| *** yamahata has quit IRC | 00:01 | |
| *** sweston has joined #openstack-infra | 00:02 | |
| *** yamahata has joined #openstack-infra | 00:04 | |
| mordred | morning clarkb jhesketh | 00:06 |
|---|---|---|
| mordred | anything I should peep? | 00:06 |
| clarkb | mordred: just maybe the stack at https://review.openstack.org/#/c/101283/ | 00:09 |
| *** lcheng has quit IRC | 00:16 | |
| *** CaptTofu_ has quit IRC | 00:16 | |
| *** CaptTofu_ has joined #openstack-infra | 00:17 | |
| *** CaptTof__ has joined #openstack-infra | 00:18 | |
| *** dkehn_ is now known as dkehnx | 00:20 | |
| *** CaptTofu_ has quit IRC | 00:22 | |
| zaro | jhesketh, clarkb : why remove envinject? i think lots of projects use it. | 00:22 |
| zaro | but can help if we decide to remove. | 00:23 |
| clarkb | zaro: bceause gearman plugin doesn't need it | 00:25 |
| clarkb | and the current version of envinject doesn't work | 00:26 |
| clarkb | FYI I held node 476349 for the tripleo folks. They need access to a running test to debug a thing | 00:26 |
| *** flaviof is now known as flaviof_zzz | 00:26 | |
| *** matsuhashi has joined #openstack-infra | 00:28 | |
| mordred | clarkb: looking | 00:28 |
| mordred | clarkb: zomg. that's so many patches | 00:28 |
| mordred | clarkb, zaro: where did envinject come from? | 00:29 |
| *** ramashri has quit IRC | 00:30 | |
| clarkb | mordred: last night jhesketh found that envinject chomps newlines in the hmac string for swift uploads | 00:31 |
| mordred | AWESOME | 00:31 |
| clarkb | we can either upgrade envinject or remove it | 00:31 |
| clarkb | jeblair's suggest is remove it | 00:31 |
| mordred | well, we don't use it anymore for anything, do we? | 00:31 |
| clarkb | which gives us the other benefit of masking the hmac out so that it can't be abused | 00:31 |
| clarkb | or easily abused | 00:31 |
| mordred | clarkb: ooh. I like the s/bare// | 00:32 |
| *** sbalukoff has joined #openstack-infra | 00:32 | |
| clarkb | mordred: hrm? | 00:33 |
| clarkb | did I remove bare somewhere? I am not sure that was intentional | 00:33 |
| mordred | clarkb: in one place | 00:33 |
| clarkb | rally? | 00:33 |
| mordred | but now I'm noticing it wasn't pervasive | 00:33 |
| clarkb | I think I decided in a couple places that the rally jobs needed hammering | 00:33 |
| mordred | clarkb: https://review.openstack.org/#/c/101303/4/modules/openstack_project/files/zuul/layout.yaml | 00:34 |
| clarkb | I'm not sure that is a good thing. We should probably try to be more uniform even with stackforge | 00:34 |
| *** arun has quit IRC | 00:34 | |
| clarkb | ya | 00:34 |
| mordred | ++ | 00:36 |
| mordred | clarkb: actually, boris-42 said he was goign to start working through making us patches to start making more things use templates | 00:36 |
| clarkb | cool | 00:36 |
| clarkb | that would be great | 00:36 |
| mordred | clarkb: something about this set of patches you had to make seems clunky to me. I don't have a suggestion for how we rework things, and I don't think it's important to | 00:38 |
| mordred | clarkb: but it makes me think our data model or abstraction is missing a beat somewhere | 00:38 |
| clarkb | I thnik it is as well | 00:38 |
| mordred | clarkb: any reason I should hold off on +A's on this stack? | 00:39 |
| clarkb | mordred: yes, they need to go in in lock step | 00:39 |
| mordred | ok. | 00:39 |
| clarkb | we want to do first change, let zuul update, do second change, let jenkins update, then do third change | 00:39 |
| clarkb | and each step may require debugging | 00:39 |
| clarkb | so I think we will wait for tomorrow when jeblair and sdague are more around too | 00:40 |
| clarkb | mordred: I think the abstraction we are having trouble with is we define the node in two places | 00:40 |
| clarkb | mordred: we should move towards only having zuul set the node | 00:40 |
| clarkb | mordred: but, I am doing it this way to hopefully give us a little extra protection against breakage | 00:40 |
| clarkb | as removing the node labels from jjb now would make all the test fail that get the wrong node | 00:40 |
| *** basha has joined #openstack-infra | 00:41 | |
| *** basha has quit IRC | 00:43 | |
| mordred | clarkb: yah. | 00:43 |
| mordred | clarkb: weirdly, I like the new code in openstack-functions - because it shows a little more that you can do that | 00:44 |
| *** arun has joined #openstack-infra | 00:45 | |
| *** arun has joined #openstack-infra | 00:45 | |
| *** annegent_ has joined #openstack-infra | 00:46 | |
| mordred | clarkb: how does set_node_options_default_precise get called? | 00:46 |
| mordred | oh. nevermind. duch | 00:46 |
| clarkb | ya if you look at an older patchset for the last change you will see why I split those | 00:46 |
| clarkb | but that last change is now used for something slightly different and what it was doing in that patchset needs to become change number 5 | 00:47 |
| *** mrda has quit IRC | 00:47 | |
| clarkb | but I haven't written that yet | 00:47 |
| *** matsuhashi has quit IRC | 00:48 | |
| *** melwitt has quit IRC | 00:48 | |
| *** yamahata has quit IRC | 00:49 | |
| *** matsuhashi has joined #openstack-infra | 00:49 | |
| *** annegent_ has quit IRC | 00:50 | |
| *** mrda has joined #openstack-infra | 00:53 | |
| *** matsuhashi has quit IRC | 00:54 | |
| *** otter768 has joined #openstack-infra | 00:55 | |
