 04:50 |
| YorikSar | mordred: Hmmm... Ham sounds good before breakfast :) | 04:51 |
| *** nati_ueno has joined #openstack-infra | 04:52 | |
| YorikSar | mordred: It's actually not that hard to do. All you need is PEP8 logs (we have them) and API to publish inline comments (it's there in newer versions of Gerrit) | 04:53 |
| *** talluri has joined #openstack-infra | 04:53 | |
| clarkb | you probably want to do that as part of the job itself | 04:53 |
| clarkb | but definitely doable | 04:55 |
| *** afazekas is now known as _afazekas_away | 04:56 | |
| *** chandan_kumar has joined #openstack-infra | 05:00 | |
| *** signed8bit has quit IRC | 05:04 | |
| *** gokrokve has joined #openstack-infra | 05:07 | |
| *** rwsu has quit IRC | 05:07 | |
| *** nati_ueno has quit IRC | 05:07 | |
| *** nosnos has quit IRC | 05:08 | |
| YorikSar | clarkb: I don't think that should be in jobs... Jenkins nodes shouldn't access Gerrit directly. | 05:09 |
| YorikSar | clarkb: But it might be part of Zuul, for example. | 05:09 |
| clarkb | I don't think that belongs in zuul | 05:10 |
| clarkb | :) | 05:10 |
| *** yfried has joined #openstack-infra | 05:10 | |
| YorikSar | clarkb: Yeah... So maybe a separate bot is a good fit :) | 05:10 |
| *** gokrokve has quit IRC | 05:11 | |
| *** marcoemorais has quit IRC | 05:16 | |
| YorikSar | clarkb: I've jusr read through jeblair's message on ML. Why is WIP label kept transient? | 05:21 |
| *** e0ne has joined #openstack-infra | 05:22 | |
| *** rwsu has joined #openstack-infra | 05:23 | |
| clarkb | YorikSar: I suppose you could propose we keep the -1 vote like we keep a code review -2 | 05:25 |
| clarkb | YorikSar: but I think it may be to keep the old behavior | 05:25 |
| *** UtahDave has quit IRC | 05:25 | |
| *** e0ne has quit IRC | 05:26 | |
| YorikSar | clarkb: I wonder if anyone's happy about it. | 05:29 |
| YorikSar | clarkb: Ok, I'll write it to ML then. | 05:29 |
| *** nati_ueno has joined #openstack-infra | 05:33 | |
| *** amotoki has joined #openstack-infra | 05:33 | |
| *** nosnos has joined #openstack-infra | 05:35 | |
| Alex_Gaynor | jeblair, clarkb, mordred: FYI https://github.com/pypa/pip/pull/1743 is the pip fix for the wacko pypy+setuptools issue we hit | 05:49 |
| *** renlt has joined #openstack-infra | 05:52 | |
| *** nati_ueno has quit IRC | 05:58 | |
| clarkb | I love that a method called clobber appaers to be constructive | 05:59 |
| clarkb | Alex_Gaynor: thanks for the heads up. I assume that means we can make pypy go again when we have new pip? | 05:59 |
| clarkb | assuming this PR merges | 05:59 |
| Alex_Gaynor | clarkb: yeah, I'll definitely follow up once that's out :-) | 06:00 |
| clarkb | mordred: jeblair SergeyLukjanov so sdague's d-g bc removal is being erally stubborn in the gate | 06:01 |
| clarkb | I am tempted to force it through but not right before I go to bed | 06:01 |
| clarkb | but something to consider for morning | 06:02 |
| *** jcoufal has joined #openstack-infra | 06:02 | |
| *** nati_ueno has joined #openstack-infra | 06:05 | |
| *** Longgeek has quit IRC | 06:05 | |
| *** gyee has quit IRC | 06:05 | |
| *** Longgeek has joined #openstack-infra | 06:06 | |
| yjiang | Hi,Is there any core member could help?I want to apply a 3rd part test account.My apply email have sent for 2 weeks and no response at all. | 06:06 |
| *** Longgeek has quit IRC | 06:06 | |
| clarkb | yjiang: sorry about that we have been very busy with a variety of things and shorthanded | 06:06 |
| clarkb | third party test accounts have been a low priority the last few weeks | 06:07 |
| *** Longgeek has joined #openstack-infra | 06:07 | |
| *** gokrokve has joined #openstack-infra | 06:07 | |
| yjiang | clarkb: OK.thx.If I want to renew some part of the apply information,Should I sent another email? | 06:11 |
| clarkb | yjiang: is the information changing? | 06:11 |
| yjiang | clarkb: Yes. | 06:11 |
| clarkb | if so yes, just respond to your original email | 06:11 |
| *** gokrokve has quit IRC | 06:12 | |
| clarkb | also please avoid needing to change the info once the requiest is complete | 06:12 |
| *** Longgeek_ has joined #openstack-infra | 06:12 | |
| yjiang | clarkb: OK,thx a lot. | 06:12 |
| *** pcrews has quit IRC | 06:13 | |
| *** ildikov has quit IRC | 06:15 | |
| *** Longgeek has quit IRC | 06:16 | |
| *** jlibosva has joined #openstack-infra | 06:18 | |
| *** zehicle_at_dell has quit IRC | 06:23 | |
| *** harlowja is now known as harlowja_away | 06:30 | |
| renlt | hi infra team, I want add a new project in stackforge, I already read the process http://ci.openstack.org/stackforge.html, now I'll commit my initial project config, my question is do I need create a Launchpad bug for this commit? Thanks in advance. | 06:30 |
| clarkb | renlt: no you don't but please make sure your gerrit topic is 'new-project' | 06:31 |
| clarkb | renlt: you can do that by doing `git review -t new-project` when you push your change | 06:31 |
| *** afazekas has joined #openstack-infra | 06:32 | |
| *** flaper87|afk is now known as flaper87 | 06:38 | |
| renlt | clarkb: 'new-project' literally, not 'my-project-name' right? | 06:39 |
| clarkb | correct | 06:40 |
| clarkb | we use that topic to filter all the new projects in gerrit | 06:40 |
| clarkb | makes it easier to batch them up | 06:40 |
| renlt | clarkb: many thanks! | 06:40 |
| rcarrillocruz | morning | 06:51 |
| *** _nadya_ has joined #openstack-infra | 06:52 | |
| SergeyLukjanov | <day_time> folks! | 06:56 |
| *** sweston is now known as sweston|away | 06:57 | |
| *** sweston|away is now known as sweston | 06:57 | |
| *** sandywalsh has quit IRC | 07:03 | |
| openstackgerrit | A change was merged to openstack-infra/config: Handle additional reqs push failure as non failing https://review.openstack.org/89924 | 07:04 |
| *** sandywalsh has joined #openstack-infra | 07:04 | |
| *** amcrn has quit IRC | 07:04 | |
| *** lukego has joined #openstack-infra | 07:08 | |
| *** gokrokve has joined #openstack-infra | 07:08 | |
| *** ihrachyshka has joined #openstack-infra | 07:09 | |
| *** UtahDave has joined #openstack-infra | 07:10 | |
| *** e0ne has joined #openstack-infra | 07:11 | |
| *** marcoemorais has joined #openstack-infra | 07:12 | |
| *** gokrokve has quit IRC | 07:12 | |
| *** e0ne has quit IRC | 07:16 | |
| *** e0ne has joined #openstack-infra | 07:17 | |
| *** _nadya_ has quit IRC | 07:18 | |
| *** ihrachyshka has quit IRC | 07:19 | |
| *** ihrachyshka has joined #openstack-infra | 07:20 | |
| *** praneshp has quit IRC | 07:20 | |
| *** _nadya_ has joined #openstack-infra | 07:21 | |
| *** reed has joined #openstack-infra | 07:24 | |
| *** olaph has quit IRC | 07:28 | |
| *** rediskin has joined #openstack-infra | 07:33 | |
| openstackgerrit | Le Tian Ren proposed a change to openstack-infra/config: Create powervc-driver project under Stackforge https://review.openstack.org/90017 | 07:37 |
| *** arosen2 has quit IRC | 07:41 | |
| *** hashar has joined #openstack-infra | 07:47 | |
| *** andreaf has joined #openstack-infra | 07:50 | |
| openstackgerrit | Le Tian Ren proposed a change to openstack-infra/config: Create powervc-driver project under Stackforge https://review.openstack.org/90017 | 07:51 |
| *** rcarrill` has joined #openstack-infra | 07:53 | |
| *** rcarrillocruz has quit IRC | 07:56 | |
| *** ildikov has joined #openstack-infra | 07:56 | |
| *** sandywalsh_ has joined #openstack-infra | 08:04 | |
| *** matrohon has joined #openstack-infra | 08:06 | |
| *** andreaf has quit IRC | 08:06 | |
| *** sandywalsh has quit IRC | 08:06 | |
| *** gokrokve has joined #openstack-infra | 08:09 | |
| *** pblaho has joined #openstack-infra | 08:11 | |
| *** jamielennox is now known as jamielennox|away | 08:12 | |
| johnthetubaguy | jeblair: for when you are up, did you get the performance node issues sorted? If you used PVHVM based images (probably a good idea) the base disk doesn't get auto sized upto 40 GB right now (I have a patch pending in gerrit to fix that, if you need it) | 08:12 |
| *** mrmartin_ has joined #openstack-infra | 08:13 | |
| *** gokrokve has quit IRC | 08:13 | |
| mrmartin_ | re | 08:14 |
| mrmartin_ | SergeyLukjanov are you here? | 08:14 |
| *** e0ne has quit IRC | 08:15 | |
| *** ildikov_ has joined #openstack-infra | 08:16 | |
| *** jpich has joined #openstack-infra | 08:17 | |
| *** e0ne has joined #openstack-infra | 08:17 | |
| *** nosnos has quit IRC | 08:18 | |
| *** derekh has joined #openstack-infra | 08:18 | |
| *** dizquierdo has joined #openstack-infra | 08:19 | |
| *** ildikov has quit IRC | 08:20 | |
| SergeyLukjanov | mrmartin_, yup | 08:22 |
| *** julienvey has joined #openstack-infra | 08:22 | |
| mrmartin_ | SergeyLukjanov: can you check the zuul? | 08:22 |
| SergeyLukjanov | mrmartin_, what's the problem? | 08:23 |
| *** darraghb has joined #openstack-infra | 08:24 | |
| mrmartin_ | https://review.openstack.org/#/c/90018 | 08:27 |
| mrmartin_ | I cannot see this job in zuul status page. | 08:28 |
| mrmartin_ | and some of the jobs there running more then 20 hours | 08:29 |
| mrmartin_ | can you check that? | 08:29 |
| *** _nadya_ has quit IRC | 08:30 | |
| SergeyLukjanov | mrmartin_, there were some problems with gear, but it works ok now afaik | 08:32 |
| SergeyLukjanov | mrmartin_, re your patch, it already have +1 from jenkins | 08:33 |
| mrmartin_ | SergeyLukjanov, ok great, thank you | 08:33 |
| mrmartin_ | if you have a little time today, may I ask a review for this two patches: https://review.openstack.org/#/c/88323/ https://review.openstack.org/#/c/88219/ | 08:34 |
| *** UtahDave has quit IRC | 08:34 | |
| mrmartin_ | both of them sitting there since a week | 08:34 |
| *** gmoro has quit IRC | 08:35 | |
| mrmartin_ | we need to clone you somehow, I want to see more infra-team members in the EU timezone | 08:35 |
| mrmartin_ | :) | 08:35 |
| *** jooools has joined #openstack-infra | 08:37 | |
| *** andreaf has joined #openstack-infra | 08:40 | |
| *** markmc has joined #openstack-infra | 08:42 | |
| davidlenwell | hey SergeyLukjanov.. what do you think about being able to see images commited from review pages? i.e. https://review.openstack.org/#/c/88587/ | 08:45 |
| *** medieval1 has quit IRC | 08:45 | |
| SergeyLukjanov | mrmartin_, please, don't clone me :) | 08:45 |
| *** medieval1 has joined #openstack-infra | 08:45 | |
| SergeyLukjanov | davidlenwell, hey | 08:46 |
| davidlenwell | hi | 08:46 |
| *** nati_ueno has quit IRC | 08:46 | |
| davidlenwell | thought I'd talk about it with a core member of infra before maybe filing it as a bug then maybe coding it | 08:46 |
| openstackgerrit | A change was merged to openstack-infra/storyboard: Creator field added to a Task model https://review.openstack.org/87626 | 08:46 |
| *** gmoro has joined #openstack-infra | 08:47 | |
| SergeyLukjanov | davidlenwell, it sounds good, but it's really "nice to have" prio level | 08:47 |
| *** reed has quit IRC | 08:47 | |
| SergeyLukjanov | davidlenwell, probably we could search for some existing gerrit plugins | 08:47 |
| davidlenwell | sure .. but its also kinda low hanging | 08:47 |
| *** reed has joined #openstack-infra | 08:47 | |
| davidlenwell | kinda what I was thinking .. since I don't really know how gerrit builds that page .. I probably don't wanna know | 08:48 |
| *** e0ne_ has joined #openstack-infra | 08:48 | |
| *** dkehn has quit IRC | 08:49 | |
| SergeyLukjanov | davidlenwell, after short googling it seems like it's already implemented and we just should enable it | 08:50 |
| SergeyLukjanov | davidlenwell, https://gerrit-review.googlesource.com/Documentation/config-gerrit.html#mimetype | 08:50 |
| davidlenwell | neat | 08:50 |
| davidlenwell | is that file in config? | 08:50 |
| SergeyLukjanov | davidlenwell, yup, I'll take a look on current config file and send CR to discuss it with other infra folks | 08:51 |
| *** e0ne has quit IRC | 08:51 | |
| davidlenwell | one place I can start to help is on any interface stuff... | 08:52 |
| SergeyLukjanov | davidlenwell, there was a CR to embed zuul results into the change request page in gerrit | 08:53 |
| davidlenwell | I like that idea | 08:54 |
| *** rediskin has left #openstack-infra | 08:54 | |
| davidlenwell | like I said I don't currently know how that page builds .. but we use forked gerrit right? | 08:54 |
| SergeyLukjanov | davidlenwell, yup, it's forked, you can fid sources in openstack-infra/gerrit | 08:55 |
| SergeyLukjanov | review.o.o will be upgraded to the latest 2.8.X next week | 08:55 |
| *** dkehn has joined #openstack-infra | 08:56 | |
| davidlenwell | cool | 08:56 |
| *** ihrachyshka has quit IRC | 08:58 | |
| openstackgerrit | Sergey Lukjanov proposed a change to openstack-infra/config: Display images in reviews https://review.openstack.org/90030 | 08:59 |
| SergeyLukjanov | davidlenwell, ^^ | 08:59 |
| davidlenwell | nice .. thanks.. thats bad ass .. | 08:59 |
| davidlenwell | I was totally looking for where to put that .. you beat me to it .. | 08:59 |
| SergeyLukjanov | davidlenwell, oops, sorry | 09:01 |
| davidlenwell | no worries ... | 09:01 |
| *** Longgeek_ has quit IRC | 09:01 | |
| openstackgerrit | Nikita Konovalov proposed a change to openstack-infra/storyboard: Introducing the event timeline https://review.openstack.org/87959 | 09:01 |
| davidlenwell | its still awesome to be able to bring it up in irc then BAM .. patch .. kick ass | 09:02 |
| openstackgerrit | Sergey Lukjanov proposed a change to openstack-infra/config: Fix indention in gerrit.config.erb https://review.openstack.org/90031 | 09:03 |
| *** marcoemorais has quit IRC | 09:06 | |
| openstackgerrit | A change was merged to openstack-infra/devstack-gate: change time calculation to not require bc https://review.openstack.org/89920 | 09:06 |
| openstackgerrit | Nikita Konovalov proposed a change to openstack-infra/storyboard: Update a sample projects file https://review.openstack.org/90032 | 09:07 |
| *** _nadya_ has joined #openstack-infra | 09:08 | |
| *** ominakov has joined #openstack-infra | 09:09 | |
| *** renlt has quit IRC | 09:09 | |
| *** _nadya_ has quit IRC | 09:09 | |
| *** Longgeek has joined #openstack-infra | 09:09 | |
| *** ociuhandu has quit IRC | 09:10 | |
| *** gokrokve has joined #openstack-infra | 09:10 | |
| *** asselin2 has joined #openstack-infra | 09:12 | |
| *** maxbit has joined #openstack-infra | 09:12 | |
| *** gokrokve has quit IRC | 09:14 | |
| *** sweston has quit IRC | 09:16 | |
| *** asselin2 has quit IRC | 09:16 | |
| *** hashar has quit IRC | 09:19 | |
| openstackgerrit | Sergey Lukjanov proposed a change to openstack-infra/config: Extract experimental-tripleo jobs to template https://review.openstack.org/84026 | 09:20 |
| *** oomichi has quit IRC | 09:20 | |
| *** miarmak has quit IRC | 09:23 | |
| *** ociuhandu has joined #openstack-infra | 09:24 | |
| *** miarmak has joined #openstack-infra | 09:27 | |
| *** zhiyan is now known as zhiyan_ | 09:32 | |
| *** marcoemorais has joined #openstack-infra | 09:33 | |
| *** marcoemorais1 has joined #openstack-infra | 09:35 | |
| *** marcoemorais has quit IRC | 09:38 | |
| *** marcoemorais1 has quit IRC | 09:39 | |
| *** hashar has joined #openstack-infra | 09:52 | |
| reed | does anyone here know where the skin of wiki.openstack.org lives? | 09:53 |
| *** rcarrillocruz has joined #openstack-infra | 09:54 | |
| *** rcarrill` has quit IRC | 09:56 | |
| *** zhiyan_ is now known as zhiyan | 09:58 | |
| *** xchu_ has joined #openstack-infra | 10:01 | |
| openstackgerrit | A change was merged to openstack-infra/config: Add example of comment formatting https://review.openstack.org/85930 | 10:02 |
| *** yjiang has quit IRC | 10:04 | |
| *** ihrachyshka has joined #openstack-infra | 10:06 | |
| *** xchu_ has quit IRC | 10:06 | |
| openstackgerrit | Sean Dague proposed a change to openstack-infra/devstack-gate: Revert "Revert "update grenade branches"" https://review.openstack.org/90043 | 10:07 |
| *** dims has quit IRC | 10:08 | |
| *** zhiyan is now known as zhiyan_ | 10:08 | |
| *** ihrachyshka has quit IRC | 10:09 | |
| *** ihrachyshka has joined #openstack-infra | 10:10 | |
| *** hashar has quit IRC | 10:10 | |
| *** gokrokve has joined #openstack-infra | 10:11 | |
| *** Longgeek has quit IRC | 10:12 | |