| *** dims has joined #openstack-infra | 00:58 | |
| *** matsuhashi has joined #openstack-infra | 01:01 | |
| *** dims has quit IRC | 01:02 | |
| *** CaptTof__ has quit IRC | 01:04 | |
| *** CaptTofu_ has joined #openstack-infra | 01:04 | |
| *** penguinRaider has joined #openstack-infra | 01:04 | |
| *** dims_ has joined #openstack-infra | 01:05 | |
| *** Abhishek has quit IRC | 01:07 | |
| *** CaptTofu_ has quit IRC | 01:08 | |
| *** otter768 has quit IRC | 01:08 | |
| *** penguinRaider has quit IRC | 01:09 | |
| *** matsuhashi has quit IRC | 01:09 | |
| *** prad_ has quit IRC | 01:09 | |
| *** matsuhashi has joined #openstack-infra | 01:10 | |
| *** yjiang5 is now known as yjiang5_away | 01:11 | |
| *** zehicle_at_dell has quit IRC | 01:12 | |
| *** annegent_ has joined #openstack-infra | 01:15 | |
| openstackgerrit | A change was merged to openstack-infra/devstack-gate: Use the DEVSTACK_GATE_USE_APACHE flag to disable https://review.openstack.org/100744 | 01:17 |
| *** dennis_^^ has joined #openstack-infra | 01:17 | |
| *** gmurphy has joined #openstack-infra | 01:18 | |
| *** annegent_ has quit IRC | 01:20 | |
| *** reed has quit IRC | 01:20 | |
| *** gokrokve has quit IRC | 01:22 | |
| *** lcostantino has joined #openstack-infra | 01:25 | |
| *** nosnos has joined #openstack-infra | 01:30 | |
| *** chianingwang has joined #openstack-infra | 01:33 | |
| *** sweston has quit IRC | 01:33 | |
| *** annegent_ has joined #openstack-infra | 01:36 | |
| *** Ryan_Lane has quit IRC | 01:36 | |
| *** chianingwang_ has quit IRC | 01:37 | |
| *** yamahata has joined #openstack-infra | 01:39 | |
| *** dims_ has quit IRC | 01:39 | |
| *** Sukhdev has quit IRC | 01:39 | |
| *** CaptTofu_ has joined #openstack-infra | 01:44 | |
| *** marcoemorais has quit IRC | 01:46 | |
| *** gokrokve has joined #openstack-infra | 01:47 | |
| *** yaguang has joined #openstack-infra | 01:52 | |
| *** yaguang has quit IRC | 01:52 | |
| *** yaguang has joined #openstack-infra | 01:52 | |
| *** annegent_ has quit IRC | 01:53 | |
| *** CaptTof__ has joined #openstack-infra | 01:55 | |
| *** CaptTofu_ has quit IRC | 01:57 | |
| *** annegent_ has joined #openstack-infra | 02:03 | |
| *** otter768 has joined #openstack-infra | 02:04 | |
| *** lcostantino has quit IRC | 02:09 | |
| *** penguinRaider has joined #openstack-infra | 02:12 | |
| *** penguinRaider has quit IRC | 02:16 | |
| *** annegent_ has quit IRC | 02:18 | |
| *** bnemec is now known as bnemec_away | 02:20 | |
| * nibalizer trying to fix the cacti node | 02:21 | |
| nibalizer | its soo upset | 02:21 |
| *** CaptTof__ has quit IRC | 02:22 | |
| *** CaptTofu_ has joined #openstack-infra | 02:22 | |
| *** rwsu has quit IRC | 02:24 | |
| *** nati_ueno has quit IRC | 02:26 | |
| *** CaptTofu_ has quit IRC | 02:26 | |
| openstackgerrit | Spencer Krum proposed a change to openstack-infra/config: Fix cacti node and add os check https://review.openstack.org/101389 | 02:28 |
| nibalizer | that fixes at least the apache.conf file and the http://cacti.o.o not redirecting problems | 02:28 |
| nibalizer | as for why the ghraps fail idunno | 02:28 |
| *** annegent_ has joined #openstack-infra | 02:29 | |
| *** adalbas has quit IRC | 02:30 | |
| notmyname | mordred: I'll be available to do branch stuff in about 10-15 minutes. can we work on it this evening? | 02:30 |
| *** praneshp has quit IRC | 02:30 | |
| *** dkehn_ has joined #openstack-infra | 02:35 | |
| *** Johnson_Yang has joined #openstack-infra | 02:35 | |
| *** dkehn has quit IRC | 02:35 | |
| *** otherwiseguy has joined #openstack-infra | 02:36 | |
| *** Johnson_Yang has quit IRC | 02:36 | |
| *** dkehnx has quit IRC | 02:36 | |
| *** dkehn has joined #openstack-infra | 02:37 | |
| *** penguinRaider has joined #openstack-infra | 02:38 | |
| *** talluri has joined #openstack-infra | 02:42 | |
| *** penguinRaider has quit IRC | 02:44 | |
| notmyname | mordred: ready when you are | 02:45 |
| *** zhiyan_ is now known as zhiyan | 02:49 | |
| openstackgerrit | David Pursehouse proposed a change to openstack-infra/jenkins-job-builder: Consistency updates in documentation https://review.openstack.org/101394 | 02:51 |
| *** CaptTofu_ has joined #openstack-infra | 02:51 | |
| *** harlowja is now known as harlowja_away | 02:56 | |
| *** annegent_ has quit IRC | 03:00 | |
| *** weshay has quit IRC | 03:00 | |
| *** matsuhashi has quit IRC | 03:02 | |
| *** penguinRaider has joined #openstack-infra | 03:03 | |
| *** matsuhashi has joined #openstack-infra | 03:03 | |
| *** matsuhashi has quit IRC | 03:07 | |
| *** zehicle_at_dell has joined #openstack-infra | 03:09 | |
| *** Sukhdev has joined #openstack-infra | 03:14 | |
| *** mestery has quit IRC | 03:17 | |
| *** changbl has quit IRC | 03:17 | |
| *** nosnos has quit IRC | 03:18 | |
| kashyap | Hi, who would know what URL is used to download Fedora images? | 03:19 |
| clarkb | kashyap: what fedora images? | 03:19 |
| kashyap | IIRC, Sean Dague said in a recent IRC conversation on #openstack-qa that there's reliability issues for downloads of Fedora cloud images | 03:19 |
| *** changbl has joined #openstack-infra | 03:19 | |
| clarkb | oh ya the ones used by devstack | 03:19 |
| kashyap | I mentioned this to Fedora infrastructure to see if we can do anything to alleviate these | 03:19 |
| kashyap | clarkb, Yes | 03:20 |
| clarkb | listed in stackrc | 03:20 |
| *** changbl has quit IRC | 03:20 | |
| * kashyap git greps | 03:20 | |
| clarkb | you can find urls there | 03:20 |
| *** gyee has quit IRC | 03:20 | |
| kashyap | Thank you, Clark. Looking | 03:21 |
| *** dkehn__ has joined #openstack-infra | 03:22 | |
| kashyap | Found it :-) | 03:22 |
| *** dkehn_ has quit IRC | 03:23 | |
| kashyap | clarkb, The same DevStack setup is used to CI infra VMs too, right through nodepool? | 03:24 |
| *** dkehn has quit IRC | 03:24 | |
| *** dkehn has joined #openstack-infra | 03:24 | |
| kashyap | Also, nodepool seems to _cache_ the images -- https://github.com/openstack-dev/devstack/blob/master/stackrc#L360 | 03:26 |
| clarkb | yes but we cache those images on our slaves | 03:26 |
| clarkb | yup | 03:26 |
| kashyap | I don't see why there would be reliability issues once you do the first download | 03:26 |
| *** otherwiseguy has quit IRC | 03:26 | |
| clarkb | its with humans without caches | 03:26 |
| clarkb | and building the caches | 03:26 |
| clarkb | we build the caches from.scratch every 24 hours | 03:26 |
| kashyap | I see | 03:27 |
| clarkb | that will get smarter when we use dib | 03:27 |
| openstackgerrit | A change was merged to openstack/requirements: Pin httpretty's precise version. https://review.openstack.org/101214 | 03:31 |
| *** praneshp has joined #openstack-infra | 03:32 | |
| *** annegent_ has joined #openstack-infra | 03:34 | |
| *** CaptTofu_ has quit IRC | 03:34 | |
| *** praneshp_ has joined #openstack-infra | 03:34 | |
| *** changbl has joined #openstack-infra | 03:35 | |
| *** CaptTofu_ has joined #openstack-infra | 03:35 | |
| *** praneshp has quit IRC | 03:37 | |
| *** praneshp_ is now known as praneshp | 03:37 | |
| *** Johnson_Yang has joined #openstack-infra | 03:39 | |
| *** CaptTofu_ has quit IRC | 03:40 | |
| *** otter768 has quit IRC | 03:44 | |
| kashyap | clarkb, On dib, recall from a previous discussion elsewhere that dib uses losetup and mount to mount disk images, do you know top off your head if it's still the case? | 03:44 |
| *** Abhishek has joined #openstack-infra | 03:45 | |
| clarkb | I think so | 03:46 |
| mordred | notmyname: hey | 03:47 |
| notmyname | mordred: good evening | 03:48 |
| mordred | notmyname: sorry - been out most of the afternoon - you still aroudn? | 03:48 |
| notmyname | mordred: yup I'm here | 03:48 |
| *** matsuhashi has joined #openstack-infra | 03:53 | |
| *** talluri has quit IRC | 03:54 | |
| *** mwagner_lap has joined #openstack-infra | 03:55 | |
| openstackgerrit | Morgan Fainberg proposed a change to openstack-infra/config: Add keystonemiddleware project https://review.openstack.org/101406 | 03:57 |
| *** bookwar has quit IRC | 03:58 | |
| zaro | clarkb: yo, IIRC i created a few jobs with envinject. so i think it's used in a few places at least. | 03:58 |
| *** bookwar has joined #openstack-infra | 03:58 | |
| clarkb | zaro: which ones? everything is triggered by gearman | 03:59 |
| openstackgerrit | Morgan Fainberg proposed a change to openstack-infra/config: Add keystonemiddleware project https://review.openstack.org/101406 | 04:00 |
| morganfainberg | mordred, ^ so we can remove memcache (and other silly things) from the deps of keystoneclient :) | 04:01 |
| *** arnaud__ has quit IRC | 04:02 | |
| *** matsuhashi has quit IRC | 04:02 | |
| zaro | clarkb: wait, what does gearman have to do with envinject. i believe i use it to read a file with ENV vars and inject that into the build. | 04:02 |
| morganfainberg | mordred, i cant remember who else asked me that question at the summit but i swear no less than three people did (why memcache was a dep of keystoneclient) | 04:02 |
| *** matsuhashi has joined #openstack-infra | 04:02 | |
| clarkb | zaro env inject interprets the gearman parameters | 04:03 |
| clarkb | and it does so wrongly | 04:03 |
| clarkb | this is why we want to get rid of it | 04:03 |
| zaro | clarkb: maven-plugin-jobs.yaml , jenkins-plugin-jobs.yaml and gearman-plugin-jobs.yaml | 04:04 |
| clarkb | thanks | 04:04 |
| zaro | a few more places as well. do a git grep inject | 04:06 |
| *** matsuhashi has quit IRC | 04:06 | |
| openstackgerrit | Morgan Fainberg proposed a change to openstack-infra/config: Add keystonemiddleware project https://review.openstack.org/101406 | 04:07 |
| zaro | clarkb: have you completely updated build-timeout plugin? | 04:07 |
| clarkb | so I wonder if we can make it leave gearman alone | 04:08 |
| clarkb | zaro: no only 07 is done | 04:08 |
| clarkb | still no trouble so we can start the others | 04:08 |
| zaro | clarkb: or we can use a different plugin. i think we can use envfile plugin instead. | 04:10 |
| *** nosnos has joined #openstack-infra | 04:11 | |
| zaro | i mean remove envinject and refactor those jobs to use envfile plugin instead. | 04:12 |
| mordred | clarkb: notmyname was asking if there was a way to do <ul> in a review comment | 04:12 |
| openstackgerrit | Morgan Fainberg proposed a change to openstack-infra/config: Add keystonemiddleware project https://review.openstack.org/101406 | 04:13 |
| clarkb | there is a small markup language available iirc | 04:13 |
| clarkb | zaro gotcha | 04:13 |
| clarkb | you can use * for list items for example | 04:15 |