| openstackgerrit | Nikita Konovalov proposed a change to openstack-infra/storyboard: Added test for migrations https://review.openstack.org/90047 | 10:12 |
| davidlenwell | SergeyLukjanov: any idea why this failed pep8? https://review.openstack.org/#/c/90024/ the logs aren't making sense to me. | 10:13 |
| SergeyLukjanov | davidlenwell, from logs: | 10:14 |
| SergeyLukjanov | ./nova/virt/libvirt/utils.py:230:80: E501 line too long (81 > 79 characters) | 10:14 |
| openstackgerrit | Nikita Konovalov proposed a change to openstack-infra/storyboard: Update a sample projects file https://review.openstack.org/90032 | 10:14 |
| *** gokrokve has quit IRC | 10:15 | |
| *** nosnos has joined #openstack-infra | 10:15 | |
| *** dims has joined #openstack-infra | 10:19 | |
| *** Longgeek has joined #openstack-infra | 10:20 | |
| *** e0ne_ has quit IRC | 10:20 | |
| *** mburned is now known as mburned_out | 10:23 | |
| openstackgerrit | Nikita Konovalov proposed a change to openstack-infra/storyboard: Added test for migrations https://review.openstack.org/90047 | 10:25 |
| *** ihrachyshka has quit IRC | 10:30 | |
| *** marcoemorais has joined #openstack-infra | 10:36 | |
| openstackgerrit | Bob Ball proposed a change to openstack-infra/nodepool: Check for stale PID lock when starting https://review.openstack.org/90052 | 10:39 |
| *** marcoemorais has quit IRC | 10:40 | |
| *** jcoufal has quit IRC | 10:41 | |
| *** waterkinfe has joined #openstack-infra | 10:45 | |
| *** waterkinfe has quit IRC | 10:46 | |
| *** waterkinfe has joined #openstack-infra | 10:47 | |
| *** darraghb has quit IRC | 10:48 | |
| *** darraghb has joined #openstack-infra | 10:48 | |
| *** waterkinfe has quit IRC | 10:52 | |
| *** waterkinfe has joined #openstack-infra | 10:52 | |
| openstackgerrit | Bob Ball proposed a change to openstack-infra/nodepool: Check for stale PID lock when starting https://review.openstack.org/90052 | 10:53 |
| *** mburned_out is now known as mburned | 10:58 | |
| openstackgerrit | Sean Dague proposed a change to openstack-infra/config: index n-sched in grenade runs https://review.openstack.org/90054 | 11:00 |
| openstackgerrit | Sean Dague proposed a change to openstack-infra/elastic-recheck: adjust the cgroup bug to use n-cpu https://review.openstack.org/90057 | 11:03 |
| *** Clabbe has quit IRC | 11:03 | |
| *** Clabbe has joined #openstack-infra | 11:04 | |
| *** e0ne has joined #openstack-infra | 11:05 | |
| *** lukego has quit IRC | 11:09 | |
| *** e0ne has quit IRC | 11:10 | |
| *** gokrokve has joined #openstack-infra | 11:11 | |
| *** e0ne has joined #openstack-infra | 11:12 | |
| *** gokrokve has quit IRC | 11:16 | |
| openstackgerrit | Nikita Konovalov proposed a change to openstack-infra/storyboard: Introducing the event timeline https://review.openstack.org/87959 | 11:19 |
| openstackgerrit | A change was merged to openstack-infra/elastic-recheck: adjust the cgroup bug to use n-cpu https://review.openstack.org/90057 | 11:34 |
| *** jgallard has joined #openstack-infra | 11:36 | |
| *** marcoemorais has joined #openstack-infra | 11:37 | |
| *** Clabbe has quit IRC | 11:39 | |
| *** ihrachyshka has joined #openstack-infra | 11:40 | |
| *** Clabbe has joined #openstack-infra | 11:41 | |
| *** marcoemorais has quit IRC | 11:41 | |
| *** adalbas has joined #openstack-infra | 11:42 | |
| openstackgerrit | A change was merged to openstack-infra/devstack-gate: Add os-cloud-config into PROJECTS https://review.openstack.org/89978 | 11:44 |
| *** chandan_kumar has quit IRC | 11:45 | |
| mattymo | SergeyLukjanov, why is the launchpad+gerrit bug broken? :( | 11:47 |
| SergeyLukjanov | mattymo, terrorists probably | 11:49 |
| SergeyLukjanov | mattymo, is it not working for all events or for specific CR/bug/bp? | 11:51 |
| *** Longgeek_ has joined #openstack-infra | 11:52 | |
| mattymo | there is a specific one here: https://review.openstack.org/89832 | 11:53 |
| *** Longgeek has quit IRC | 11:53 | |
| *** HenryG has joined #openstack-infra | 11:55 | |
| *** Longgeek_ has quit IRC | 11:57 | |
| *** ildikov has joined #openstack-infra | 11:59 | |
| *** rfolco has joined #openstack-infra | 12:02 | |
| SergeyLukjanov | mattymo, comment was posted to the issue | 12:02 |
| *** pdmars has joined #openstack-infra | 12:02 | |
| *** ildikov_ has quit IRC | 12:02 | |
| SergeyLukjanov | mattymo, and comment was added to https://blueprints.launchpad.net/fuel/+spec/pack-all-eggs-and-gems | 12:03 |
| SergeyLukjanov | so, I don't see the issue | 12:03 |
| mattymo | oh ok | 12:03 |
| mattymo | carry on :) | 12:03 |
| *** Longgeek has joined #openstack-infra | 12:04 | |
| *** Longgeek_ has joined #openstack-infra | 12:04 | |
| *** doug-fish has left #openstack-infra | 12:04 | |
| *** hashar has joined #openstack-infra | 12:07 | |
| *** Longgeek has quit IRC | 12:08 | |
| *** salv-orlando has joined #openstack-infra | 12:09 | |
| *** lazy_prince has joined #openstack-infra | 12:09 | |
| lazy_prince | Hi all.. I see that couple of tasks are still stuck over 20hrs.. is there any problem with Zuul..? | 12:12 |
| *** gokrokve has joined #openstack-infra | 12:12 | |
| *** AJaeger has quit IRC | 12:12 | |
| *** gokrokve has quit IRC | 12:17 | |
| hashar | lazy_prince: I have no clue but potentially you could retrigger the openstack-infra/config check change https://review.openstack.org/82384 | 12:20 |
| hashar | by commenting "recheck no bug" | 12:20 |
| hashar | that might reenqueue it | 12:20 |
| hashar | and thus discard the stuck change | 12:20 |
| hashar | for the two tripleo changes I have no idea. The same trick might work as well or the job might depends on tripleo instances or something | 12:20 |
| *** prad has joined #openstack-infra | 12:22 | |
| *** mbacchi has joined #openstack-infra | 12:24 | |
| lazy_prince | Well.. I could see that other tasks are getting executed.. but there are couple of them that are queued but not yet executed. I will retrigger the checks again to see if it helps. | 12:24 |
| lazy_prince | I checked and seems like someone already added recheck no bug and it is still in the queue.. | 12:25 |
| *** AJaeger has joined #openstack-infra | 12:31 | |
| *** rlandy has joined #openstack-infra | 12:31 | |
| *** prad has quit IRC | 12:31 | |
| *** marcoemorais has joined #openstack-infra | 12:37 | |
| *** dizquierdo has quit IRC | 12:39 | |
| SergeyLukjanov | lazy_prince, if recheck no bug doesn't help, try to push new patch set | 12:39 |
| sdague | lazy_prince: you have to push a new patch set | 12:41 |
| sdague | there was a zuul issue yesterday where it lost track of some changes | 12:41 |
| *** nosnos has quit IRC | 12:42 | |
| *** marcoemorais has quit IRC | 12:42 | |
| *** waterkinfe has quit IRC | 12:45 | |
| openstackgerrit | David Caro proposed a change to openstack-infra/jenkins-job-builder: Added parallelization options https://review.openstack.org/75514 | 12:46 |
| sdague | oh, damn, we passed 90k | 12:47 |
| sdague | https://review.openstack.org/#/c/90000 | 12:48 |
| *** lukego has joined #openstack-infra | 12:50 | |
| *** ildikov has quit IRC | 12:50 | |
| *** ildikov has joined #openstack-infra | 12:51 | |
| *** mark0n has joined #openstack-infra | 12:52 | |
| *** weshay has joined #openstack-infra | 12:52 | |
| SergeyLukjanov | sdague, yeah, 100k is coming | 12:56 |
| *** ildikov has quit IRC | 12:56 | |
| *** liam__ has joined #openstack-infra | 12:57 | |
| sdague | 43 days since 80k | 12:57 |
| sdague | IIRC that's about the same window as 70k -> 80k | 12:58 |
| *** NobodyCam has joined #openstack-infra | 12:58 | |
| sdague | so ~ June 6th | 12:58 |
| sdague | +/- a couple of days | 12:58 |
| openstackgerrit | Nikita Konovalov proposed a change to openstack-infra/storyboard-webclient: TimeLine events support https://review.openstack.org/88405 | 12:59 |
| *** skolekonov has quit IRC | 13:00 | |
| *** alexpilotti has joined #openstack-infra | 13:01 | |
| *** zehicle_at_dell has joined #openstack-infra | 13:02 | |
| *** signed8bit has joined #openstack-infra | 13:02 | |
| lazy_prince | sdague: SergeyLukjanov: Thanks. I will push a new change soon. but what will happen to the check thats running right now.. | 13:05 |
| lazy_prince | will it get cancelled once new change is pushed..? | 13:06 |
| SergeyLukjanov | lazy_prince, I'll be canceled because of new patch set | 13:06 |
| lazy_prince | okay.. | 13:06 |
| *** ildikov has joined #openstack-infra | 13:06 | |
| mark0n | Jenkins Job Builder: It seems like "python setup.py install" does not install pbr automatically - the README doesn't mention this dependency but when I run JJB it fails with "pkg_resources.DistributionNotFound: pbr>=0.5.21,<1.0". When I run "easyinstall pbr" I think if the installer cannot install this automatically this additional step should be documented. | 13:07 |
| *** Clabbe has quit IRC | 13:08 | |
| *** yfried has quit IRC | 13:09 | |
| *** gokrokve has joined #openstack-infra | 13:13 | |
| *** coolsvap is now known as coolsvap|afk | 13:13 | |
| *** jhesketh has quit IRC | 13:13 | |
| openstackgerrit | David Caro proposed a change to openstack-infra/jenkins-job-builder: Added recursive option https://review.openstack.org/85106 | 13:13 |
| *** coolsvap|afk is now known as coolsvap | 13:13 | |
| *** bknudson has quit IRC | 13:14 | |
| phschwartz | morning infra | 13:14 |
| *** dprince has joined #openstack-infra | 13:15 | |
| *** signed8b_ has joined #openstack-infra | 13:15 | |
| *** AJaeger has quit IRC | 13:16 | |
| *** mriedem has joined #openstack-infra | 13:17 | |
| *** AJaeger has joined #openstack-infra | 13:17 | |
| *** gokrokve has quit IRC | 13:17 | |
| *** signed8bit has quit IRC | 13:18 | |
| *** markmcclain has joined #openstack-infra | 13:19 | |
| *** julim has joined #openstack-infra | 13:20 | |
| *** markmcclain has quit IRC | 13:20 | |
| anteaya | morning phschwartz | 13:22 |
| *** yaguang has quit IRC | 13:25 | |
| *** jhesketh has joined #openstack-infra | 13:25 | |
| *** Clabbe has joined #openstack-infra | 13:30 | |
| *** changbl has quit IRC | 13:31 | |
| *** dkranz has joined #openstack-infra | 13:32 | |
| *** smarcet has joined #openstack-infra | 13:35 | |
| *** lazy_prince has quit IRC | 13:35 | |
| *** zehicle_at_dell has quit IRC | 13:36 | |
| anteaya | mark0n: I agree, would you like to document it? | 13:37 |
| anteaya | http://git.openstack.org/cgit/openstack-infra/jenkins-job-builder/tree/README.rst | 13:38 |
| *** marcoemorais has joined #openstack-infra | 13:38 | |
| *** marun has joined #openstack-infra | 13:39 | |
| anteaya | reed I do not know where the skin of wiki.o.o lives | 13:41 |
| *** bknudson has joined #openstack-infra | 13:41 | |
| reed | anteaya, thanks... Ryan_Lane should know | 13:41 |
| *** yolanda has joined #openstack-infra | 13:42 | |
| *** marcoemorais has quit IRC | 13:42 | |
| *** doug-fish has joined #openstack-infra | 13:43 | |
| *** _nadya_ has joined #openstack-infra | 13:45 | |
| *** _nadya_ has quit IRC | 13:45 | |
| *** _nadya_ has joined #openstack-infra | 13:45 | |
| *** jaypipes has joined #openstack-infra | 13:46 | |
| *** devoid has joined #openstack-infra | 13:46 | |
| *** zns has joined #openstack-infra | 13:48 | |
| *** zns has quit IRC | 13:48 | |
| *** zns has joined #openstack-infra | 13:48 | |
| anteaya | reed: yes, he probably will | 13:48 |
| *** zns has quit IRC | 13:51 | |
| *** thomasem has joined #openstack-infra | 13:51 | |
| *** zns has joined #openstack-infra | 13:54 | |
| *** markwash has joined #openstack-infra | 13:55 | |
| *** zhiyan_ is now known as zhiyan | 13:55 | |
| *** coolsvap is now known as coolsvap|afk | 13:57 | |
| *** unicell has quit IRC | 13:58 | |
| *** che-arne has joined #openstack-infra | 13:58 | |
| openstackgerrit | Marc Abramowitz proposed a change to openstack-infra/jenkins-job-builder: Refactor the `matches` function https://review.openstack.org/88678 | 14:01 |
| *** gokrokve has joined #openstack-infra | 14:01 | |
| *** zns has quit IRC | 14:03 | |
| *** homeless has joined #openstack-infra | 14:03 | |
| *** zz_gondoi is now known as gondoi | 14:03 | |
| *** zns has joined #openstack-infra | 14:07 | |
| *** prad has joined #openstack-infra | 14:14 | |
| *** thedodd has joined #openstack-infra | 14:15 | |
| *** wchrisj has joined #openstack-infra | 14:16 | |
| *** wchrisj has left #openstack-infra | 14:16 | |
| openstackgerrit | Martin Konrad proposed a change to openstack-infra/jenkins-job-builder: Document installation of PBR library https://review.openstack.org/90092 | 14:17 |
| mark0n | anteaya: done | 14:18 |
| openstackgerrit | João Vale proposed a change to openstack-infra/jenkins-job-builder: Allow using template variables in "defaults" field. https://review.openstack.org/85336 | 14:19 |
| anteaya | mark0n: thank you | 14:23 |
| *** atiwari has joined #openstack-infra | 14:24 | |
| *** rcarrill` has joined #openstack-infra | 14:25 | |
| openstackgerrit | Vladislav Kuzmin proposed a change to openstack-infra/nodepool: Add layer for work with KVM/QEMU hosts https://review.openstack.org/88554 | 14:26 |
| *** devoid has quit IRC | 14:26 | |
| *** rcarrillocruz has quit IRC | 14:27 | |
| *** signed8b_ is now known as signed8bit | 14:27 | |
| *** markmcclain has joined #openstack-infra | 14:28 | |
| *** UtahDave has joined #openstack-infra | 14:28 | |
| *** mwagner_lap has quit IRC | 14:30 | |
| openstackgerrit | Victor Sergeyev proposed a change to openstack-infra/config: Import oslo.db repository https://review.openstack.org/78955 | 14:31 |
| *** adalbas has quit IRC | 14:32 | |
| *** UtahDave has quit IRC | 14:36 | |
| *** marcoemorais has joined #openstack-infra | 14:39 | |
| derekh | hi, would anybody mind taking a look at https://review.openstack.org/#/c/86842/ I think I've rebased it correctly this time | 14:39 |
| *** UtahDave has joined #openstack-infra | 14:40 | |
| *** marcoemorais has quit IRC | 14:43 | |
| *** adalbas has joined #openstack-infra | 14:44 | |
| SergeyLukjanov | derekh, hey, it already have 2 x +2, so, it's waiting for someone to babysitting | 14:45 |
| openstackgerrit | afazekas proposed a change to openstack-infra/elastic-recheck: Cirros boot failed https://review.openstack.org/90100 | 14:45 |
| derekh | SergeyLukjanov: ahh ok, it just got the second one there, thanks | 14:46 |
| anteaya | okay so I am clicking around on review-dev and currently when I try to open any patch via the GUI I am told I don't have access: https://review-dev.openstack.org/#/q/status:open,n,z | 14:47 |
| anteaya | I am trying to distill the goodness sweston has offered in this etherpad: https://etherpad.openstack.org/p/ai9irqk3FC | 14:48 |
| jeblair | oops; i'll check and see if the apache config was messed up again | 14:49 |
| *** hashar has quit IRC | 14:49 | |
| pleia2 | good morning | 14:49 |
| *** pmathews has joined #openstack-infra | 14:49 | |
| jeblair | pleia2: good morning | 14:50 |
| anteaya | jeblair: thanks and good morning | 14:51 |
| anteaya | pleia2: and good morning to you too | 14:51 |
| *** pcrews has joined #openstack-infra | 14:51 | |
| Longgeek_ | I want to cry. | 14:53 |
| jeblair | ah this is the problem: https://review.openstack.org/#/c/60893 | 14:53 |
| Longgeek_ | https://bugs.launchpad.net/openstack-ci/+bug/1306967 | 14:53 |
| jeblair | it's based on a WIP change | 14:53 |
| uvirtbot | Launchpad bug 1306967 in openstack-ci "thstack-ci service account" [Undecided,New] | 14:53 |
| pleia2 | Longgeek_: usually fungi is the one who handles those and he's on vacation, hopefully someone else will get to it soon, sorry for the delay | 14:54 |
| *** mrmartin_ has quit IRC | 14:54 | |
| *** aglarendil|nb has joined #openstack-infra | 14:54 | |
| anteaya | pleia2 Longgeek_ clarkb is aware and it is on his list | 14:54 |
| pleia2 | anteaya: thanks :) | 14:54 |
| anteaya | Longgeek_: I will attract your attention to my comments | 14:55 |
| anteaya | I have a concern about the name that will appear in the comments | 14:55 |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/config: Gerrit-2.8: Allow encoded path separators in URLs https://review.openstack.org/60893 | 14:55 |
| pleia2 | anteaya: well it's the company name.. | 14:55 |
| pleia2 | seems a bit hard to get around that | 14:55 |