| clarkb | but I think its completely undocumented as to what is supported | 04:15 |
| notmyname | https://wiki.hpdd.intel.com/display/PUB/2012/02/16/Comment+formatting+on+Gerrit | 04:15 |
| *** talluri has joined #openstack-infra | 04:15 | |
| *** eharney has quit IRC | 04:20 | |
| *** homeless_ has quit IRC | 04:25 | |
| zaro | notmyname: i don't even think that works anymore. that documentation has been removed. | 04:26 |
| notmyname | just did :-) | 04:26 |
| zaro | haha, buried treasure. | 04:27 |
| *** otherwiseguy has joined #openstack-infra | 04:29 | |
| *** eharney has joined #openstack-infra | 04:33 | |
| *** annegent_ has quit IRC | 04:35 | |
| *** arnaud has joined #openstack-infra | 04:42 | |
| *** CaptTofu_ has joined #openstack-infra | 04:44 | |
| *** CaptTofu_ has quit IRC | 04:48 | |
| *** Longgeek has joined #openstack-infra | 04:50 | |
| *** harlowja_at_home has joined #openstack-infra | 04:52 | |
| *** trinaths has joined #openstack-infra | 04:52 | |
| *** ildikov has quit IRC | 04:53 | |
| *** hdd_ has joined #openstack-infra | 04:56 | |
| *** openstackgerrit has quit IRC | 04:57 | |
| *** penguinRaider has quit IRC | 04:59 | |
| *** penguinRaider has joined #openstack-infra | 05:03 | |
| *** Abhishek has quit IRC | 05:05 | |
| *** basha has joined #openstack-infra | 05:07 | |
| *** Abhishek has joined #openstack-infra | 05:08 | |
| *** Longgeek has quit IRC | 05:11 | |
| *** Longgeek has joined #openstack-infra | 05:12 | |
| *** annegent_ has joined #openstack-infra | 05:13 | |
| *** Sukhdev has quit IRC | 05:14 | |
| *** Longgeek has quit IRC | 05:14 | |
| *** Longgeek has joined #openstack-infra | 05:14 | |
| *** sweston_cloud has joined #openstack-infra | 05:15 | |
| *** hdd_ has quit IRC | 05:17 | |
| *** penguinRaider has quit IRC | 05:18 | |
| *** praneshp has quit IRC | 05:19 | |
| *** otherwiseguy has quit IRC | 05:19 | |
| *** annegent_ has quit IRC | 05:20 | |
| *** harlowja_at_home has quit IRC | 05:20 | |
| *** basha has quit IRC | 05:23 | |
| *** dkehn_ has joined #openstack-infra | 05:23 | |
| *** harlowja_at_home has joined #openstack-infra | 05:25 | |
| *** sweston_cloud has quit IRC | 05:26 | |
| *** dkehn has quit IRC | 05:26 | |
| *** sweston_cloud has joined #openstack-infra | 05:27 | |
| *** dkehn__ has quit IRC | 05:27 | |
| *** dkehn has joined #openstack-infra | 05:28 | |
| *** adrian_otto has joined #openstack-infra | 05:28 | |
| *** sweston_cloud has quit IRC | 05:28 | |
| *** sweston_cloud has joined #openstack-infra | 05:30 | |
| *** sweston_cloud has quit IRC | 05:34 | |
| *** sweston has joined #openstack-infra | 05:34 | |
| *** ildikov has joined #openstack-infra | 05:34 | |
| *** harlowja_at_home has quit IRC | 05:35 | |
| *** sweston has quit IRC | 05:38 | |
| *** chandan_kumar has joined #openstack-infra | 05:40 | |
| *** basha has joined #openstack-infra | 05:42 | |
| *** _nadya_ has joined #openstack-infra | 05:42 | |
| *** _nadya_ has quit IRC | 05:45 | |
| *** mrda is now known as mrda-away | 05:48 | |
| *** adrian_otto has quit IRC | 05:50 | |
| *** talluri has quit IRC | 05:53 | |
| *** talluri has joined #openstack-infra | 05:53 | |
| *** talluri has quit IRC | 05:57 | |
| *** basha has quit IRC | 06:08 | |
| afazekas | I do not really follow what kind of change is required in 'devstack' for allocating nodes ? https://review.openstack.org/#/c/100309/ | 06:08 |
| *** nati_ueno has joined #openstack-infra | 06:09 | |
| *** gcha has quit IRC | 06:09 | |
| *** otherwiseguy has joined #openstack-infra | 06:11 | |
| *** talluri has joined #openstack-infra | 06:11 | |
| *** Abhishek has quit IRC | 06:14 | |
| *** rcarrillocruz has joined #openstack-infra | 06:18 | |
| *** otherwiseguy has quit IRC | 06:19 | |
| *** rcarrill` has quit IRC | 06:20 | |
| mattoliverau | afazekas: it's not about allocating nodes directly, looks like there was/is a bug with devstack on F20 nodes. They get to the top of the gate queue and stop everything from moving forwards. Because hold up got big we reached quota on how many nodes we can create. Moving F20 nodes into the experimental took them off the check and gate pipelinse allowing the stuck/queued changes to start flowing again. | 06:22 |
| mattoliverau | Which in turn allowed nodepool to start reap and build new nodes. | 06:22 |
| *** lukego has joined #openstack-infra | 06:25 | |
| *** Longgeek has quit IRC | 06:30 | |
| *** Longgeek_ has joined #openstack-infra | 06:30 | |
| afazekas | mattoliverau: there was an issue with different cloud images used on HP and rackspace cloud on one the firewalld was installed , on the other the selinux was enabled | 06:32 |
| *** CaptTofu_ has joined #openstack-infra | 06:32 | |
| afazekas | Now the differnces expected to handled on devstack | 06:32 |
| *** doude has joined #openstack-infra | 06:33 | |
| afazekas | https://review.openstack.org/#/c/99559/ selinux permissive | 06:34 |
| afazekas | https://review.openstack.org/#/c/99047/ | 06:37 |
| afazekas | mattoliverau: are there any other issues ? | 06:38 |
| afazekas | ianw: ^^ | 06:39 |
| *** basha has joined #openstack-infra | 06:40 | |
| *** tkelsey has joined #openstack-infra | 06:40 | |
| ianw | afazekas: it's not that, see the comments in https://review.openstack.org/#/c/101110/ | 06:40 |
| ianw | the problem is nodepool not allocating f20 nodes, not any issues with devstack on f20 | 06:42 |