| anteaya | when I see thstack it looks too much like thestack, which I wonder if others will think is our system | 14:55 |
| pleia2 | and fwiw, I don't make that connection | 14:56 |
| anteaya | I just want to ensure new devs aren't confused about what set of tests they reference | 14:56 |
| Longgeek_ | thestack is openstack system? | 14:56 |
| anteaya | it is possible it is just me | 14:56 |
| anteaya | Longgeek_: it doesn't take much to confuse new developers | 14:57 |
| anteaya | and they don't tend to ask questions | 14:57 |
| anteaya | they just operate on incorrect information and get frustrated | 14:57 |
| anteaya | then the post blogs about their frustration | 14:57 |
| AJaeger | ;( | 14:57 |
| anteaya | exactly | 14:57 |
| anteaya | perhaps it is just me | 14:57 |
| Longgeek_ | "ThStack" is our company name, and we design 'Th' as the prefix to connect 'Stack' to show we based on OpenStack. It's really different between 'Th' and 'The', we want to express "Think" not "The", it's used to show our brief about cloud and OpenStack contribution, unfortunately we can't apply 'thinkstack.com' as our domain name, so 'th' is a good choice. thanks | 14:58 |
| *** mfer has joined #openstack-infra | 14:58 | |
| Longgeek_ | I believe you can see our email address | 14:58 |
| anteaya | I'm not finding fault with your company name | 14:58 |
| anteaya | or why you chose it | 14:58 |
| anteaya | just trying to think like a new developer | 14:59 |
| AJaeger | anteaya, you have a good point. The question is how to avoid this confusion. | 15:00 |
| *** markwash has quit IRC | 15:00 | |
| anteaya | right | 15:00 |
| anteaya | which includes many options | 15:00 |
| AJaeger | Longgeek_, www.thstack.com does not exist as webpage | 15:00 |
| anteaya | one of which is renaming our system from Jenkins to something taht is more obvious to new developers | 15:00 |
| *** pmathews has quit IRC | 15:00 | |
| AJaeger | "OpenStack official CI" ;) | 15:01 |
| anteaya | or just OpenStack | 15:01 |
| anteaya | but yes, something along those lines | 15:01 |
| anteaya | Steve I think was an option at one point | 15:01 |
| anteaya | :D | 15:01 |
| Longgeek_ | Our site has not yet developed, still in the design phase. The company was founded less than a month. | 15:01 |
| anteaya | Longgeek_: I am not finding fault with your company or any of your naming choices | 15:02 |
| *** pmathews has joined #openstack-infra | 15:02 | |
| anteaya | since we have so many 3rd party testing accounts, we need to help developers sort though the comments | 15:02 |
| *** sweston has joined #openstack-infra | 15:02 | |
| Longgeek_ | How to solve that problem? | 15:02 |
| jeblair | anteaya: theirs is no more confusing than any of the other 40 account names | 15:03 |
| *** dims has quit IRC | 15:03 | |
| *** wenlock has joined #openstack-infra | 15:03 | |
| anteaya | jeblair: okay, I seem to be the only one | 15:03 |
| anteaya | which is great, it is easy to solve my problem | 15:03 |
| anteaya | Longgeek_: yes, it is a new problem since we have so many accounts now | 15:04 |
| *** chandan_kumar has joined #openstack-infra | 15:04 | |
| anteaya | it will come up at the summit, I am sure | 15:04 |
| jeblair | it's scheduled | 15:04 |
| anteaya | yay | 15:04 |
| anteaya | thank you jeblair | 15:04 |
| Longgeek_ | I do not want to wait until the end of the summit, right? | 15:05 |
| *** dims has joined #openstack-infra | 15:05 | |
| krtaylor | yes, glad it is scheduled, saw that! | 15:05 |
| jeblair | Longgeek_: no, someone will make the account soon | 15:06 |
| krtaylor | I'll add descriptive naming to the etherpad I'm keeping | 15:06 |
| fifieldt | fyi launchpad looks like it's having problems | 15:07 |
| * fifieldt just saw a launchpad an error he'd never seen before! | 15:07 | |
| Longgeek_ | ok, thank you | 15:08 |
| *** rediskin has joined #openstack-infra | 15:10 | |
| *** ihrachyshka has quit IRC | 15:10 | |
| anteaya | fifieldt: thanks, comcast subscribers were having lp issues the other day | 15:10 |
| anteaya | fifieldt: you aren't using comcast are you? | 15:10 |
| fifieldt | erm | 15:11 |
| fifieldt | I don't think comcast coverage reaches this far west :) | 15:11 |
| pleia2 | fifieldt: I'm having trouble with other bits of the canonical-run infrastructure as well, all seems slow today | 15:11 |
| fifieldt | yeah, I saw a maint page with their IRC channel name on it | 15:11 |
| fifieldt | which was odd :) | 15:11 |
| pleia2 | yeah, I saw that yesterday while trying to file a bug | 15:12 |
| *** pmathews has quit IRC | 15:12 | |
| *** pmathews has joined #openstack-infra | 15:12 | |
| fifieldt | anteaya, my provider is "中華電信“ | 15:12 |
| AJaeger | fifieldt, how do I pronounce that one? ;( | 15:13 |
| anteaya | fifieldt: so not comcast | 15:13 |
| fifieldt | Zhōnghuá Diànxìn :) | 15:13 |
| fifieldt | aka ChungHwa Telecom | 15:13 |
| fifieldt | anyway, just wanted to report in :) | 15:14 |
| krtaylor | ls | 15:14 |
| anteaya | sweston: I am selecting picking from your work on this etherpad: https://etherpad.openstack.org/p/ai9irqk3FC | 15:14 |
| anteaya | fifieldt: thank you | 15:14 |
| openstackgerrit | A change was merged to openstack-infra/config: Gerrit-2.8: Allow encoded path separators in URLs https://review.openstack.org/60893 | 15:15 |
| *** zns has quit IRC | 15:15 | |
| *** coolsvap|afk is now known as coolsvap | 15:15 | |
| *** aglarendil|nb has left #openstack-infra | 15:16 | |
| sweston | anteaya: in a meeting at the moment, may I get back to you? | 15:16 |
| *** david-lyle has joined #openstack-infra | 15:16 | |
| *** chandan_kumar has quit IRC | 15:16 | |
| anteaya | sweston: of course | 15:17 |
| *** mrodden has joined #openstack-infra | 15:17 | |
| openstackgerrit | Marc Abramowitz proposed a change to stackforge/python-jenkins: Add tests, make it work on Python 3 https://review.openstack.org/90114 | 15:18 |
| mordred | morning all | 15:19 |
| *** afazekas has quit IRC | 15:21 | |
| ttx | morn | 15:21 |
| rediskin | jeblair fungi sdague mordred im really looking forward to +A this tiny patch @_@ https://review.openstack.org/#/c/89562/ | 15:21 |
| anteaya | morning mordred | 15:22 |
| rediskin | to finally can see fancy html charts | 15:22 |
| *** msabramo has joined #openstack-infra | 15:23 | |
| *** rcarrill` is now known as rcarrillocruz | 15:25 | |
| * anteaya assumes jeblair is still figuring out apache config for review-dev, since she can't yet see the patch review view | 15:25 | |
| *** markmc has quit IRC | 15:25 | |
| jeblair | anteaya: i linked to the patch earlier that fixes it. it's in the process of merging. | 15:25 |
| devananda | is there a guideline for what kinds of external libs should be in requirements, compared to optional / third-party things that shouldn't be, and how to test those? | 15:25 |
| devananda | g'morning, mordred | 15:25 |
| anteaya | jeblair: ah okay, I misunderstood, thank you | 15:25 |
| msabramo | jeblair: Jenkins jobs for python-jenkins will go live tomorrow (Friday)? | 15:26 |
| devananda | anteaya: my neutron question to you yesterday is: does it support other dhcp providers (besides dnsmasq)? | 15:26 |
| msabramo | jeblair: I was able to submit a review (https://review.openstack.org/90114) but it fired off a "noop" | 15:26 |
| anteaya | jeblair: you removed the wip dependency? | 15:26 |
| jeblair | msabramo: no, creating projects was the only thing we were doing weekly; changes to jobs will happen as people approve changes | 15:27 |
| *** mburned is now known as mburned_out | 15:27 | |
| jeblair | msabramo: there might be a bit of a delay with that because we are short-staffed and those of us that are here are largely focusing on the gerrit upgrade | 15:27 |
| msabramo | jeblair: Ah okday | 15:27 |
| msabramo | okay | 15:27 |
| anteaya | devananda: ah that would be a neutron code question, which I am embarrassed to say I never figured out | 15:28 |
| anteaya | devananda: so I do not know the answer to that question | 15:28 |
| anteaya | :( | 15:28 |
| devananda | anteaya: ah, np. i thought there might be some community knowledge there too. | 15:29 |
| anteaya | right, reasonable assumption | 15:29 |
| anteaya | I'm empty | 15:29 |
| *** maxbit has quit IRC | 15:30 | |
| jeblair | mordred, clarkb: so i'm wondering if we shouldn't actually put the jeepyb cache and local git mirror in /opt, since they are both recreatable? (i mean, actually var is a great place for them, aside from the disk space problem) | 15:32 |
| jeblair | mordred, clarkb: and /opt seems to be our "place to put things that take up a lot of space but we don't care about preserving" | 15:33 |
| jeblair | since it's where we typically mount the ephemeral disks | 15:33 |
| mordred | jeblair: sounds great to me | 15:34 |
| *** mburned_out is now known as mburned | 15:34 | |
| mordred | it also has the benefit of being locally attached and therefor probably quicker | 15:34 |
| *** Longgeek_ has quit IRC | 15:34 | |
| mordred | so for the apache git mirror, you know, no backend iscsi hop | 15:34 |
| *** devoid has joined #openstack-infra | 15:35 | |
| anteaya | so puppetdb report times? browser default time? http://puppetdb.openstack.org/report/review-dev.openstack.org/8715afee9349ca94963f4666e336456d58904c70 | 15:35 |
| jeblair | mordred: are you sure ephemeral disks are local? | 15:35 |
| jeblair | anteaya: i believe so; i think i opened an issue about that | 15:35 |
| mordred | jeblair: I _believe_ so - I believe that's why they are considered 'ephemeral' | 15:35 |
| anteaya | jeblair: k, thanks | 15:35 |
| mordred | jeblair: but I'm frequently wrong | 15:36 |
| jeblair | mordred: yeah, though the system disk is presumably local and is not | 15:36 |
| jeblair | but then again, maybe neither of them are | 15:36 |
| *** changbl has joined #openstack-infra | 15:36 | |
| mordred | jogo, sdague ^^ ? | 15:36 |
| jeblair | i think someone at rax would be the right person to answer that :) | 15:37 |
| sdague | they should be local, unless your cloud provider did something crazy with nova | 15:37 |
| *** mwagner_lap has joined #openstack-infra | 15:38 | |
| anteaya | jeblair: I don't see a bug about it: http://bit.ly/1iiFD6G | 15:38 |
| *** msabramo has quit IRC | 15:38 | |
| mordred | BWAHAHAHAHAHAAHHAAH | 15:38 |
| jeblair | i'm not at all sure why nova should care _where_ the disks backing the storage it uses are :) | 15:38 |
| anteaya | I can file unless I missed it | 15:38 |
| mordred | I spent an hour trying to figure out why a script in my dib elements wasn't being run | 15:38 |
| *** yfried has joined #openstack-infra | 15:38 | |
| mordred | it's quite simply because I did not actually reference the element anywhere | 15:38 |
| jeblair | anteaya: it's not our bug, it's puppetboard's | 15:38 |
| anteaya | ah | 15:38 |
| mordred | it had nothign to do with naming or sequencing or me misunderstanding the system | 15:39 |
| *** rediskin has left #openstack-infra | 15:39 | |
| *** adarazs has quit IRC | 15:40 | |
| anteaya | jeblair: found it: https://github.com/nedap/puppetboard/issues/63 | 15:40 |
| *** adarazs has joined #openstack-infra | 15:41 | |
| anteaya | yay, I can see patches in review-dev again | 15:41 |
| *** mrmartin has joined #openstack-infra | 15:41 | |
| anteaya | how do I override "server certificate verification failed" to clone from review-dev? http://paste.openstack.org/show/76962/ | 15:44 |
| *** msabramo has joined #openstack-infra | 15:45 | |
| *** ominakov has quit IRC | 15:49 | |
| mordred | anteaya: there is a git setting ... | 15:50 |
| mordred | anteaya: although you can also just clone with ssh | 15:50 |
| anteaya | mordred: yes, trying to remember where I put the clone over ssh command | 15:51 |
| anteaya | I will find it | 15:51 |
| *** markmcclain has quit IRC | 15:53 | |
| *** MarkAtwood has joined #openstack-infra | 15:54 | |
| anteaya | how long after I add a new ssh public key to review-dev before it sees it? | 15:55 |
| mordred | should be immediate | 15:55 |
| *** ihrachyshka has joined #openstack-infra | 15:56 | |
| *** pmathews1 has joined #openstack-infra | 15:56 | |
| anteaya | or do I not have access to stream-events, ls-projects or show-queue? | 15:56 |
| anteaya | I remember I don't have access to stream-events now | 15:56 |
| anteaya | as a registered user | 15:56 |
| anteaya | are the others verboten to registered users as well? | 15:57 |
| *** zns has joined #openstack-infra | 15:57 | |
| *** pmathews has quit IRC | 15:57 | |
| *** melwitt has joined #openstack-infra | 15:58 | |
| *** jlibosva has quit IRC | 15:58 | |
| mordred | not sure ... | 15:59 |
| anteaya | k | 15:59 |
| anteaya | it might just be me | 15:59 |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/config: Make gerrit git replica and jeepyb paths configurable https://review.openstack.org/90128 | 16:00 |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/config: Move local git replica on review to /opt https://review.openstack.org/90129 | 16:00 |
| jeblair | i believe we need to grant access to stream-events | 16:00 |
| jeblair | anteaya: try now | 16:01 |
| jeblair | mordred, clarkb: would you please review https://review.openstack.org/90128 ? (and https://review.openstack.org/90129 though less important) | 16:02 |
| *** dangers_away has quit IRC | 16:02 | |
| *** gothicmindfood has joined #openstack-infra | 16:02 | |
| *** ihrachyshka has quit IRC | 16:02 | |
| *** dims has quit IRC | 16:03 | |
| mordred | jeblair: on it | 16:03 |
| *** dangers_away has joined #openstack-infra | 16:03 | |
| *** ihrachyshka has joined #openstack-infra | 16:03 | |
| *** jlibosva has joined #openstack-infra | 16:04 | |
| *** sweston has quit IRC | 16:05 | |
| *** dims has joined #openstack-infra | 16:06 | |
| *** krtaylor has quit IRC | 16:08 | |
| anteaya | anita@anita-hp:~$ ssh -p 29418 anteaya@review-dev.openstack.org gerrit stream-events | 16:09 |
| anteaya | Permission denied (publickey). | 16:09 |
| anteaya | :( | 16:09 |
| jeblair | anteaya: that's more likely to be related to having the right public key configured than gerrit permissions | 16:09 |
| jeblair | anteaya: does "ls-projects" work? | 16:09 |
| anteaya | I did cat .ssh/id_rsa.pub and copied the output from my terminal and pasted | 16:10 |
| *** tru_tru has quit IRC | 16:10 | |
| anteaya | like I have done before | 16:10 |
| anteaya | ls-projects and show-queue give me the same response | 16:10 |
| anteaya | permission denied (publickey) | 16:10 |
| *** tru_tru has joined #openstack-infra | 16:10 | |
| anteaya | I will repaste my publickey | 16:10 |
| anteaya | nope | 16:11 |
| anteaya | no change | 16:11 |
| *** hashar has joined #openstack-infra | 16:12 | |
| *** homeless has quit IRC | 16:12 | |
| jeblair | anteaya: check your username | 16:12 |
| anteaya | ah | 16:12 |
| anteaya | that be it | 16:12 |
| anteaya | d'oh | 16:12 |
| *** imcsk8 has quit IRC | 16:13 | |
| anteaya | username, I am still akuno to review-dev | 16:13 |
| *** jaypipes has quit IRC | 16:13 | |
| *** imcsk8 has joined #openstack-infra | 16:13 | |
| *** homeless has joined #openstack-infra | 16:14 | |
| *** _afazekas_away is now known as afazekas | 16:14 | |
| *** marcoemorais has joined #openstack-infra | 16:17 | |
| *** dims has quit IRC | 16:17 | |
| *** dims has joined #openstack-infra | 16:17 | |
| *** zns has quit IRC | 16:18 | |
| *** pmathews1 has quit IRC | 16:18 | |
| *** pblaho has quit IRC | 16:19 | |
| jeblair | I've pasted the ssh host key warning that people will get into the etherpad at https://etherpad.openstack.org/p/ai9irqk3FC | 16:19 |
| *** pmathews has joined #openstack-infra | 16:19 | |
| jeblair | along with the command that people should use in order to resolve the error | 16:20 |
| clarkb | ++ to /opt for jeepyb and mirror | 16:21 |
| clarkb | I'm reviewing puppet change now | 16:21 |
| openstackgerrit | A change was merged to openstack/requirements: Add PyECLib as a requirement https://review.openstack.org/76068 | 16:24 |
| *** chandan_kumar has joined #openstack-infra | 16:25 | |
| *** jgallard has quit IRC | 16:25 | |
| anteaya | jeblair: great thanks | 16:26 |
| *** gyee has joined #openstack-infra | 16:26 | |
| *** lakshmi has joined #openstack-infra | 16:27 | |
| zaro | monring | 16:28 |