| *** basha has quit IRC | 06:43 | |
| afazekas | I did not read the topic fully yet, cannot be just increasing the min pool size ? | 06:43 |
| *** CaptTofu_ has quit IRC | 06:43 | |
| ianw | afazekas: it would have to be about 50/50 ratio to avoid issues, which isn't feasible. that change, or some variant of it, is i think the correct way forward | 06:47 |
| *** basha has joined #openstack-infra | 06:47 | |
| afazekas | Is the relative gap between the minimum and the current node num is calculated into the priority ? | 06:54 |
| *** gokrokve_ has joined #openstack-infra | 06:58 | |
| *** mrmartin has joined #openstack-infra | 06:58 | |
| ianw | afazekas: not directly. the allocator doesn't know about minimums | 07:00 |
| afazekas | In similar problem set the error from the P-tag usually good for starting priority values: http://en.wikipedia.org/wiki/PID_controller | 07:01 |
| markmc | afazekas, this from ianw summarizes the zuul bug beautifully: http://lists.openstack.org/pipermail/openstack-infra/2014-June/001358.html | 07:02 |
| *** gokrokve has quit IRC | 07:02 | |
| *** mrmartin has quit IRC | 07:04 | |
| *** jlibosva has joined #openstack-infra | 07:04 | |
| *** e0ne has joined #openstack-infra | 07:05 | |
| *** bookwar has quit IRC | 07:06 | |
| *** bodepd has quit IRC | 07:07 | |
| *** basha has quit IRC | 07:07 | |
| *** bookwar has joined #openstack-infra | 07:07 | |
| *** bodepd has joined #openstack-infra | 07:08 | |
| ianw | afazekas: while PID algorithms etc are cool, I don't think it's wise to start there (KISS). however, i do think that with the history tracking provided by https://review.openstack.org/#/c/101110/ , you could expand it to something like this if it became necessary | 07:09 |
| *** freyes has joined #openstack-infra | 07:10 | |
| afazekas | if you want 1 fedora node and you get 0 is 100% error, when you wanted 30 node of Ubuntu you got 20 is 66% error | 07:10 |
| afazekas | 33% | 07:10 |
| *** shardy has quit IRC | 07:12 | |
| *** annegent_ has joined #openstack-infra | 07:14 | |
| afazekas | ianw: the simplest algorithm for this kind of allocation to getting the flavor which has the highest error rate | 07:14 |
| afazekas | getting back 0 from something is the highest error | 07:15 |
| *** chianingwang has quit IRC | 07:16 | |
| *** Johnson_Yang has quit IRC | 07:16 | |
| ianw | afazekas: in the worse case, if everyone has the same error and you only have one node, you want to make sure you round-robbin between them. that's why you need to remember who you gave nodes to last time, which is hopefully what my change does | 07:19 |
| *** annegent_ has quit IRC | 07:19 | |
| *** ihrachyshka has joined #openstack-infra | 07:19 | |
| *** chianingwang has joined #openstack-infra | 07:19 | |
| *** camunoz has quit IRC | 07:23 | |
| *** gcha has joined #openstack-infra | 07:23 | |
| *** ihrachyshka has quit IRC | 07:26 | |
| *** talluri has quit IRC | 07:26 | |
| *** talluri has joined #openstack-infra | 07:27 | |
| *** zzelle has joined #openstack-infra | 07:27 | |
| *** ihrachyshka has joined #openstack-infra | 07:28 | |
| *** zzelle has quit IRC | 07:29 | |
| *** talluri has quit IRC | 07:31 | |
| *** ihrachyshka has quit IRC | 07:31 | |
| *** rdopieralski has joined #openstack-infra | 07:32 | |
| *** andreykurilin_ has joined #openstack-infra | 07:33 | |
| *** rkukura has quit IRC | 07:34 | |
| *** rkukura has joined #openstack-infra | 07:36 | |
| *** e0ne has quit IRC | 07:41 | |
| *** _nadya_ has joined #openstack-infra | 07:45 | |
| *** Hal_ has joined #openstack-infra | 07:48 | |
| *** dizquierdo has joined #openstack-infra | 07:54 | |
| *** jpich has joined #openstack-infra | 07:57 | |
| *** zhiyan is now known as zhiyan_ | 08:00 | |
| *** jcoufal has joined #openstack-infra | 08:00 | |
| *** ihrachyshka has joined #openstack-infra | 08:00 | |
| *** alexpilotti has joined #openstack-infra | 08:02 | |
| *** andreykurilin_ has quit IRC | 08:06 | |
| *** IvanBerezovskiy has joined #openstack-infra | 08:10 | |
| *** _nadya_ has quit IRC | 08:12 | |
| *** hashar has joined #openstack-infra | 08:14 | |
| *** annegent_ has joined #openstack-infra | 08:14 | |
| *** rgerganov has joined #openstack-infra | 08:15 | |
| *** hashar has quit IRC | 08:16 | |
| *** zz_johnthetubagu is now known as johnthetubaguy | 08:20 | |
| *** annegent_ has quit IRC | 08:20 | |
| *** CaptTofu_ has joined #openstack-infra | 08:21 | |
| *** _nadya_ has joined #openstack-infra | 08:22 | |
| *** CaptTofu_ has quit IRC | 08:26 | |
| *** e0ne has joined #openstack-infra | 08:26 | |
| *** talluri has joined #openstack-infra | 08:27 | |
| *** penguinRaider has joined #openstack-infra | 08:28 | |
| *** arnaud has quit IRC | 08:28 | |
| *** talluri has quit IRC | 08:30 | |
| *** talluri has joined #openstack-infra | 08:30 | |
| *** persia has quit IRC | 08:36 | |
| *** e0ne has quit IRC | 08:36 | |
| *** e0ne has joined #openstack-infra | 08:39 | |
| *** e0ne_ has joined #openstack-infra | 08:41 | |
| *** trinaths has quit IRC | 08:42 | |
| *** jp_at_hp has joined #openstack-infra | 08:43 | |
| *** persia__ has joined #openstack-infra | 08:43 | |
| *** persia__ has quit IRC | 08:44 | |
| *** persia__ has joined #openstack-infra | 08:44 | |
| *** persia__ is now known as persia | 08:44 | |