| *** talluri has quit IRC | 16:31 | |
| clarkb | devananda requirements should contain compatibly licensed packages that are actively supported with support for python 2 and 3 | 16:31 |
| zaro | fooy, i can't rax slave. thinking about turning off gerrit tests | 16:31 |
| *** mburned is now known as mburned_out | 16:31 | |
| clarkb | not everything meets that criteria because a lot of stuff was grandfathered in. but additions will hopefully be able to do that | 16:32 |
| devananda | clarkb: "actively supported" -- how does that apply to vendor published libraries? | 16:32 |
| jeblair | clarkb, devananda: https://wiki.openstack.org/wiki/Requirements | 16:32 |
| *** zns has joined #openstack-infra | 16:32 | |
| *** gokrokve_ has joined #openstack-infra | 16:32 | |
| anteaya | morning zaro | 16:32 |
| *** zehicle_at_dell has joined #openstack-infra | 16:33 | |
| mordred | jeblair: +2 - you wanna land it? | 16:33 |
| devananda | jeblair: heh, nice. python-seamicroclient is the reason i'm asking, but i'm sure it'll come up for other libs | 16:33 |
| anteaya | jeblair: I blended in the instructions for the ssh host key change | 16:33 |
| jeblair | mordred: thx done | 16:33 |
| anteaya | edits welcome | 16:33 |
| * anteaya goes to eat some lunch | 16:33 | |
| clarkb | devananda well are they licensed compatibly? if so they should be open source. you can see what their dev is like | 16:34 |
| clarkb | looks like apache2 hosted on github | 16:35 |
| *** mburned_out is now known as mburned | 16:35 | |
| devananda | clarkb: yes | 16:35 |
| zaro | anybody willing to review https://review.openstack.org/#/c/88026 ? would like to get that in and tested before gerrit upgrade. | 16:35 |
| clarkb | with no issues or pull requests but last commit is from 4 days ago | 16:36 |
| devananda | clarkb: is "apache2 and on github" enough to be added to global req's, and tested in ironic's gate? that seems iffy to me | 16:36 |
| *** gokrokve has quit IRC | 16:36 | |
| clarkb | no see my other criteria :) | 16:37 |
| *** liam__ has joined #openstack-infra | 16:37 | |
| *** Ryan_Lane has quit IRC | 16:37 | |
| clarkb | but its certainly a step in the right direction | 16:37 |
| devananda | clarkb: put another way - gate runs unit and functional tests, but we can't run functional tests of a third-party driver that requires special hardware | 16:38 |
| *** liam__ has quit IRC | 16:38 | |
| openstackgerrit | A change was merged to openstack-infra/config: Make gerrit git replica and jeepyb paths configurable https://review.openstack.org/90128 | 16:39 |
| clarkb | right Iam not sure where the gate got involved | 16:39 |
| devananda | clarkb: if the library isn't in global-req's, we can't run those unit tests in the gate. if it is, we can. | 16:40 |
| clarkb | openstack/requirements is a list of dependencies. it helps keep all projects in syncon common depsand makes it easier to know what needs packaging | 16:40 |
| devananda | clarkb: but fast-forward to when ironic graduates, now all projects are gated on eg. python-seamicroclient. | 16:40 |
| clarkb | sure but you cant run those tests regardless | 16:40 |
| clarkb | so its not worth worrying about from a gate perspective | 16:41 |
| devananda | why can't we run those unit tests? | 16:43 |
| *** jlibosva has quit IRC | 16:43 | |
| clarkb | no no you said functional tests with special hardware | 16:43 |
| devananda | right | 16:43 |
| devananda | i mean, we can run the unit tests IFF that lib is in global requiremetns. but should we do that, considering that we /cant/ run the functional tests | 16:43 |
| clarkb | I think those unittests belong in python seamicroclient | 16:44 |
| clarkb | ironic doesnt unittest external libs | 16:44 |
| devananda | clarkb: i maen, run ironic's unit tests for the driver that uses said external lib | 16:45 |
| *** zns has quit IRC | 16:45 | |
| mordred | right. you need the lib to be in global reqs if ironic needs the lib for ironic's unittests | 16:45 |
| devananda | clarkb: http://git.openstack.org/cgit/openstack/ironic/tree/ironic/drivers/modules/seamicro.py | 16:45 |
| *** zns has joined #openstack-infra | 16:46 | |
| clarkb | what do you need the lib for if you dont have hardware? | 16:46 |
| *** marun_ has joined #openstack-infra | 16:46 | |
| clarkb | I think we are using "unittest" very liberally here | 16:46 |
| *** marun has quit IRC | 16:47 | |
| *** marun_ is now known as marun | 16:47 | |
| devananda | clarkb: ah | 16:47 |
| clarkb | any test that needs aspecial lib to talk to special hardware is a functional test | 16:48 |
| devananda | clarkb: the driver could be unit tested without the lib by mocking it, yes | 16:49 |
| clarkb | right | 16:49 |
| devananda | clarkb: so i think my question is answered -- third-party drivers that can't be functionally tested in the gate should not have their external dependencies added to requirements, and unit tests just need to mock them | 16:50 |
| *** hashar has quit IRC | 16:50 | |
| devananda | thanks | 16:50 |
| *** dkliban is now known as dkliban_afk | 16:50 | |
| *** harlowja_away is now known as harlowja | 16:51 | |
| *** _nadya_ has quit IRC | 16:52 | |
| *** lakshmi has quit IRC | 16:52 | |
| *** LarsN has joined #openstack-infra | 16:53 | |
| *** shardy is now known as shardy_afk | 16:57 | |
| *** jpich has quit IRC | 16:58 | |
| *** jpeeler has quit IRC | 16:58 | |
| *** jpeeler has joined #openstack-infra | 16:59 | |
| *** jpeeler has quit IRC | 16:59 | |
| *** jpeeler has joined #openstack-infra | 16:59 | |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/config: Gerrit: Ensure /opt/lib exists if used https://review.openstack.org/90147 | 16:59 |
| jeblair | clarkb, mordred: ^ minor tested fix to previous patch | 16:59 |
| clarkb | jeblair: can you just do if not defined File[$replicate_path] then file { $replicate_path: ensure => directory } | 17:00 |
| *** zns has quit IRC | 17:01 | |
| *** signed8bit has quit IRC | 17:01 | |
| jeblair | clarkb: hrm, i was responding to this error: | 17:01 |
| *** sweston has joined #openstack-infra | 17:01 | |
| jeblair | err: /Stage[main]/Openstack_project::Gerrit/File[/opt/lib/git]/ensure: change from absent to directory failed: Cannot create /opt/lib/git; parent directory /opt/lib does not exist | 17:01 |
| *** e0ne has quit IRC | 17:01 | |
| clarkb | oh I see, this is a problem with the root of the path | 17:01 |
| clarkb | and puppet doesn't mkdir -p | 17:02 |
| clarkb | that change is a bit hacky but I can't come up with anything better right now +2 | 17:02 |
| mordred | +@ | 17:02 |
| mordred | +2 | 17:02 |
| clarkb | I approved since mordred got the first +2 | 17:03 |
| jeblair | clarkb: i'd love to see the correct and elegant way to do that in puppet | 17:03 |
| *** zhiyan is now known as zhiyan_ | 17:03 | |
| *** darraghb has quit IRC | 17:04 | |
| clarkb | me too. I'm guessing it involves a git repo class that sorts out the recursive path stuff for you | 17:04 |
| *** thedodd has quit IRC | 17:04 | |
| jeblair | it's also a little weird because we ensure that local_git_dir is created in openstack_project::gerrit | 17:04 |
| mordred | exec { 'make dir': command => "mkdir -p $path", creates => $path } | 17:05 |
| jeblair | i'm not sure why the idea of local replication is split across those two modules | 17:05 |
| *** signed8bit has joined #openstack-infra | 17:05 | |
| *** signed8bit has quit IRC | 17:05 | |
| mordred | jeblair: ++ should probably be consolidated | 17:05 |
| jeblair | mordred: so you think that exec should replace the existing File['/opt/lib/git']? | 17:05 |
| *** thedodd has joined #openstack-infra | 17:06 | |
| *** signed8bit has joined #openstack-infra | 17:06 | |
| *** sweston has quit IRC | 17:06 | |
| jeblair | mordred: but then you'd have to "requires => Exec['make dir']", yeah? | 17:06 |
| *** SumitNaiksatam has quit IRC | 17:06 | |
| mordred | yeah | 17:07 |
| mordred | it's not really good | 17:07 |
| mordred | we could make a Path type that does that - but then we're going way over the top | 17:07 |
| clarkb | http://www.puppetcookbook.com/posts/creating-a-directory-tree.html | 17:07 |
| clarkb | apparently you pass an array of dirs | 17:07 |
| jeblair | clarkb: i don't think that actually solves the problem | 17:08 |
| *** coolsvap is now known as coolsvap|afk | 17:08 | |
| jeblair | clarkb: our target directory is a variable | 17:08 |
| clarkb | jeblair: right so anytime your target dir changes your array variable needs updating | 17:08 |
| jeblair | we can't set it to be an array and then go and use it in templates | 17:08 |
| *** gokrokve_ has quit IRC | 17:08 | |
| clarkb | this doesn't work well with the templates | 17:08 |
| clarkb | ya | 17:08 |
| jeblair | perhaps what we should actually do is not give the value a default | 17:09 |
| jeblair | and have the api for the gerrit module be "the caller ensures the directory is created" | 17:09 |
| clarkb | wfm | 17:09 |
| mordred | ++ | 17:10 |
| mordred | and also, we should bitch that puppet doesn't have a mkdir -p | 17:11 |
| * jeblair glares at puppet | 17:11 | |
| *** msabramo1 has joined #openstack-infra | 17:12 | |
| *** morganfainberg_Z is now known as morganfainberg | 17:12 | |
| clarkb | NobodyCam: have you been able to confirm that ironic is happy on the tripleo ci images now? | 17:13 |
| *** msabramo has quit IRC | 17:13 | |
| NobodyCam | clarkb: ironic is not happy.. the the changes are in the nodes now... Ty | 17:13 |
| *** derekh has quit IRC | 17:15 | |
| openstackgerrit | Khai Do proposed a change to openstack-infra/config: temporary disable gerrit tests https://review.openstack.org/90154 | 17:16 |
| zaro | jeblair: ^ need your help to get the gerrit patches in. | 17:17 |
| *** Ajaeger1 has joined #openstack-infra | 17:17 | |
| *** ildikov has quit IRC | 17:18 | |
| *** chandan_kumar has quit IRC | 17:19 | |
| *** chandan_kumar has joined #openstack-infra | 17:19 | |
| *** doug-fish1 has joined #openstack-infra | 17:20 | |
| devananda | clarkb, jeblair: should "can be functionally tested upstream" be added to the requirements guidelines? | 17:20 |
| anteaya | clarkb: when you have a chance, how do we access the important changes custom dashboard you created? https://etherpad.openstack.org/p/ai9irqk3FC | 17:20 |
| clarkb | devananda: I am not entirely sure it is that black and white in all cases | 17:20 |
| clarkb | devananda: thinking of test libs themselves and so on | 17:20 |
| *** SumitNaiksatam has joined #openstack-infra | 17:20 | |
| devananda | clarkb: i'm looking at some other requirements entries like hp3parclient | 17:21 |
| *** zns has joined #openstack-infra | 17:21 | |
| *** doug-fish has quit IRC | 17:21 | |
| devananda | if none of the tests in the gate exercise a requirement, how do we know it actually works, and if we can't know that, why is it required? | 17:22 |
| clarkb | devananda: fwiw I was trying to decouple gating from requirements :) I don't think we gate on it but I can see the lib being in requirements anyways | 17:22 |
| openstackgerrit | Marton Kiss proposed a change to openstack-infra/config: Create groups-static-pages project https://review.openstack.org/89734 | 17:22 |
| jeblair | zaro: wfm | 17:22 |
| clarkb | devananda: the requirements repo itself should document why it is required | 17:22 |
| jeblair | clarkb, mordred: can you review/aprv https://review.openstack.org/90154 for zaro? | 17:23 |
| anteaya | clarkb: thank you | 17:23 |
| clarkb | jeblair: sure | 17:23 |
| *** maxbit has joined #openstack-infra | 17:23 | |
| clarkb | jeblair: zaro done | 17:24 |
| devananda | clarkb: eg, http://git.openstack.org/cgit/openstack/requirements/commit/?id=669056d88937a8c629020ee1b89602843cb6f0c4 | 17:24 |
| clarkb | devananda: yes. I have been pushing pretty hard when I review requirements that people properly document stuff in the commits themselves | 17:25 |
| clarkb | devananda: it is really important for figuring out if a thing should be removed or if an alternative might be better | 17:25 |
| *** Ryan_Lane has joined #openstack-infra | 17:25 | |
| *** talluri has joined #openstack-infra | 17:25 | |
| devananda | clarkb: i dont feel that i have a good understanding of when vendor libraries for vendor drivers should be added to requirements or not | 17:25 |
| mordred | WOOT | 17:26 |
| mordred | finally got puppet running in the dib | 17:26 |
| clarkb | devananda: I think we need to reframe why requirements exists for you | 17:26 |
| devananda | clarkb: there are at least three for ironic that I expect to have to address during Juno: seamicro, ilo, and pyghmi | 17:26 |
| mordred | (foudn why it wasnt' working before- I now expect a 95% functional thing in maybe 2-3 ish more hours) | 17:26 |
| devananda | and others which are being discussed but for which there's no code yet | 17:26 |
| clarkb | devananda: it isn't related to gating (well it actually is but only for sanity not as a goal) | 17:26 |
| clarkb | it exists so that projects will use acommon set of tools | 17:27 |
| devananda | right | 17:27 |
| clarkb | on a common set of versions | 17:27 |
| devananda | that makes all kinds of sense for common requirements :) | 17:27 |
| clarkb | so that we stop having incompatible deps across openstack (which was a huge problem prior to the reqs repo) | 17:27 |
| devananda | it doesn't seem to make sense for external libs only needed by one project | 17:27 |
| clarkb | but also so that we can avoid having 4 libs that do the same thing | 17:27 |
| devananda | eg, ironic or cinder needing specific bindings for specific hardware | 17:27 |
| *** thedodd has quit IRC | 17:28 | |
| clarkb | devananda: where it does begin to make sense is when you consider the distros | 17:28 |
| clarkb | they can refer to requirements as the source of truth for all the things you might want to package for openstack | 17:28 |
| mordred | right. and for the most part, those are considered depllioy time reqs and aren't listed in requirements.txt | 17:28 |
| devananda | mordred: right. and at least for ironic, it's not the default | 17:28 |
| mordred | because a person installing ironic and using ipmi probably doesn't also need to install the seamicro lib | 17:28 |
| devananda | exactly | 17:28 |
| mordred | ++ | 17:28 |
| clarkb | thats fine | 17:28 |
| clarkb | that doesn't mean you exclude it from the requirements file though | 17:29 |
| clarkb | because you do that then ubuntu doesn't package pyghmi for you | 17:29 |
| clarkb | and now all the people deploying ironic have to figure that out | 17:29 |
| clarkb | *all the people using pyghmi | 17:29 |
| zaro | can anyone tell me why the name-filter: 'Performance' is only applied to rax VMs? | 17:29 |
| devananda | pyghmi may be a bit of a special case -- i'd like it to replace ipmitool | 17:29 |
| devananda | clarkb: but the same argument applies to python-seamicroclient | 17:29 |
| clarkb | zaro: because rax has performance flavors and regular flavors | 17:30 |
| openstackgerrit | Roger Luethi proposed a change to openstack-infra/git-review: Obey --version outside of git tree https://review.openstack.org/90160 | 17:30 |
| devananda | and hp3parclient | 17:30 |
| clarkb | zaro: but none of the other clouds do | 17:30 |
| clarkb | devananda: yup, I think they go in the requirements file | 17:30 |
| devananda | clarkb: but there are a host of other deploy time requirements tha tdeplpoyers need to figure out | 17:30 |
| zaro | so what's the difference between prformance vs regular? | 17:30 |
| devananda | clarkb: like tftpd | 17:30 |
| clarkb | I don't think you gate on seamicro though | 17:30 |
| clarkb | zaro: it depends on the particular flavor but in general you get more cpu cycles and less diskj | 17:30 |
| devananda | clarkb: then the entry in requirements is meaningless IMO because it's never tested | 17:31 |
| *** gyee has quit IRC | 17:31 | |
| clarkb | devananda: no go back to what I was saying before | 17:31 |
| clarkb | forget the gate completely | 17:31 |
| devananda | clarkb: i get the packaging impact | 17:31 |
| clarkb | this is a packaging thing | 17:31 |
| devananda | clarkb: but why tell packages to package a thing that we have no idea whether it works? | 17:31 |
| devananda | s/taht/when/ | 17:31 |
| clarkb | because if ironic hides their use of seamicro from everyone a year from now cinder will use a different seamicro lib | 17:31 |
| clarkb | devananda: that is where third party testing comes in | 17:32 |
| clarkb | but also ^ | 17:32 |
| mordred | s/seamicro/wsgi/ | 17:32 |
| *** mrodden has quit IRC | 17:32 | |
| devananda | mordred: wsgi we can test upstream | 17:32 |
| clarkb | you don't want cinder and ironic and neutron to all end up in a place with vendor libs that they were in with all the other libs | 17:32 |
| devananda | mordred: seamicro we can't | 17:32 |
| mordred | and you have the problem ina nutshell | 17:32 |
| openstackgerrit | A change was merged to openstack-infra/config: Gerrit: Ensure /opt/lib exists if used https://review.openstack.org/90147 | 17:32 |
| clarkb | and you know what happens in this case? oslo.vmware | 17:33 |
| mordred | anybody seen this before: | 17:33 |
| mordred | err: /Stage[main]/Openstack_project::Template/Service[rsyslog]: Could not evaluate: undefined method `[]' for nil:NilClass | 17:33 |
| *** Ryan_Lane1 has joined #openstack-infra | 17:33 | |
| devananda | clarkb: fair 'nuf. /me updates requirements to indicate that it's there for packaging and not for testing | 17:34 |
| mordred | clarkb, jeblair ^^ I just saw that go by in running puppet for the slave dib - I wonder if it's happening all the time but we don't notice? | 17:34 |
| devananda | clarkb: and that ironic doesn't actually import it :) | 17:34 |
| clarkb | I really don't want anymore oslo.$vendor libs | 17:34 |
| clarkb | devananda: right note that it is an optional dependency for ironic to talk to seamicro nodes | 17:34 |
| clarkb | and that seamicro nodes work fine with pyghmi or whatever | 17:34 |
| clarkb | however ^ raises a more interesting qusetion to me | 17:35 |
| clarkb | should we completely avoid vendor specific libs in the first place and stick to standard tooling? | 17:35 |
| jeblair | mordred: you could check the nodepool image build logs | 17:35 |
| mordred | jeblair: k | 17:36 |
| clarkb | mordred: I don't think so, if puppet errors it should fail the image build | 17:36 |
| mordred | clarkb: that err did not cause puppet to return error | 17:36 |
| clarkb | eg the metadata things we see in hpcloud | 17:36 |
| clarkb | mordred: oh | 17:36 |
| *** mrodden has joined #openstack-infra | 17:36 | |
| mordred | beacuse, you know, sometimes puppet does that for $reasons | 17:36 |
| openstackgerrit | A change was merged to openstack-infra/config: temporary disable gerrit tests https://review.openstack.org/90154 | 17:36 |
| *** thedodd has joined #openstack-infra | 17:37 | |
| devananda | clarkb: "should we completely avoid vendor specific libs" -- unless you're asking if opensatck should refrain from ever utilizing vendor-specific APIs, I'm not sure what you mean | 17:37 |
| mordred | clarkb: I think the no-vendor-libs approach would not ultimately be helpful in the goal of building an open cloud | 17:37 |
| mordred | clarkb: if you think of openstack as a datacenter os | 17:37 |
| *** zehicle_at_dell has quit IRC | 17:37 | |
| *** Ryan_Lane1 has quit IRC | 17:38 | |
| clarkb | mordred: maybe I should rephrase | 17:38 |
| clarkb | mordred: don't lock into specific libs when there are alternatives that work for all the things | 17:38 |
| mordred | then sometimes you have hardware that you need to interact with via some drivers - but YOU want to just use openstack to manipulate it | 17:38 |
| mordred | clarkb: well sure, I think we do tend to prefer alternatives that work for all of the things | 17:38 |
| clarkb | if python seamicro does something that ironic can't do with ipmi it poses two problems. First the testing one. Second now you have no feature parity | 17:38 |
| mordred | but hardware vendors kinda suck a lot - so the existence of such things is less common | 17:39 |
| *** dkliban_afk is now known as dkliban | 17:39 | |
| mordred | clarkb: what if python-seamicro lets you control hardware devices that don't have an IPMI interface | 17:39 |
| mordred | but your ironic interace to that hardware is now the same? | 17:39 |
| *** UtahDave has quit IRC | 17:39 | |
| mordred | in essence, WE are becoming the thing that is the common thing that works with all of the things | 17:40 |
| clarkb | mordred: while I would like that to be the case history has shown we are completely incapable of doing that | 17:40 |
| mordred | clarkb: well, we can keep trying :) | 17:40 |
| clarkb | as soon as you use $other thing feature parity is thrown out the window | 17:40 |
| devananda | clarkb: there are not alternatives that work for all things | 17:40 |
| clarkb | I have a summit session to hopeflly point a spotlight on this problem and get peopel to consider it seriously | 17:40 |
| *** gokrokve has joined #openstack-infra | 17:40 | |
| devananda | clarkb: specifically, anything beyond IPMI 2.0 is not standardized _at_all_ -- but is really helpful and nearly everyone implements it | 17:40 |
| devananda | clarkb: eg, mount virtual media | 17:41 |
| *** mrodden1 has joined #openstack-infra | 17:41 | |
| clarkb | but I am skeptical that adding more non feature consist bits is a good thing | 17:41 |
| mordred | ^^ ++ | 17:41 |
| devananda | clarkb: every hardware mgmt interface has it but it's not part of IPMI | 17:41 |
| *** mrodden has quit IRC | 17:41 | |
| devananda | clarkb: and it's a planned component of secure boot, if ironic ever gets around to that | 17:41 |
| clarkb | devananda: right so I am not speaknig as an expert | 17:41 |
| clarkb | merely poitning out the myriad of issues we have in this space | 17:41 |
| clarkb | bceause no one does the thing mordred wants them to do | 17:41 |
| devananda | clarkb: heh. well, i'm trying | 17:41 |
| mordred | yup. I will stick up for devananda here, because I know he's trying to have ironic do the right thing | 17:42 |
| devananda | my goal with ironic is to drive the vendors towards feature parity for things which the common libraries DONT implement today | 17:42 |
| mordred | ++ | 17:42 |
| devananda | so seamicro, ilo, etc -- they implement some of the same functions but their HW APIs are different | 17:43 |
| clarkb | ya definitely not singling out ironic here | 17:43 |
| *** _nadya_ has joined #openstack-infra | 17:43 | |
| clarkb | or seamicro or 3par | 17:43 |
| devananda | because HW vendors dont like to agree | 17:43 |
| devananda | but all their customers want the same things | 17:43 |
| *** maxbit has quit IRC | 17:44 | |
| clarkb | and you'll note that I said add it to requirements :) | 17:44 |
| devananda | yep | 17:44 |
| jeblair | clarkb, mordred: does manage-projects run as root or gerrit2? | 17:44 |
| clarkb | I just find this massive feature matrix that we build for ourselves to be more interesting on a whole than should $individual lib be included | 17:45 |
| clarkb | jeblair: its run by puppet so root | 17:45 |
| mordred | jeblair: should runas gerrit2 | 17:45 |
| *** krtaylor has joined #openstack-infra | 17:45 | |
| clarkb | the puppet exec could change uers but puppet makes that difficultish | 17:45 |
| mordred | oh. | 17:45 |
| mordred | hrm | 17:45 |
| devananda | clarkb: im starting to feel your pain. there are some feature matrixes in ironic that are cropping up with the new agent | 17:45 |
| clarkb | anything relying on user env stuff will need to be very explicit | 17:45 |
| clarkb | eg where ssh keys are located and so on | 17:46 |
| jeblair | zaro: have you run manage-projects on review-dev recently? | 17:46 |
| jeblair | paramiko.BadHostKeyException: Host key for server localhost does not match! | 17:46 |
| jeblair | i get that error | 17:46 |
| clarkb | definitely doable though and ++ to having gerrit2 do it | 17:46 |
| jeblair | mordred: ^ ? | 17:46 |
| mordred | jeblair: zaro was having that problem on review-dev | 17:46 |
| jeblair | it was not a philosophical question. :) | 17:46 |
| mordred | I know | 17:46 |
| mordred | oh. jeez. | 17:47 |
| mordred | 17:46:22 jeblair | zaro: have you run manage-projects on review-dev recently? | 17:47 |
| mordred | that's how much I suck | 17:47 |
| *** lakshmi has joined #openstack-infra | 17:47 | |
| zaro | jeblair, mordred : i tried that yesterday but got an mismatched key error | 17:47 |
| *** zns has quit IRC | 17:48 | |
| jeblair | zaro: have you tested it against 2.8? | 17:48 |
| zaro | was going to try and update the project owner key and redo | 17:48 |
| anteaya | I am starting to become pleased with the information for developers regarding the Gerrit 2.8 upgrade found here: https://etherpad.openstack.org/p/ai9irqk3FC | 17:48 |
| zaro | jeblair: i ran on review-dev.o.o | 17:48 |
| anteaya | looking for feedback | 17:48 |
| zaro | jeblair: which is 2.8 | 17:48 |
| anteaya | it is clear? | 17:48 |
| jeblair | zaro: yes, but did you run it and did it _work_? | 17:49 |
| jeblair | i'm wondering if someone did update the project owner key but didn't do it completely or correctly | 17:49 |
| jeblair | (for heartbleed) | 17:49 |
| *** mburned is now known as mburned_out | 17:49 | |
| zaro | jeblair: no, failed with mismatched keys. clarkb said i needed to update gerrit project creator key | 17:49 |
| clarkb | I thought it was because the key in gerrit's DB wasn't updated | 17:49 |
| jeblair | clarkb: yes, that makes sense | 17:50 |
| clarkb | but the key on disk was. But I didn't actually check. that was just a hunch | 17:50 |
| jeblair | i will fix it | 17:50 |
| *** nati_ueno has joined #openstack-infra | 17:52 | |
| *** mark0n has quit IRC | 17:52 | |
| *** mburned_out is now known as mburned | 17:53 | |
| jeblair | i think it's probably actually ~/.ssh/known_hosts | 17:53 |
| jeblair | we will need to do that on review on monday | 17:54 |
| jeblair | ssh-keygen -f "/root/.ssh/known_hosts" -R [localhost]:29418 | 17:56 |
| jeblair | fixed that | 17:56 |
| jeblair | BadCredentialsException: 401 {u'documentation_url': u'https://developer.github.com/v3', u'message': u'Bad credentials'} | 17:56 |
| jeblair | nowe we have that | 17:56 |
| clarkb | oh mordred swapped out github creds and must not have updated review-dev? | 17:57 |
| clarkb | review-dev has its own token under the same account iirc | 17:57 |
| *** zns has joined #openstack-infra | 17:57 | |
| clarkb | so it is segregated but will need to be updated independent of review.o.o which is working best I can tell | 17:57 |
| clarkb | oh wait no | 17:57 |
| clarkb | mange-projects doesn't use the tokens because tehy couldn't create projects. They can now, we should fix manage projects | 17:57 |
| clarkb | mordred: ^ do you know what is going on there? | 17:58 |
| * clarkb needs to AFK to transplant from home to elsewhere | 17:58 | |
| *** thedodd has quit IRC | 17:59 | |
| openstackgerrit | Michael Krotscheck proposed a change to openstack-infra/storyboard-webclient: Favicon Whitelabel https://review.openstack.org/90167 | 17:59 |
| jeblair | erm, the password for that account in the password file is wrong | 18:00 |
| *** thedodd has joined #openstack-infra | 18:01 | |
| jeblair | mordred: ^ | 18:01 |
| *** sweston has joined #openstack-infra | 18:01 | |
| mordred | jeblair: I remember updating that :( | 18:02 |
| *** andreaf has quit IRC | 18:02 | |
| openstackgerrit | Michael Krotscheck proposed a change to openstack-infra/storyboard-webclient: Removed logo images https://review.openstack.org/90169 | 18:02 |
| *** jcoufal has joined #openstack-infra | 18:03 | |
| *** e0ne has joined #openstack-infra | 18:03 | |
| jeblair | mordred: the pw in the file matches hiera which you did update | 18:04 |
| jeblair | mordred: it's possible github still has the old password; will try that | 18:04 |
| *** talluri has quit IRC | 18:04 | |
| jeblair | mordred: yep, that was it. you made a new pw but didn't actually change it on github | 18:05 |
| *** talluri has joined #openstack-infra | 18:05 | |
| jeblair | fixing that now | 18:05 |
| mordred | jeblair: wow. sorry about that | 18:05 |
| jeblair | yay git :) | 18:05 |
| *** sweston has quit IRC | 18:05 | |
| jeblair | done | 18:06 |
| jeblair | and now manage-projects on review-dev runs | 18:06 |
| *** nati_ueno has quit IRC | 18:06 | |
| anteaya | yay | 18:06 |
| mordred | w00t! | 18:07 |
| jeblair | (i don't know if it can create projects or change acls or anything though) | 18:07 |
| *** ihrachyshka has quit IRC | 18:07 | |
| *** talluri has quit IRC | 18:07 | |
| anteaya | but it runs | 18:08 |
| *** MarkAtwood has quit IRC | 18:08 | |
| *** talluri has joined #openstack-infra | 18:08 | |
| *** amcrn has joined #openstack-infra | 18:09 | |
| zaro | jeblair: thanks for fixing. did it pass the entire puppet run? will you start the agent on review-dev? | 18:10 |
| *** _nadya_ has quit IRC | 18:11 | |
| devananda | clarkb: if you can spare a moment, I'd like your input as to whether https://review.openstack.org/90174 is a sane way to do what we just talked about | 18:12 |
| jeblair | zaro: done | 18:12 |
| *** praneshp has joined #openstack-infra | 18:12 | |
| *** che-arne has quit IRC | 18:13 | |
| *** nati_ueno has joined #openstack-infra | 18:13 | |
| jeblair | i'm approving the replica move to /opt on production; i just rsynced the existing contents over | 18:13 |
| jeblair | when it lands, i will restart gerrit and then trigger replication to catch it up | 18:14 |
| *** msabramo1 has quit IRC | 18:14 | |
| *** nati_ueno has quit IRC | 18:14 | |
| *** zns has quit IRC | 18:14 | |
| sdague | Alex_Gaynor: so what's the story with the pypy issue? | 18:15 |
| Alex_Gaynor | sdague: there's a PR for pip, dstufft could opine on when it's likely to land / be in a release | 18:15 |
| sdague | that seems to have been abandoned to the point where it feels like it's time to just delete the jobs, as they have been running non voting for a long time | 18:16 |
| sdague | Alex_Gaynor: that's not going to help us though, right? because we're pinned on an earlier pip. | 18:16 |
| dstufft | Likely it'll land when Marcus feels it's ready and merges it, the PR is only a few days old. | 18:16 |
| openstackgerrit | A change was merged to openstack-infra/config: Move local git replica on review to /opt https://review.openstack.org/90129 | 18:17 |
| sdague | dstufft: but we're using pip 1.4.1, so that means that doesn't solve it for infra, right? | 18:17 |
| openstackgerrit | Devananda van der Veen proposed a change to openstack/requirements: Add notes about packaging of external libraries https://review.openstack.org/90176 | 18:18 |
| dstufft | Not generally, but probably I can be convinced to fiddle some release bits if someone makes a PR against 1.4.X once Marcus's PR lands | 18:18 |
| *** jcoufal has quit IRC | 18:19 | |
| Alex_Gaynor | sdague: I haven't heard anything from the debian maintainer who did the patching, unfortunately :/ | 18:19 |
| sdague | Alex_Gaynor: well, it doesn't seem like it's doing us any good to keep running these jobs non voting that don't work | 18:20 |
| jlk | clarkb: in j-j-b, I've got a job-group, where i want to have a list in one attribute, which would create a job for every item in that list, using a specific job template. Is there a way to have multiple attributes as lists in a job-group? to create some sort of union or cartesian product? | 18:20 |
| *** e0ne has quit IRC | 18:21 | |
| *** darraghb has joined #openstack-infra | 18:22 | |
| *** e0ne has joined #openstack-infra | 18:22 | |
| *** nati_ueno has joined #openstack-infra | 18:22 | |
| jlk | clarkb: ah, I do it in - project | 18:24 |
| dstufft | sdague: you could install PyPy from tarball or turn off wheels for PyPy too | 18:24 |
| *** darraghb has quit IRC | 18:25 | |
| *** rcarrill` has joined #openstack-infra | 18:25 | |
| Alex_Gaynor | dstufft: --no-use-wheel would fix it I suppose? | 18:26 |
| Alex_Gaynor | or, PIP_NO_USE_WHEEL=1 ? | 18:26 |
| dstufft | yea | 18:27 |
| dstufft | well | 18:27 |
| dstufft | it's pip 1.4 | 18:27 |
| dstufft | oh wait | 18:27 |
| *** msabramo has joined #openstack-infra | 18:27 | |
| dstufft | did 1.4 have wheels on by default I forget | 18:27 |
| *** rcarrillocruz has quit IRC | 18:27 | |
| Alex_Gaynor | are we specifically enabling wheels somewhere? | 18:27 |
| dstufft | I think that was 1.5 | 18:27 |