| *** e0ne has quit IRC | 08:45 | |
| *** alexpilotti has quit IRC | 08:50 | |
| *** oomichi has quit IRC | 08:53 | |
| *** oomichi has joined #openstack-infra | 08:54 | |
| *** oomichi has quit IRC | 08:55 | |
| *** chandan_kumar has quit IRC | 09:09 | |
| *** ominakov has joined #openstack-infra | 09:12 | |
| *** nati_ueno has quit IRC | 09:13 | |
| *** yamahata_ has quit IRC | 09:13 | |
| *** _nadya_ has quit IRC | 09:14 | |
| *** annegent_ has joined #openstack-infra | 09:14 | |
| *** andreaf_ has joined #openstack-infra | 09:18 | |
| *** annegent_ has quit IRC | 09:19 | |
| *** e0ne_ has quit IRC | 09:20 | |
| *** e0ne has joined #openstack-infra | 09:22 | |
| *** e0ne_ has joined #openstack-infra | 09:23 | |
| *** lcheng has joined #openstack-infra | 09:26 | |
| *** ociuhandu has quit IRC | 09:26 | |
| *** isviridov|away is now known as isviridov | 09:27 | |
| *** davidlenwell has quit IRC | 09:27 | |
| *** e0ne has quit IRC | 09:27 | |
| *** lcheng has quit IRC | 09:28 | |
| *** amotoki has quit IRC | 09:34 | |
| *** jamielennox is now known as jamielennox|away | 09:35 | |
| *** talluri has quit IRC | 09:39 | |
| *** pelix has joined #openstack-infra | 09:42 | |
| *** davidlenwell_ has joined #openstack-infra | 09:52 | |
| *** davidlenwell_ is now known as davidlenwell | 09:52 | |
| *** davidlenwell is now known as davidlenwell_ | 09:52 | |
| *** davidlenwell_ is now known as davidlenwell | 09:52 | |
| *** katyafervent_awa is now known as katyafervent | 09:53 | |
| *** dguerri is now known as _dguerri | 09:53 | |
| *** _dguerri is now known as dguerri | 09:53 | |
| salv-orlando | Hi I remember a recent communication re: not merging changes like https://review.openstack.org/#/c/101112/ because the gate was backed up. | 09:58 |
| salv-orlando | It seems quite quiet right now - but I was wondering if we’d still refrain from merging this kind of patches. | 09:59 |
| salv-orlando | I really have no opinion, because changes like these really dont add neither remove anything, apart from granting free summit tickets. | 09:59 |
| *** habib has joined #openstack-infra | 10:00 | |
| *** yjiang|2 has quit IRC | 10:02 | |
| *** habib has quit IRC | 10:02 | |
| *** habib has joined #openstack-infra | 10:07 | |
| *** ociuhandu has joined #openstack-infra | 10:08 | |
| *** CaptTofu_ has joined #openstack-infra | 10:09 | |
| *** talluri has joined #openstack-infra | 10:10 | |
| *** talluri has quit IRC | 10:12 | |
| *** talluri_ has joined #openstack-infra | 10:12 | |
| *** _nadya_ has joined #openstack-infra | 10:15 | |
| *** CaptTofu_ has quit IRC | 10:15 | |
| *** talluri has joined #openstack-infra | 10:16 | |
| *** talluri_ has quit IRC | 10:16 | |
| *** habib has quit IRC | 10:35 | |
| *** habib_ has joined #openstack-infra | 10:35 | |
| *** katyafervent is now known as katyafervent_awa | 10:35 | |
| *** _nadya_ has quit IRC | 10:36 | |
| *** _nadya_ has joined #openstack-infra | 10:37 | |
| jpich | salv-orlando: From what I understand we're all clear now, http://lists.openstack.org/pipermail/openstack-dev/2014-June/038083.html | 10:44 |
| salv-orlando | ok, thanks | 10:45 |
| *** Longgeek_ has quit IRC | 10:49 | |
| *** flaper87|afk is now known as flaper87 | 10:51 | |
| sdague | salv-orlando: yeh, go for it | 10:52 |
| salv-orlando | thanks sdague. | 10:53 |
| sdague | or more importantly, use whatever judgement you would on the value of the patch | 10:53 |
| sdague | but the gate isn't a concern atm | 10:53 |
| sdague | also, if we merge another 100 patches this week we'll break the record :) | 10:53 |
| *** openstackgerrit has joined #openstack-infra | 10:56 | |
| *** alexpilotti has joined #openstack-infra | 10:57 | |
| *** alexpilotti has quit IRC | 10:58 | |
| SergeyLukjanov | morning folks | 10:58 |
| *** katyafervent_awa is now known as katyafervent | 10:58 | |
| *** Longgeek has joined #openstack-infra | 11:02 | |
| *** dprince has joined #openstack-infra | 11:02 | |
| *** yamahata_ has joined #openstack-infra | 11:02 | |
| *** ociuhandu has quit IRC | 11:03 | |
| *** nosnos has quit IRC | 11:03 | |
| *** habib_ has quit IRC | 11:03 | |
| *** habib_ has joined #openstack-infra | 11:04 | |
| *** dguerri is now known as _dguerri | 11:06 | |
| *** _dguerri is now known as dguerri | 11:07 | |
| *** chianingwang has quit IRC | 11:08 | |
| *** freyes has quit IRC | 11:09 | |
| boris-42 | SergeyLukjanov hi | 11:13 |
| boris-42 | SergeyLukjanov btw are there unit tests for openstack-inrfa/config ? | 11:13 |
| *** freyes has joined #openstack-infra | 11:14 | |
| *** chianingwang has joined #openstack-infra | 11:14 | |
| *** mbacchi has joined #openstack-infra | 11:15 | |
| SergeyLukjanov | boris-42, there are some tests that could be executed using tox | 11:17 |
| boris-42 | SergeyLukjanov not able to find them.. | 11:17 |
| SergeyLukjanov | boris-42, just check different tox envs | 11:18 |
| SergeyLukjanov | boris-42, there are not always unit tests | 11:18 |
| SergeyLukjanov | boris-42, for ex. there is a test that downloads jjb and zuul and check jobs configs | 11:18 |
| boris-42 | SergeyLukjanov can we add just plain py27 tests? | 11:19 |
| boris-42 | SergeyLukjanov unit tests* | 11:19 |
| SergeyLukjanov | boris-42, we always can do it ;) | 11:19 |
| boris-42 | SergeyLukjanov okay I just would like to improve zuul layout | 11:19 |