| Alex_Gaynor | no, it had them opt-in | 18:27 |
| dstufft | so wheels must be getting explicitly enabled somewhere | 18:28 |
| zaro | anteaya: gerrit upgrade doc looks good, gave a few suggestions but not much. thanks | 18:28 |
| dstufft | at least as far as we know the bug only happens in the wheel install path | 18:28 |
| anteaya | zaro: thanks for the suggestions | 18:28 |
| *** blamar_ has joined #openstack-infra | 18:29 | |
| openstackgerrit | Michael Krotscheck proposed a change to openstack-infra/storyboard-webclient: Fixed padding on refresh icon. https://review.openstack.org/90182 | 18:29 |
| *** blamar_ is now known as blamar | 18:29 | |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/config: Gerrit: fix cron syntax errors https://review.openstack.org/90184 | 18:30 |
| jeblair | mordred, clarkb: ^ tested brown bag fix | 18:30 |
| anteaya | zaro: do we have a link to the gerrit search docs for 2.8.4 that you would like to use? | 18:30 |
| *** darraghb has joined #openstack-infra | 18:30 | |
| jeblair | mordred, clarkb: that produces: 7 4 * * 0 find /opt/lib/git -type d -name "*.git" -print -exec git --git-dir="{}" repack -afd \; | 18:31 |
| *** derekh has joined #openstack-infra | 18:31 | |
| mordred | jeblair: +2 | 18:31 |
| zaro | anteaya: it's on the site documentation->searching | 18:32 |
| jeblair | i'm restarting gerrit now to pick up the replication config change | 18:32 |
| *** praneshp_ has joined #openstack-infra | 18:33 | |
| *** markvoelker has joined #openstack-infra | 18:33 | |
| anteaya | zaro: _that's_ what that documention button doea | 18:34 |
| anteaya | does | 18:34 |
| anteaya | I had no idea, thank you | 18:34 |
| *** darraghb has quit IRC | 18:34 | |
| *** markwash has joined #openstack-infra | 18:35 | |
| *** igor__ has joined #openstack-infra | 18:35 | |
| *** UtahDave has joined #openstack-infra | 18:35 | |
| *** praneshp has quit IRC | 18:35 | |
| *** praneshp_ is now known as praneshp | 18:35 | |
| jeblair | i triggered a global replication push | 18:36 |
| *** zns has joined #openstack-infra | 18:37 | |
| *** thedodd has quit IRC | 18:37 | |
| *** wenlock has quit IRC | 18:38 | |
| jeblair | cloning from /p/... seems to work | 18:39 |
| *** _nadya_ has joined #openstack-infra | 18:39 | |
| openstackgerrit | Marc Abramowitz proposed a change to stackforge/python-jenkins: Add .gitignore https://review.openstack.org/90188 | 18:39 |
| *** thedodd has joined #openstack-infra | 18:40 | |
| jeblair | i think i'm finally ready to make a cinder volume. :) | 18:40 |
| devoid | is there a guide for donating VMs to the testing pool? | 18:40 |
| *** jooools has quit IRC | 18:41 | |
| openstackgerrit | Michael Krotscheck proposed a change to openstack-infra/storyboard-webclient: Added loading UI for current user name https://review.openstack.org/90189 | 18:41 |
| jeblair | devoid: nothing published, but send email to <jeblair@openstack.org> and i can work with you on it | 18:41 |
| *** wenlock has joined #openstack-infra | 18:42 | |
| devoid | jeblair: will do | 18:43 |
| jeblair | i'll make a 40G ssd volume for /home/gerrit2 (it's currently 5.5G) | 18:45 |
| jeblair | after lunch | 18:45 |
| *** doddstack has joined #openstack-infra | 18:46 | |
| *** _nadya_ has quit IRC | 18:46 | |
| *** hashar has joined #openstack-infra | 18:49 | |
| *** thedodd has quit IRC | 18:49 | |
| *** nati_ueno has quit IRC | 18:49 | |
| clarkb | sounds good | 18:51 |
| sdague | dstufft / Alex_Gaynor: so my concern is that I turned that job completely non voting on March 18th | 18:51 |
| sdague | 6 weeks later, no resolution. | 18:51 |
| sdague | and no one seems to be owning solving it | 18:51 |
| *** chandan_kumar has quit IRC | 18:52 | |
| *** ominakov has joined #openstack-infra | 18:53 | |
| *** lukego has quit IRC | 18:53 | |
| *** zns has quit IRC | 18:54 | |
| *** zns has joined #openstack-infra | 18:55 | |
| *** zehicle_at_dell has joined #openstack-infra | 18:55 | |
| *** talluri has quit IRC | 18:57 | |
| openstackgerrit | Monty Taylor proposed a change to openstack-infra/config: Add support for disk-image-builder in nodepool https://review.openstack.org/88479 | 18:58 |
| * mordred afks | 18:58 | |
| *** jaypipes has joined #openstack-infra | 18:58 | |
| *** ildikov has joined #openstack-infra | 18:59 | |
| mtreinish | ttx: if you're still around, is there anything I have to do to give the qa track a topic and color on sched.org? | 19:00 |
| *** BadCub has joined #openstack-infra | 19:02 | |
| zaro | anteaya: did you see the 1st abandoned change on the list on review-dev? | 19:02 |
| anteaya | ahhhhh | 19:02 |
| zaro | anteaya: no X in W column. | 19:02 |
| sdague | mtreinish: you got white because you happen to be in that part of the color rotation | 19:02 |
| sdague | it will change the next time something alphabetically < q pushes a schedule | 19:03 |
| anteaya | zaro: was looking at this: https://review-dev.openstack.org/#/q/status:abandoned,n,z | 19:03 |
| zaro | anteaya: yeah, that's what i'm referring to | 19:03 |
| anteaya | zaro: then was in conversation, haven't figured out the pattern around how abandoned patches show up | 19:03 |
| mtreinish | sdague: oh ok, but there isn't a topic listed either which is why I thought I might be missing something | 19:03 |
| mtreinish | on the key in the upper right of the page | 19:04 |
| sdague | oh, there might be an issue | 19:04 |
| zaro | anteaya: abandone is unrelated to wip. you have have abadoned changes in their work in progress or ready for review state | 19:04 |
| zaro | their/either | 19:04 |
| anteaya | hmmmm | 19:05 |
| anteaya | maybe I will just leave that sentence out then | 19:05 |
| *** lukego has joined #openstack-infra | 19:06 | |
| *** liam__ has joined #openstack-infra | 19:06 | |
| *** liam__ has quit IRC | 19:06 | |
| zaro | anteaya: abandoned and wip are states. being in one doesn't necessarily mean you are in another. | 19:06 |
| anteaya | zaro: I removed that sentence | 19:06 |
| anteaya | the big direction was clarifying work in progress which I think I have done | 19:06 |
| *** Ajaeger1 has quit IRC | 19:07 | |
| *** hartsocks has joined #openstack-infra | 19:07 | |
| anteaya | zaro: how are states different from workflow? | 19:07 |
| *** markmcclain has joined #openstack-infra | 19:07 | |
| anteaya | or should I have that conversation after the upgrade? | 19:07 |
| *** exz has quit IRC | 19:07 | |
| *** StevenK has quit IRC | 19:07 | |
| *** StevenK has joined #openstack-infra | 19:07 | |
| *** simonmcc_ has quit IRC | 19:08 | |
| *** simonmcc_ has joined #openstack-infra | 19:08 | |
| zaro | anteaya: i think that's semantics. workflow contains states and items can be in multiple states at the same time. like a change can be in wip and abandoned state at the same time. | 19:09 |
| *** Adri2000 has quit IRC | 19:10 | |
| *** Adri2000 has joined #openstack-infra | 19:10 | |
| *** serverascode has quit IRC | 19:11 | |
| *** cyoyo has quit IRC | 19:11 | |
| *** dims has quit IRC | 19:11 | |
| anteaya | zaro: let's have the semantics conversation over dinner | 19:11 |
| anteaya | if the etherpad is good for now, that is my main concern | 19:12 |
| *** blamar has quit IRC | 19:12 | |
| openstackgerrit | Salvatore Orlando proposed a change to openstack-infra/config: Make neutron full job voting (neutron gate only) https://review.openstack.org/88289 | 19:12 |
| zaro | anteaya: looking forward to it. etherpad wfm | 19:12 |
| *** exz has joined #openstack-infra | 19:13 | |
| *** cyeoh has joined #openstack-infra | 19:13 | |
| *** serverascode has joined #openstack-infra | 19:13 | |
| anteaya | zaro: great thanks | 19:13 |
| *** chandan_kumar has joined #openstack-infra | 19:15 | |
| *** prad has quit IRC | 19:16 | |
| *** prad has joined #openstack-infra | 19:17 | |
| *** MarkAtwood has joined #openstack-infra | 19:17 | |
| *** _nadya_ has joined #openstack-infra | 19:17 | |
| *** marcoemorais has quit IRC | 19:17 | |
| jogo | mordred: wanna do some hacking reviews so I cannot do any more giant rebases: https://review.openstack.org/#/q/status:open+project:openstack-dev/hacking+branch:master+topic:rev,n,z | 19:22 |
| *** chandan_kumar has quit IRC | 19:23 | |
| *** david-lyle_ has joined #openstack-infra | 19:26 | |
| openstackgerrit | Matt Riedemann proposed a change to openstack-infra/elastic-recheck: Re-add query for bug 1251784 https://review.openstack.org/90200 | 19:26 |
| uvirtbot | Launchpad bug 1251784 in tripleo "nova+neutron scheduling error: Connection to neutron failed: Maximum attempts reached" [Critical,Fix released] https://launchpad.net/bugs/1251784 | 19:26 |
| *** ihrachyshka has joined #openstack-infra | 19:27 | |
| mtreinish | sdague: well it looks like it fixed itself now | 19:28 |
| sdague | oh, maybe on delay | 19:28 |
| *** saschpe has quit IRC | 19:29 | |
| *** mdenny has quit IRC | 19:29 | |
| *** wenlock has quit IRC | 19:29 | |
| *** mdenny has joined #openstack-infra | 19:30 | |
| *** KurtMartin has joined #openstack-infra | 19:30 | |
| *** wenlock has joined #openstack-infra | 19:30 | |
| *** nati_ueno has joined #openstack-infra | 19:30 | |
| *** kmartin has quit IRC | 19:33 | |
| *** vhoward has joined #openstack-infra | 19:33 | |
| *** dims has joined #openstack-infra | 19:33 | |
| anteaya | cyeoh: yayyyayayayayay | 19:34 |
| *** marcoemorais has joined #openstack-infra | 19:36 | |
| *** saschpe has joined #openstack-infra | 19:38 | |
| *** vhoward has left #openstack-infra | 19:39 | |
| *** sdague has quit IRC | 19:41 | |
| *** arosen has joined #openstack-infra | 19:41 | |
| *** sdague has joined #openstack-infra | 19:42 | |
| *** coolsvap|afk has quit IRC | 19:44 | |
| openstackgerrit | A change was merged to openstack-infra/config: Gerrit: fix cron syntax errors https://review.openstack.org/90184 | 19:44 |
| openstackgerrit | Matt Riedemann proposed a change to openstack-infra/elastic-recheck: Add check-grenade-dsvm-neutron-icehouse to bug 1307344 query https://review.openstack.org/90207 | 19:45 |
| uvirtbot | Launchpad bug 1307344 in neutron "neutron migrations are not idempotent or invariant to enabled services" [Critical,In progress] https://launchpad.net/bugs/1307344 | 19:45 |
| clarkb | I restarted firefox and lost all my tabs :( | 19:46 |
| clarkb | may as well update and reboot now | 19:46 |
| *** ihrachyshka has quit IRC | 19:47 | |
| *** hashar has quit IRC | 19:47 | |
| *** ihrachyshka has joined #openstack-infra | 19:47 | |
| *** MarkAtwood has quit IRC | 19:49 | |
| clarkb | devananda: yup something like that is what I would do | 19:51 |
| cody-somerville | clarkb: I hate when that happens to me :( | 19:51 |
| *** dprince has quit IRC | 19:51 | |
| clarkb | devananda: its a similar problem to what neutron faces for linux only things | 19:51 |
| clarkb | devananda: they need to be mocked if people want to run unittests on windows or OS X | 19:52 |
| clarkb | cody-somerville: I had to double check I have the save my tabs option selected (and I do) | 19:54 |
| clarkb | not sure what happened but I restarted it because it was unhappy so might be related | 19:54 |
| *** _nadya_ has quit IRC | 19:54 | |
| sdague | clarkb: using the built in tab restore, or tabmixplus? | 19:56 |
| sdague | I found I could never trust built in tab saving in firefox | 19:56 |
| *** david-lyle_ has quit IRC | 19:56 | |
| clarkb | sdague: built in tab restore | 19:57 |
| *** julim has quit IRC | 19:58 | |
| *** ociuhandu has quit IRC | 19:58 | |
| *** melwitt has quit IRC | 20:00 | |
| openstackgerrit | A change was merged to openstack-dev/hacking: Make test_doctest use entry points https://review.openstack.org/86692 | 20:02 |
| *** melwitt has joined #openstack-infra | 20:02 | |
| soren | clarkb: That only works when Firefox crashes, doesn't it? If you restart it gracefully, the tabs are gone IME. | 20:02 |
| openstackgerrit | A change was merged to openstack-dev/hacking: Move comment checks into their own module https://review.openstack.org/86693 | 20:03 |
| openstackgerrit | A change was merged to openstack-dev/hacking: Move except checks into their own module https://review.openstack.org/86694 | 20:03 |
| boris-42 | clarkb hi | 20:03 |
| *** mburned is now known as mburned_out | 20:04 | |
| *** zehicle_at_dell has quit IRC | 20:05 | |
| openstackgerrit | A change was merged to openstack-dev/hacking: Only compile regex for hacking_except_format once https://review.openstack.org/86695 | 20:05 |
| openstackgerrit | Michael Krotscheck proposed a change to openstack-infra/storyboard: Users can be filtered with LIKE clauses https://review.openstack.org/90217 | 20:06 |
| *** harlowja is now known as harlowja_away | 20:06 | |
| *** marcoemorais has quit IRC | 20:06 | |
| clarkb | soren: there is an option under preferences to save tabs always | 20:07 |
| *** mburned_out is now known as mburned | 20:07 | |
| clarkb | boris-42: hi | 20:07 |
| openstackgerrit | Khai Do proposed a change to openstack-infra/config: use buck from master branch https://review.openstack.org/90219 | 20:08 |
| openstackgerrit | A change was merged to openstack-infra/elastic-recheck: Re-add query for bug 1251784 https://review.openstack.org/90200 | 20:08 |
| uvirtbot | Launchpad bug 1251784 in tripleo "nova+neutron scheduling error: Connection to neutron failed: Maximum attempts reached" [Critical,Fix released] https://launchpad.net/bugs/1251784 | 20:08 |
| openstackgerrit | A change was merged to openstack-dev/hacking: Speed up license checks https://review.openstack.org/86726 | 20:08 |
| soren | clarkb: Does that somehow protect you when you have another window open (some popup or whatever) and then you close your main window with all the tabs? | 20:09 |
| soren | clarkb: I've done that more times than I care to think about. | 20:09 |
| *** marcoemorais has joined #openstack-infra | 20:09 | |
| clarkb | soren: I'm not sure. I am a one window user | 20:09 |
| *** HenryG has quit IRC | 20:11 | |
| soren | clarkb: I find that sometimes various web sites "helpfully" open a help window or something in another window (rather than in another tab).. Then I leave it alone, forget all about it, and close my main browser window and find it sitting there, taunting me. | 20:11 |
| openstackgerrit | A change was merged to openstack-infra/elastic-recheck: Cirros boot failed https://review.openstack.org/90100 | 20:12 |
| openstackgerrit | A change was merged to openstack-dev/hacking: Move python 2->3 checks into their own module https://review.openstack.org/88440 | 20:13 |
| openstackgerrit | A change was merged to openstack-dev/hacking: Compile python 2-> regular expressions https://review.openstack.org/88441 | 20:14 |
| *** mrmartin has quit IRC | 20:15 | |
| *** sweston has joined #openstack-infra | 20:15 | |
| devananda | clarkb: ack, thanks | 20:18 |
| openstackgerrit | Matthew Treinish proposed a change to openstack-infra/config: Move tempest master-havana job to experimental https://review.openstack.org/90225 | 20:19 |
| openstackgerrit | Sean Dague proposed a change to openstack-infra/config: move tempest-master-*-havana to experimental https://review.openstack.org/90226 | 20:19 |
| mtreinish | sdague: ^^^ I win this time... | 20:19 |
| mtreinish | :) | 20:19 |
| * zaro takes a walk. bbib | 20:19 | |
| sdague | heh | 20:19 |
| sdague | you should say something :) | 20:19 |
| sdague | mtreinish: your patch isn't right though :) | 20:19 |
| *** _sweston_ has joined #openstack-infra | 20:20 | |
| *** sweston has quit IRC | 20:20 | |
| mtreinish | oh, I didn't realize we ran it on devstack too | 20:20 |
| openstackgerrit | A change was merged to openstack-dev/hacking: Move import checks into their own file https://review.openstack.org/89454 | 20:20 |
| openstackgerrit | A change was merged to openstack-dev/hacking: Move docstring checks into their own file https://review.openstack.org/89713 | 20:20 |
| mtreinish | sdague: I'll just abandon mine, just so I can contribute to your stat padding for samsung | 20:21 |
| *** ociuhandu has joined #openstack-infra | 20:23 | |
| boris-42 | clarkb could you pls take a look at that patch with publishing plots | 20:23 |
| boris-42 | clarkb we really would like to finish this part of job | 20:23 |
| boris-42 | clarkb pls pls pls =) | 20:23 |
| boris-42 | clarkb it's quite simple patch -> https://review.openstack.org/#/c/89562/ | 20:23 |
| *** SumitNaiksatam has quit IRC | 20:24 | |
| *** Ryan_Lane1 has joined #openstack-infra | 20:27 | |
| *** bknudson has quit IRC | 20:28 | |
| *** Ryan_Lane1 has quit IRC | 20:31 | |
| *** weshay has quit IRC | 20:34 | |
| *** changbl has quit IRC | 20:36 | |
| *** bknudson has joined #openstack-infra | 20:37 | |
| *** derekh has quit IRC | 20:39 | |
| *** marun has quit IRC | 20:40 | |
| *** changbl has joined #openstack-infra | 20:41 | |
| openstackgerrit | Clark Boylan proposed a change to openstack-infra/zuul: Handle NullChange "merges" gracefully. https://review.openstack.org/85217 | 20:42 |
| openstackgerrit | Clark Boylan proposed a change to openstack-infra/zuul: Add a remote url override location https://review.openstack.org/76057 | 20:42 |
| clarkb | sdague: did we end up not needing https://review.openstack.org/#/c/84828/1 for branchless tempest? | 20:45 |
| clarkb | I am going to create longgeeks service acocunt now Bug 1306967 | 20:47 |
| uvirtbot | Launchpad bug 1306967 in openstack-ci "thstack-ci service account" [Undecided,New] https://launchpad.net/bugs/1306967 | 20:47 |
| *** zehicle_at_dell has joined #openstack-infra | 20:47 | |
| anteaya | clarkb: yay, he will be very happy | 20:47 |
| *** changbl has quit IRC | 20:50 | |
| *** che-arne has joined #openstack-infra | 20:50 | |
| _sweston_ | anteaya: ok, finally out of meetings for now. I looked over the etherpad, can you tell me what else you need? | 20:50 |
| *** jaypipes has quit IRC | 20:52 | |
| anteaya | _sweston_: I think we are good | 20:52 |
| anteaya | thanks so much for putting together the seed docs | 20:52 |
| anteaya | I'm really grateful | 20:52 |
| *** ominakov has quit IRC | 20:52 | |
| anteaya | what is your availablity for testing the gerrit upgrade instructions? | 20:53 |
| *** prad has quit IRC | 20:53 | |
| _sweston_ | anteaya: I am grateful for your help as well, thank you :-) | 20:53 |
| anteaya | anytime | 20:53 |
| anteaya | I enjoy working with you | 20:53 |
| clarkb | its done | 20:54 |
| clarkb | anteaya: ^ | 20:54 |
| *** changbl has joined #openstack-infra | 20:54 | |
| *** SumitNaiksatam has joined #openstack-infra | 20:54 | |
| _sweston_ | :-) what do we need to test the upgrade? do we want to start with a base 2.4.8 ? | 20:54 |
| clarkb | _sweston_: I think jeblairs plan is to migrate to 2.4.8 not to upgrade directly to it | 20:55 |
| *** _nadya_ has joined #openstack-infra | 20:55 | |
| clarkb | _sweston_: there is a new machine jeblair is working on to do that | 20:55 |
| clarkb | which should allow us to check it before we go live and minimize the actual downtime | 20:55 |
| _sweston_ | clarkb: what are we migrating from? | 20:55 |
| clarkb | 2.4.4 | 20:56 |
| clarkb | er sorry I totally flipped those numbers before | 20:56 |
| jeblair | clarkb: 2.4.8 != 2.8.4 | 20:56 |
| clarkb | ya | 20:56 |
| clarkb | my eyes were convinced they were the same though | 20:56 |
| anteaya | clarkb: yay | 20:57 |
| _sweston_ | hehe, ok so help me understand what I need on my base box then | 20:58 |
| anteaya | jeblair: is there anything _sweston_ and I can work on to help at this point? | 20:58 |
| _sweston_ | i.e. where are we starting and where are we going? | 20:58 |
| clarkb | _sweston_: 2.4.4 is what we currently have. you can see the version on the bottom right of https://review.openstack.org/ | 20:58 |
| clarkb | _sweston_: that way is available at http://tarballs.openstack.org/ci/ | 20:59 |
| clarkb | s/way/war/ | 20:59 |
| anteaya | _sweston_: https://etherpad.openstack.org/p/gerrit_upgrade_script and https://etherpad.openstack.org/p/remaining-gerrit-upgrade-changes | 20:59 |
| *** mwagner_lap has quit IRC | 20:59 | |
| _sweston_ | clarkb: anteaya: ok, so I can bring up a box with 2.4.4 and then test the upgrade path from there? | 21:00 |
| clarkb | _sweston_: ya, but we should check with jeblair if that is useful | 21:00 |
| * anteaya nods | 21:01 | |
| clarkb | _sweston_: I think the current plan is not an actual in place upgrade which makes migration and roll back really easy | 21:01 |
| *** harlowja_away is now known as harlowja | 21:01 | |
| *** Ryan_Lane has quit IRC | 21:01 | |
| anteaya | we just point revew.o.o at our server of choice | 21:02 |
| *** pdmars has quit IRC | 21:02 | |
| _sweston_ | ok | 21:02 |
| * _sweston_ is waiting to proceed | 21:03 | |
| *** Ryan_Lane has joined #openstack-infra | 21:04 | |
| *** rfolco has quit IRC | 21:05 | |
| * _sweston_ and reading etherpads | 21:05 | |
| *** marcoemorais has quit IRC | 21:05 | |
| openstackgerrit | Clark Boylan proposed a change to openstack-infra/devstack-gate: update grenade branches https://review.openstack.org/90231 | 21:06 |
| clarkb | sdague: dtroyer jogo I think ^ is ready now as the grenade change merged | 21:06 |
| clarkb | sdague: dtroyer jogo we actually test d-g on the partial grenade test now so it should self gate properly | 21:06 |
| jeblair | clarkb: i expect the process is that we stop puppet on review, merge the upgrade changes, stop gerrit on review, change dns, run puppet on new-review, do whatever upgrade steps need to be done, repeat as necessary, start gerrit on new-review | 21:07 |
| clarkb | jeblair: sounds good | 21:08 |
| *** maxbit has joined #openstack-infra | 21:09 | |
| *** marcoemorais has joined #openstack-infra | 21:09 | |
| jeblair | so we'll perform the 2.4->2.8 upgrade steps on new-review on recently made copies of all the data during the downtime | 21:09 |
| jeblair | old-review will remain in the pre-upgrade state in case a rollback is needed | 21:09 |
| anteaya | jeblair: what can _sweston_ and I do to help? | 21:11 |
| anteaya | I think the dev docs are done | 21:11 |
| anteaya | https://etherpad.openstack.org/p/ai9irqk3FC | 21:11 |
| openstackgerrit | A change was merged to openstack-infra/config: Publish rally plots https://review.openstack.org/89562 | 21:12 |
| *** smarcet has quit IRC | 21:12 | |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/config: Move cinder docs into sysadmin section https://review.openstack.org/90232 | 21:13 |
| jogo | clarkb: woot | 21:13 |
| jlk | clarkb: do you have any flow project types in jenkins job builder? | 21:14 |
| jeblair | anteaya: i think they should have more prose. i will try to work on them when i'm finished with the server. | 21:14 |
| clarkb | jlk: not really. We rely on zuul for scheduling not jenkins | 21:15 |
| jlk | gotcha. I | 21:16 |
| clarkb | jlk: it is possible to do though. I think hashar put at least on example in jjb | 21:16 |
| *** dizquierdo has joined #openstack-infra | 21:16 | |
| clarkb | jlk: https://git.openstack.org/cgit/openstack-infra/jenkins-job-builder/tree/samples/pipeline.yaml something like that maybe? | 21:16 |
| jlk | clarkb: is there any language to do list joins or anything like that? I'm looking at the example build flow, and I want the list of parallel jobs to be generated somehow | 21:16 |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/config: Move cinder docs into sysadmin section https://review.openstack.org/90232 | 21:16 |
| clarkb | jlk: not really. I think there is a change up to do more than just string types in variables | 21:17 |
| anteaya | jeblair: okay | 21:18 |
| jlk | ok. | 21:18 |
| * clarkb looks for it | 21:18 | |
| *** hashar has joined #openstack-infra | 21:18 | |
| jlk | I'm finding myself wanting to use jinja tempting inside the yaml :) | 21:18 |
| anteaya | jeblair: hmmmm, I too with think about more prose, though you probably have something composed in your head | 21:18 |
| *** maxbit_ has joined #openstack-infra | 21:19 | |
| jeblair | anteaya: not really; if you want to write, go for it. basically, i want something that 2000 developers can spend about 2 minutes reading and know how to deal with the new gerrit. | 21:19 |
| clarkb | jlk: there have been times where I want JJB to use jinja natively | 21:19 |
| clarkb | but there is some concern of having the salt problem | 21:20 |
| jlk | what's the salt problem? | 21:20 |
| jeblair | anteaya: also, they will not be motivated to read it. | 21:20 |
| anteaya | jeblair: I'll try again | 21:20 |
| clarkb | jlk: a turing complete language in your config files is insane :) | 21:20 |
| jlk | fair. | 21:20 |
| jeblair | clarkb: no it's not it's awesome | 21:20 |
| greghaynes | What, you dont just want to rewrite salt in jinja? | 21:20 |
| clarkb | jeblair: :P | 21:20 |
| *** dangers_away has quit IRC | 21:20 | |
| *** e0ne has quit IRC | 21:21 | |
| jeblair | clarkb: however, the jinja problem is no one knows how to use templates right, so doing something like "gate-{name}-docs" is really hard. and that's the problem we wanted to solve. :) | 21:21 |
| clarkb | right you would have to do for name in namevars: gate-{name}-docs and so on | 21:22 |
| clarkb | I am sure there is a way to make it less terrible | 21:22 |
| *** markmcclain has quit IRC | 21:23 | |
| openstackgerrit | Monty Taylor proposed a change to openstack-infra/config: Add flag for unbound's resolv.conf https://review.openstack.org/90234 | 21:23 |
| openstackgerrit | Monty Taylor proposed a change to openstack-infra/config: Add support for disk-image-builder in nodepool https://review.openstack.org/88479 | 21:23 |
| mordred | jeblair, clarkb: ^^ the first patch would be helpful to me | 21:23 |
| clarkb | mordred: you really didn't like my idea of the local resolver on your build host? | 21:24 |
| *** dangers_away has joined #openstack-infra | 21:24 | |
| mordred | clarkb: it doesn't help | 21:24 |
| *** yamahata has quit IRC | 21:24 | |
| clarkb | why not? | 21:24 |
| mordred | clarkb: the install.d scripts run in the node via qemu | 21:24 |
| clarkb | mordred: only as a mount though | 21:25 |
| *** dangers_away is now known as dangers | 21:25 | |
| clarkb | so your udp and tcp services on the host should be available right? | 21:25 |
| mordred | nope | 21:25 |
| clarkb | why don't they start services in the image then? | 21:25 |
| mordred | gethostbyname() is going to see /etc/resolv.conf on the image host | 21:25 |
| clarkb | mordred: which will say 127.0.0.1 | 21:26 |
| mordred | because they're making images, and running services introduces entropy issues | 21:26 |
| mordred | oh - hrm. | 21:26 |
| clarkb | 127.0.0.1:53 on the host can have a resolver | 21:26 |
| jeblair | clarkb: where does it come from? | 21:26 |
| mordred | lemme try - I follow what you're saying now | 21:26 |
| clarkb | jeblair: in the case of nodepool from nodepool puppet | 21:26 |
| clarkb | jeblair: for mordred's dib build dib would probably have to ninja that in | 21:27 |
| jeblair | clarkb: mordred's dib is for nodepool puppet; i'm not sure how they are different | 21:27 |
| clarkb | jeblair: because you can run nodepool dib without openstack puppet | 21:27 |
| mordred | if running a caching nameserver on the build hosts fixes the problem | 21:27 |
| mordred | then it'll fix the problem | 21:27 |
| mordred | in both places | 21:28 |
| mordred | but it will mean that anyone who wants to build these images will be required to have a local nameserver | 21:28 |
| jeblair | clarkb: yeah, let's not concern ourselves with that for the moment though since we won't be running nodepool dib without openstack puppet | 21:28 |
| mordred | so really - the puppet patch above keeps things simple | 21:28 |
| *** _sweston_ is now known as sweston | 21:28 | |
| clarkb | jeblair: right I don't think this is a problem for us | 21:29 |
| clarkb | it is a problem for mordred testing on his laptop though | 21:29 |
| mordred | for people - in that they should really be able to do git clone openstack-infra/config ; pip install diskimage-builder ; tools/build-image and it shold just work | 21:29 |
| *** Ryan_Lane1 has joined #openstack-infra | 21:29 | |
| mordred | and if it's a problem for me testing on my laptop, it'sa barrier for other people too | 21:29 |
| jeblair | clarkb: i don't understand how they are different. mordred's laptop == nodepool.o.o | 21:29 |
| zaro | sweston: how far along are you with upgrade? | 21:29 |
| clarkb | jeblair: it doesn't because mordred's laptop isn't managed by openstack puppet | 21:29 |
| clarkb | jeblair: openstack puppet installs a local resolver | 21:29 |
| openstackgerrit | Ruslan Kamaldinov proposed a change to openstack-infra/config: Extract Murano python jobs in a template https://review.openstack.org/87974 | 21:30 |
| clarkb | jeblair: having a local resolver already exist fixes mordreds problem | 21:30 |
| jeblair | clarkb: i was given to understand that dib runs inside of a qemu vm | 21:30 |
| clarkb | jeblair: it does not | 21:30 |
| clarkb | it runs on a qemu nbd mount or similar | 21:30 |
| anteaya | my latest attempt at developer info for the Gerrit upgrade: https://etherpad.openstack.org/p/TO1UOA0H3x | 21:30 |
| clarkb | it is more like a chroot than a VM | 21:30 |
| mordred | sorry - I was wrong - it's a qemu nbd mount | 21:30 |
| jeblair | well that's completely different | 21:31 |
| openstackgerrit | Ruslan Kamaldinov proposed a change to openstack-infra/config: Extract Murano python jobs in a template https://review.openstack.org/87974 | 21:31 |
| mordred | yeah. my bad from the earlier conversation | 21:31 |
| jeblair | and i'm actually surprised mordred's laptop doesn't have a local resolver | 21:31 |
| jeblair | networkmanagery things in ubuntu just do that these days | 21:31 |
| mordred | jeblair: nameserver 127.0.1.1 | 21:31 |
| jeblair | ha | 21:32 |
| jeblair | so close | 21:32 |
| clarkb | oh hey mine does have a local resolver | 21:32 |
| clarkb | 127.0.0.1 and 127.0.1.1 are basically the same thing though | 21:32 |
| clarkb | since the kernel will forward packets if there isn't a specific match | 21:32 |
| jeblair | my workstation has nameserver 127.0.0.1 | 21:32 |
| jeblair | it's precise though, maybe they changed. | 21:32 |
| sweston | zaro: I'd like to test out the process, but I still don't know if this will be helpful to anyone | 21:32 |
| *** signed8bit has quit IRC | 21:33 | |
| *** zns has quit IRC | 21:33 | |
| *** dangers is now known as dangers_away | 21:33 | |
| *** dangers_away has quit IRC | 21:33 | |
| *** signed8bit has joined #openstack-infra | 21:33 | |
| jeblair | sweston: any issues you can find before the downtime on monday would be very helpful | 21:33 |
| *** nati_ueno has quit IRC | 21:33 | |
| *** Ryan_Lane1 has quit IRC | 21:34 | |
| clarkb | huh in this case I don't get 127.0.1.1 via 127.0.0.1 | 21:34 |
| clarkb | is that a sockopt? | 21:34 |
| clarkb | or is it just that you don't need an explicit interface for it to work? | 21:34 |
| *** dangers_away has joined #openstack-infra | 21:35 | |
| sweston | jeblair: zaro: so why don't I start with 2.4 and then go through the steps here: https://etherpad.openstack.org/p/gerrit_upgrade_script | 21:35 |
| clarkb | woo non standard fun behavior that is useful | 21:35 |
| jeblair | erm, we don't have noatime set on static.o.o | 21:37 |
| mordred | clarkb, jeblair: so, since local resolvers on laptops seem to be very weird on current ubuntu ... do we think I shoudl keep beating against that? or just do the puppet patch? | 21:38 |
| zaro | sweston: sound like a good plan, although i would make one change. on the db upgrade part i would use the db from review.o.o | 21:38 |
| * anteaya goes for a walk | 21:39 | |
| sweston | zaro: review.o.o is ? | 21:39 |
| zaro | sweston: review.openstack.org (gerrit 2.4.4) | 21:39 |
| jeblair | zaro: except we don't just hand out the db from review.o.o; it has a lot of private info (eg, email addresses). | 21:40 |
| *** hashar has quit IRC | 21:40 | |
| sweston | zaro: jeblair: that was my thought as well. Can we figure out a way to protect that data then? | 21:41 |
| sweston | for example, do an export and scrub the addresses? | 21:41 |
| *** markmcclain has joined #openstack-infra | 21:42 | |
| jeblair | sweston: i don't believe so in the time we have remaining. however, i believe either zaro or mordred has run the mysql steps on a copy of the production data, so we at least have verified that part should be okay. | 21:42 |
| *** bknudson has quit IRC | 21:42 | |