| boris-42 | SergeyLukjanov adding special tests for it + refactoing a bit | 11:20 |
| *** CaptTofu_ has joined #openstack-infra | 11:20 | |
| SergeyLukjanov | boris-42, if you'd like to change layout.yaml format than zuul is the correct repo for you | 11:20 |
| boris-42 | SergeyLukjanov i just want to make a test that will check that JOBs are used from template if there is such template | 11:21 |
| boris-42 | SergeyLukjanov so should I make such test it in config or zuul? | 11:21 |
| *** rcarrillocruz has quit IRC | 11:25 | |
| *** habib_ has quit IRC | 11:25 | |
| *** habib_ has joined #openstack-infra | 11:29 | |
| *** dprince has quit IRC | 11:29 | |
| *** rcarrillocruz has joined #openstack-infra | 11:29 | |
| mordred | boris-42: there is a zuul layout check already | 11:31 |
| boris-42 | mordred hm and it already checks templates? | 11:31 |
| mordred | boris-42: I do not know if it does everything you want or not | 11:31 |
| boris-42 | mordred so I should write this test in zuul? | 11:32 |
| boris-42 | mordred not in config | 11:32 |
| mordred | just saying, it validates layout.yaml ... so its a good place to start | 11:32 |
| boris-42 | mordred thanks | 11:32 |
| *** sweston has joined #openstack-infra | 11:32 | |
| boris-42 | mordred will do | 11:32 |
| boris-42 | mordred I have idea about forcing everybody to use templates=) | 11:32 |
| mordred | in config, zuul-layout tox env (I think... on phone) | 11:32 |
| mordred | heh. | 11:32 |
| *** yaguang has quit IRC | 11:35 | |
| *** sweston has quit IRC | 11:38 | |
| sdague | man what tz you in now mordred ? :) | 11:39 |
| yolanda | hi, tring to install puppet modules using repos, but i get this error err: /Stage[main]/Openstack_project::Template/Service[rsyslog]: Could not evaluate: undefined method `[]' for nil:NilClass | 11:40 |
| yolanda | any idea about that? | 11:40 |
| *** amalagon has quit IRC | 11:40 | |
| *** flaper87 is now known as flaper87|afk | 11:40 | |
| sdague | yolanda: do you have a longer dump that you can pastebin? | 11:41 |
| yolanda | sdague, i can send you, just a moment | 11:41 |
| yolanda | i'm getting that trying to deploy a devstack-node on dib | 11:42 |
| sdague | ah, gotcha. | 11:42 |
| mordred | sdague: connecting on a redeye through msp | 11:43 |
| *** sweston has joined #openstack-infra | 11:43 | |
| yolanda | sdague http://paste.hpcloud.net/show/190/ | 11:45 |
| sdague | mordred: gotcha | 11:45 |
| sdague | yolanda: can you put it on paste.openstack.org instead, so I don't have to connect to the vpn :) | 11:45 |
| yolanda | ok | 11:45 |
| yolanda | http://paste.openstack.org/show/84567/ | 11:46 |
| mordred | sdague: I'll be in NYC through Wed | 11:53 |
| *** gokrokve_ has quit IRC | 11:54 | |
| *** basha has joined #openstack-infra | 11:54 | |
| yolanda | sdague : http://paste.openstack.org/show/84567/ | 11:54 |
| mordred | yolanda: we need to mark the service refresh of rsyslog to only run if ! $in_chroot | 11:54 |
| openstackgerrit | Sean Dague proposed a change to openstack-dev/hacking: remove H402 https://review.openstack.org/101497 | 11:57 |
| openstackgerrit | Sean Dague proposed a change to openstack-dev/hacking: remove H803, it's mostly an old flamewar https://review.openstack.org/101498 | 11:57 |
| openstackgerrit | Sean Dague proposed a change to openstack-dev/hacking: remove H305 because it's python version specific https://review.openstack.org/101499 | 11:57 |
| sdague | mordred: ah cool, down at the end users summit? | 11:58 |
| mordred | sdague: in theory, although I still don't have the Info's for it | 11:58 |
| mordred | sdague: friend is getting married tomorrow | 11:59 |
| sdague | cool | 11:59 |
| *** basha has quit IRC | 12:01 | |
| *** weshay has joined #openstack-infra | 12:01 | |
| *** dprince has joined #openstack-infra | 12:02 | |
| *** ArxCruz has joined #openstack-infra | 12:02 | |
| *** _nadya_ has quit IRC | 12:03 | |
| *** CaptTofu_ has quit IRC | 12:04 | |
| *** CaptTofu_ has joined #openstack-infra | 12:05 | |
| openstackgerrit | Sean Dague proposed a change to openstack-dev/hacking: remove H803, it's mostly an old flamewar https://review.openstack.org/101498 | 12:08 |
| openstackgerrit | Sean Dague proposed a change to openstack-dev/hacking: remove H305 because it's python version specific https://review.openstack.org/101499 | 12:08 |
| openstackgerrit | Sean Dague proposed a change to openstack-dev/hacking: remove H402 https://review.openstack.org/101497 | 12:08 |
| *** CaptTofu_ has quit IRC | 12:09 | |
| openstackgerrit | Nikita Konovalov proposed a change to openstack-infra/storyboard-webclient: A timeline tag usage fix https://review.openstack.org/101507 | 12:11 |
| *** chianingwang has quit IRC | 12:11 | |
| *** annegent_ has joined #openstack-infra | 12:14 | |
| *** chianingwang has joined #openstack-infra | 12:15 | |
| yolanda | mordred, i can take a look, should i file a patch to openstack-infra/config? | 12:15 |
| *** dkliban_afk is now known as dkliban | 12:16 | |
| *** annegent_ has quit IRC | 12:19 | |
| yolanda | mordred, i see we are exporting a var called FACTER_in_chroot, is that ok to use it for discriminate the rsyslog stuff? | 12:21 |
| *** jaypipes has joined #openstack-infra | 12:21 | |
| *** jaypipes is now known as leakypipes | 12:22 | |
| *** habib_ has quit IRC | 12:23 | |