| sweston | zaro: jeblair: ok, that's fine -- but I won't be able to get any extra data for you on timing without the database | 21:43 |
| zaro | i have which i can do again to double check. | 21:44 |
| zaro | it's been a while. | 21:44 |
| *** changbl has quit IRC | 21:44 | |
| jeblair | sweston: i know. it's unfortunate but i think that's just where we are now. | 21:45 |
| *** zns has joined #openstack-infra | 21:45 | |
| *** thomasem has quit IRC | 21:46 | |
| dhellmann | hi, all! | 21:46 |
| sweston | jeblair: ok, just trying to manage any expectations, and want to be clear about what I can do :-) | 21:46 |
| jeblair | the cinder volume at /home/gerrit2 is set up now | 21:47 |
| clarkb | dhellmann: hello there | 21:47 |
| *** thomasem has joined #openstack-infra | 21:47 | |
| jeblair | sweston: yeah, i think the value at this point is in finding procedural errors | 21:48 |
| *** thomasem has quit IRC | 21:48 | |
| * dhellmann heard clarkb taking oslo's name in vain earlier | 21:48 | |
| *** mriedem has quit IRC | 21:48 | |
| clarkb | dhellmann: sorry | 21:48 |
| *** ominakov has joined #openstack-infra | 21:48 | |
| dhellmann | clarkb: np, just kidding :-) | 21:48 |
| clarkb | but if we can avoid any more oslo.vendor libs it was worth it | 21:48 |
| sweston | jeblair: I understand. | 21:48 |
| dhellmann | fwiw, I wasn't excited about oslo.vmware either, but they were going to share the code one way or the other, and this way we have have some input | 21:49 |
| mordred | oslo.vmware makes perfect sense | 21:49 |
| dhellmann | are git repo imports still being done by hand, or should we plan to schedule https://review.openstack.org/#/c/78955/ landing? | 21:49 |
| mordred | dhellmann: they're fixed | 21:49 |
| boris-42 | clarkb thank you! | 21:49 |
| * sweston is getting started on the upgrade script testing | 21:49 | |
| clarkb | mordred: imo it belongs in a vmware lib not tied to openstack | 21:50 |
| dhellmann | yeah, in the end it includes code specific to vmware and openstack, and according to the devs not useful anywhere else, which exactly what we'd want in an oslo lib like that | 21:50 |
| clarkb | mordred: openstack can then consume that | 21:50 |
| clarkb | dhellmann: right the bug there is that openstack is special | 21:50 |
| jeblair | mordred: except i've disabled manage-projects on review.o.o for the upgrade | 21:50 |
| jeblair | so we're actually back to running manually if we do want to import | 21:50 |
| mordred | jeblair: good point | 21:51 |
| mordred | clarkb: I disagree | 21:51 |
| mordred | it shoudl be here | 21:51 |
| mordred | because we know how to collaborate | 21:51 |
| jeblair | also, we would have seen that weeks ago if it had "new-project" as the topic | 21:51 |
| dhellmann | the import can wait (that change isn't approved yet anyway) if you're doing an upgrade | 21:51 |
| *** _nadya_ has quit IRC | 21:51 | |
| jeblair | dhellmann: that's scheduled for monday | 21:51 |
| dhellmann | jeblair: "new-project"? | 21:51 |
| dhellmann | as the git topic? | 21:51 |
| jeblair | dhellmann: yep | 21:51 |
| dhellmann | er, gerrit topic? | 21:51 |
| jeblair | yep yep | 21:51 |
| dhellmann | is that a new thing? | 21:52 |
| * dhellmann goes to update his growing list of instructions | 21:52 | |
| mordred | dhellmann: been there for 2 years :) | 21:52 |
| mordred | dhellmann: git review -t new-project | 21:52 |
| *** e0ne has joined #openstack-infra | 21:52 | |
| jeblair | as of like 1 or 2 months ago; we sent an announcement and updated the infra instructions | 21:52 |
| *** derekh has joined #openstack-infra | 21:52 | |
| mordred | oh - I thought you were asking about gerrit topics at all - thus the snark | 21:52 |
| dhellmann | jeblair: ok, I must have missed those, I'll go look at the archives | 21:52 |
| * mordred goes back into hole | 21:52 | |
| jeblair | dhellmann: just out of curiousity (not implying any deficiency on your part) -- are you aware of the gerrit downtime and upgrade scheduled for monday? | 21:53 |
| *** gondoi is now known as zz_gondoi | 21:53 | |
| dhellmann | jeblair: I haven't been paying attention in this channel. Was it announced elsewhere? | 21:53 |
| *** zz_gondoi is now known as gondoi | 21:53 | |
| jeblair | dhellmann: twice to the openstack-dev list (so far; i anticipate one more, and then a post-upgrade announcement) | 21:54 |
| dhellmann | I don't see many unread messages in my dev mailbox, so I may have skipped over the announcement | 21:54 |
| jeblair | i'm starting to think we need an infra-announce list | 21:54 |
| *** asselin_ has quit IRC | 21:54 | |
| *** mdenny has quit IRC | 21:54 | |
| dhellmann | ah, I see the email from zaro | 21:54 |
| *** adalbas has quit IRC | 21:54 | |
| dhellmann | a perfectly clear subject line, that I probably skipped over | 21:54 |
| *** asselin_ has joined #openstack-infra | 21:55 | |
| *** mdenny has joined #openstack-infra | 21:55 | |
| dhellmann | maybe next time "DOWNTIME: gerrit upgrade $date" would be better? | 21:55 |
| dhellmann | or I could be more careful, too :-) | 21:55 |
| jeblair | dhellmann: well, it was missing "Subject: [everyone] [IMPORTANT] [please read] [nova] [cinder] [neutron] [openstack] ...." :) | 21:55 |
| dhellmann | maybe instead of a separate list, just a separate topic? | 21:55 |
| dhellmann | heh | 21:55 |
| *** dims has quit IRC | 21:56 | |
| dhellmann | a topic would make it easy to filter those to my inbox where I wouldn't miss them, but so would a separate list | 21:56 |
| dhellmann | I don't know what other people's workflow looks like so I'm not sure which is better | 21:56 |
| *** e0ne has quit IRC | 21:57 | |
| *** david-lyle has quit IRC | 21:57 | |
| jeblair | anyway, i suspect we are going to surprise several people who aren't taking next week off. oh well. | 21:58 |
| clarkb | everyone doesn't skim all the messages anyways? | 21:59 |
| clarkb | I don't trust other people to filter for me :) | 21:59 |
| dhellmann | clarkb: only the subject lines, and obviously I missed some | 21:59 |
| *** KurtMartin has quit IRC | 22:00 | |
| *** KurtMartin has joined #openstack-infra | 22:00 | |
| *** jgrimm has quit IRC | 22:02 | |
| mordred | jeblair: when we set the irc topic , and then when gerrit stops working, I betch people will notice | 22:02 |
| *** gondoi is now known as zz_gondoi | 22:03 | |
| jeblair | mordred: yep. maybe we'll have statusbot send a notice tomorrow | 22:04 |
| jeblair | (we can put the message and docs in a wiki page and have it broadcast a link) | 22:04 |
| *** mfer has quit IRC | 22:06 | |
| *** marcoemorais has quit IRC | 22:06 | |
| *** amcrn has quit IRC | 22:08 | |
| *** marcoemorais has joined #openstack-infra | 22:08 | |
| *** dims has joined #openstack-infra | 22:08 | |
| *** yamahata has joined #openstack-infra | 22:09 | |
| *** dizquierdo_ has joined #openstack-infra | 22:11 | |
| *** marcoemorais has quit IRC | 22:12 | |
| *** adalbas has joined #openstack-infra | 22:12 | |
| *** marcoemorais has joined #openstack-infra | 22:14 | |
| *** e0ne has joined #openstack-infra | 22:16 | |
| *** _nadya_ has joined #openstack-infra | 22:17 | |
| *** e0ne_ has joined #openstack-infra | 22:18 | |
| *** e0ne has quit IRC | 22:21 | |
| *** e0ne_ has quit IRC | 22:22 | |
| *** dkranz has quit IRC | 22:24 | |
| *** zns has quit IRC | 22:24 | |
| *** doug-fish1 has left #openstack-infra | 22:26 | |
| *** ihrachyshka has quit IRC | 22:27 | |
| *** zns has joined #openstack-infra | 22:28 | |
| openstackgerrit | Sean Dague proposed a change to openstack-infra/gitdm: adjusting affiliation https://review.openstack.org/90243 | 22:29 |
| clarkb | ah man I won't be around for your first day? | 22:30 |
| sdague | slacker :) | 22:31 |
| jeblair | sdague: congrats! i like this way of announcing changes... if we -1 that change do you stay at samsung? ;) | 22:31 |
| JayF | Is that yet another place to put in our info, along with stackalytics and reviewstats? | 22:31 |
| sdague | heh, I hope not :) | 22:31 |
| clarkb | JayF: there are far too many | 22:31 |
| *** dizquierdo_ has quit IRC | 22:31 | |
| sdague | reviewstats has affiliation? | 22:31 |
| *** dizquierdo has quit IRC | 22:31 | |
| JayF | it doesn't have affiliation, but iirc something had to update there when I joined? Maybe listing me as a core for ipa? | 22:32 |
| boris-42 | clarkb http://logs.openstack.org/45/86945/19/check/check-rally-dsvm-rally/f53b058/rally-plot/ !!! | 22:33 |
| boris-42 | clarkb it works~ | 22:33 |
| boris-42 | clarkb thank you=) | 22:33 |
| clarkb | cool | 22:33 |
| *** doddstack has quit IRC | 22:34 | |
| sdague | I honestly completely forgot gitdm was publishing in channel, but announcing is good. | 22:34 |
| *** dkliban is now known as dkliban_afk | 22:36 | |
| *** mriedem has joined #openstack-infra | 22:37 | |
| anteaya | sdague: did you want to add a line to http://git.openstack.org/cgit/openstack-infra/gitdm/tree/openstack-config/groups/hp as well? | 22:39 |
| sdague | anteaya: honestly, IIRC, the groups files are trumped by the email-map | 22:39 |
| sdague | the gitdm layers of trumping is pretty byzantine | 22:40 |
| pleia2 | sdague: woo, welcome-ish :) | 22:40 |
| sdague | clarkb / jeblair / mordred: on the more useful commit front - https://review.openstack.org/#/c/90226/ - to help reduce some noise in the er fails, and stop burning nodes | 22:41 |
| sdague | pleia2: thanks :) | 22:41 |
| sdague | ok, my wife just got home, time to go afk | 22:41 |
| anteaya | ah I didn't know the trumping | 22:42 |
| *** enikanorov_ has quit IRC | 22:42 | |
| *** zns has quit IRC | 22:43 | |
| *** bknudson has joined #openstack-infra | 22:44 | |
| *** derekh has quit IRC | 22:45 | |
| openstackgerrit | A change was merged to openstack-infra/elastic-recheck: Add check-grenade-dsvm-neutron-icehouse to bug 1307344 query https://review.openstack.org/90207 | 22:46 |
| uvirtbot | Launchpad bug 1307344 in neutron "neutron migrations are not idempotent or invariant to enabled services" [Critical,Fix committed] https://launchpad.net/bugs/1307344 | 22:46 |
| anteaya | so dhellmann can you do me a favour? can I get your feedback on: https://etherpad.openstack.org/p/TO1UOA0H3x | 22:48 |
| anteaya | if we were to send that out to 2000 devs, would it make sense? | 22:48 |
| openstackgerrit | A change was merged to openstack-infra/config: move tempest-master-*-havana to experimental https://review.openstack.org/90226 | 22:48 |
| *** msabramo has quit IRC | 22:48 | |
| *** signed8bit is now known as signed8bit_ZZZzz | 22:50 | |
| openstackgerrit | A change was merged to openstack-infra/reviewday: Add Sahara repos to reviewday https://review.openstack.org/88214 | 22:53 |
| openstackgerrit | Monty Taylor proposed a change to openstack-infra/config: Add support for disk-image-builder in nodepool https://review.openstack.org/88479 | 22:55 |
| openstackgerrit | Monty Taylor proposed a change to openstack-infra/config: Add flag for unbound's resolv.conf https://review.openstack.org/90234 | 22:55 |
| *** jamielennox|away is now known as jamielennox | 22:56 | |
| *** mdenny has quit IRC | 22:56 | |
| *** etoews has joined #openstack-infra | 22:57 | |
| *** amcrn has joined #openstack-infra | 22:57 | |
| *** msabramo has joined #openstack-infra | 22:59 | |
| *** dkehn_ has joined #openstack-infra | 23:00 | |
| *** pfallenop has quit IRC | 23:01 | |
| *** pfallenop has joined #openstack-infra | 23:01 | |
| *** clarkb has quit IRC | 23:01 | |
| *** clarkb has joined #openstack-infra | 23:01 | |
| clarkb | did I miss anything? | 23:02 |
| *** odyi has quit IRC | 23:02 | |
| clarkb | mordred: jeblair: yjiang over in #openstack-qa had a nova pep8 job run out of disk on an hpcloud az2 bare precise node | 23:02 |
| *** adalbas has quit IRC | 23:02 | |
| *** dkehn has quit IRC | 23:03 | |
| clarkb | we can't do anything with the ephemeral disks during image buildtime ... | 23:03 |
| *** dkehn has joined #openstack-infra | 23:03 | |
| clarkb | not sure how else we can deal with this | 23:03 |
| *** odyi has joined #openstack-infra | 23:03 | |
| clarkb | looks like jobs have 3.5GB of disk in / though | 23:03 |
| clarkb | which should be plenty for pep8 | 23:03 |
| *** dkehnx has quit IRC | 23:04 | |
| *** lifeless has quit IRC | 23:04 | |
| clarkb | woot https://review.openstack.org/#/c/90231/ passed | 23:04 |
| *** lifeless has joined #openstack-infra | 23:05 | |
| anteaya | yay | 23:05 |
| anteaya | devstack-gate has a lot of tests | 23:05 |
| *** markmcclain has quit IRC | 23:07 | |
| *** flaper87 is now known as flaper87|afk | 23:08 | |
| *** CaptTof__ has quit IRC | 23:08 | |
| *** atiwari has quit IRC | 23:09 | |
| *** CaptTofu has joined #openstack-infra | 23:09 | |
| *** msabramo has quit IRC | 23:11 | |
| *** msabramo has joined #openstack-infra | 23:11 | |
| *** melwitt has quit IRC | 23:11 | |
| *** pmathews has quit IRC | 23:12 | |
| *** CaptTofu has quit IRC | 23:14 | |
| *** mburned is now known as mburned_out | 23:14 | |
| *** lukego has quit IRC | 23:16 | |
| *** e0ne has joined #openstack-infra | 23:18 | |
| *** adalbas has joined #openstack-infra | 23:18 | |
| *** lukego has joined #openstack-infra | 23:21 | |
| clarkb | mordred pbr should do versioning like this https://twitter.com/stahnma/status/459452329637326849/photo/1 | 23:22 |
| *** e0ne has quit IRC | 23:22 | |
| mordred | wow | 23:22 |
| anteaya | ruby releasing | 23:25 |
| anteaya | never a strong point | 23:25 |
| *** MarkAtwood has joined #openstack-infra | 23:26 | |
| *** gokrokve has quit IRC | 23:26 | |
| *** devoid has quit IRC | 23:28 | |
| *** CaptTofu has joined #openstack-infra | 23:29 | |
| arosen | hrm is this a gerrit bug why there is no syntax hightlighting on this page? https://review.openstack.org/#/c/82831/13/nova/tests/virt/libvirt/test_libvirt.py | 23:31 |
| *** Ryan_Lane1 has joined #openstack-infra | 23:31 | |
| *** mwagner_lap has joined #openstack-infra | 23:31 | |
| *** ominakov has quit IRC | 23:32 | |
| *** reed has quit IRC | 23:32 | |
| *** Ryan_Lane1 has quit IRC | 23:35 | |
| anteaya | arosen: isn't that interesting, there is syntax highlighting in the other .py file in that patch | 23:35 |
| arosen | anteaya: cha | 23:36 |
| arosen | And if you click on ... skipping <<HERERER 843> common lines .. | 23:36 |
| arosen | it just adds random +1 lines. | 23:36 |
| anteaya | that's wild | 23:37 |
| anteaya | well done, I have never seen that before | 23:37 |
| anteaya | I wouldn't even know how to describe that as a bug report | 23:37 |
| *** oomichi has joined #openstack-infra | 23:38 | |
| anteaya | arosen: since we are about to upgrade next week to gerrit 2.8.4, I don't think filing a bug is worth your time | 23:39 |
| anteaya | but it is interesting | 23:39 |
| arosen | yea wasn't planning on it. Just had never seen it before | 23:39 |
| *** MarkAtwood has quit IRC | 23:39 | |
| anteaya | me either | 23:39 |
| arosen | figured i'd point it out. Not worried about it thou | 23:39 |
| anteaya | k | 23:40 |
| sdague | clarkb: I have the revert revert version of that up actually | 23:40 |
| sdague | clarkb: https://review.openstack.org/#/c/90043/ | 23:40 |
| *** ravikumar_hp has quit IRC | 23:41 | |
| clarkb | oh even better | 23:42 |
| clarkb | can you abandon mine | 23:43 |
| sdague | sure | 23:43 |
| clarkb | ty | 23:43 |
| *** e0ne has joined #openstack-infra | 23:43 | |
| sdague | actually, I don't seem to have a revert button | 23:43 |
| sdague | sorry, the abandon button | 23:44 |
| sdague | on https://review.openstack.org/#/c/90231/ | 23:44 |
| sdague | maybe because you are the committer? | 23:44 |
| clarkb | huh I thought core had it | 23:44 |
| clarkb | feel free to -2 instead | 23:45 |
| *** lukego has quit IRC | 23:45 | |
| sdague | guess not | 23:45 |
| *** gokrokve has joined #openstack-infra | 23:46 | |
| *** e0ne has quit IRC | 23:47 | |
| *** masayukig has joined #openstack-infra | 23:52 | |
| mordred | clarkb, jeblair: I wrote a script to process nodepool.log and produce a success/fail info - it's sitting in /var/log/nodepool right now and is called process.py | 23:52 |
| *** UtahDave has quit IRC | 23:53 | |
| *** yamahata has quit IRC | 23:57 | |
| jogo | heh sdague has two compeating patches | 23:58 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!