| *** pdmars has joined #openstack-infra | 12:26 | |
| *** prad has joined #openstack-infra | 12:28 | |
| *** salv-orlando has quit IRC | 12:28 | |
| *** ArxCruz has quit IRC | 12:29 | |
| *** ociuhandu has joined #openstack-infra | 12:30 | |
| *** lcostantino has joined #openstack-infra | 12:31 | |
| *** annegent_ has joined #openstack-infra | 12:32 | |
| *** ArxCruz has joined #openstack-infra | 12:32 | |
| *** gokrokve has joined #openstack-infra | 12:32 | |
| *** jcoufal has quit IRC | 12:35 | |
| *** gokrokve has quit IRC | 12:37 | |
| *** dkehn_ is now known as dkehnx | 12:39 | |
| *** yamahata has quit IRC | 12:39 | |
| *** sbalukoff has left #openstack-infra | 12:41 | |
| *** dizquierdo has quit IRC | 12:41 | |
| ianw | anyone seeing some packet loss issues to git.openstack.org? i've had a couple of jobs fail with issues accessing it, and see some packet loss pinging it on occasion | 12:45 |
| *** lcheng has joined #openstack-infra | 12:45 | |
| ianw | --- git.openstack.org ping statistics --- | 12:45 |
| ianw | 20 packets transmitted, 17 received, 15% packet loss, time 19017ms | 12:45 |
| openstackgerrit | A change was merged to openstack-infra/storyboard-webclient: A timeline tag usage fix https://review.openstack.org/101507 | 12:45 |
| *** CaptTofu_ has joined #openstack-infra | 12:45 | |
| sdague | ianw: I'm not seeing packet loss from here | 12:46 |
| sdague | ianw: maybe tracert to figure out where the packets are getting lost | 12:47 |
| ianw | sdague: yeah, it's at git.openstack.org | 12:48 |
| ianw | 20. core3.dfw1.rackspace.net 0.0% 46 201.3 201.6 198.7 209.3 1.9 | 12:48 |
| ianw | 21. aggr150a-8-core3.dfw2.rackspace.net 0.0% 46 199.9 202.1 198.7 210.3 2.2 | 12:48 |
| ianw | 22. git.openstack.org 2.2% 46 200.2 202.0 196.2 210.6 3.5 | 12:48 |
| ianw | (sorry paste of mtr output) | 12:49 |
| ianw | also i've got jobs failing | 12:49 |
| ianw | 2014-06-20 12:44:09.720 | Cloning into '/opt/stack/python-cinderclient'... | 12:49 |
| ianw | 2014-06-20 12:44:09.724 | fatal: Unable to look up git.openstack.org (port 9418) (Name or service not known) | 12:49 |
| sdague | so that's a different issue, that's a dns fail | 12:50 |
| sdague | is that on a cloud node? or in rh network? | 12:50 |
| yolanda | sdague, i think i have more or less same problem as ianw, when cloning repos from dib, i was getting a problem resolving git.openstack.org | 12:50 |
| ianw | yeah but i'm presuming it's due to packets going missing | 12:50 |
| ianw | sdague: that was from a rh internal network in boston somewhere i think | 12:51 |
| sdague | unable to lookup I'm pretty sure is dns | 12:51 |
| yolanda | i did a workaround just changing to github, because i thought it might be a temporary problem, but there could be something more | 12:51 |
| sdague | but I guess it could be rax having a lossy peering arrangement with your network | 12:51 |
| openstackgerrit | ayoung proposed a change to openstack-infra/config: Add keystonemiddleware project https://review.openstack.org/101406 | 12:52 |
| sdague | this is what my path looks like - http://paste.openstack.org/show/84572/ | 12:52 |
| ianw | i dunno, i do a lot of clones and this is the first time i've seen anything like this | 12:52 |
| sdague | in case there is something useful there | 12:52 |
| sdague | anteaya / krtaylor: the powerkvm ci seems to be posting invalid urls in test results. That should either be fixed, or it should be turned off. | 12:55 |
| *** annegent_ has quit IRC | 12:55 | |
| sdague | noting here - https://review.openstack.org/#/c/101511/ | 12:56 |
| ianw | it seems intermittent, i'm not seeing it now (my path -> http://paste.openstack.org/show/84573/). my thought was something went crazy on the box and killed it | 12:56 |
| ianw | anyway, i'm going to be and crossing my fingers that it just fixes itself :) | 12:56 |
| sdague | :) | 12:56 |
| *** dkranz has joined #openstack-infra | 13:04 | |
| *** rfolco has joined #openstack-infra | 13:06 | |
| *** hashar has joined #openstack-infra | 13:06 | |
| *** eharney has quit IRC | 13:13 | |
| *** annegent_ has joined #openstack-infra | 13:13 | |
| *** annegent_ has quit IRC | 13:14 | |
| *** annegent_ has joined #openstack-infra | 13:20 | |
| *** annegent_ has quit IRC | 13:22 | |
| *** annegent_ has joined #openstack-infra | 13:23 | |
| openstackgerrit | Antoine Musso proposed a change to openstack-infra/jenkins-job-builder: Document the Zuul triggers https://review.openstack.org/77838 | 13:24 |
| openstackgerrit | Antoine Musso proposed a change to openstack-infra/jenkins-job-builder: Document the Zuul triggers https://review.openstack.org/77838 | 13:24 |
| *** dansmith is now known as superdan | 13:24 | |
| *** changbl has quit IRC | 13:25 | |
| dprince | Is it possible to kill this zuul job: https://review.openstack.org/#/c/101378/ | 13:26 |
| dprince | The issue we were chasing there (a Neutron GRE tunnel issue) is now fixed | 13:27 |
| *** talluri has quit IRC | 13:27 | |
| *** talluri has joined #openstack-infra | 13:27 | |
| *** graham__ has quit IRC | 13:27 | |
| *** graham__ has joined #openstack-infra | 13:27 | |
| *** graham__ is now known as mugsie] | 13:28 | |
| *** mugsie] is now known as mugsie | 13:28 | |
| *** bknudson has left #openstack-infra | 13:28 | |
| *** russellb is now known as rustlebee | 13:29 | |
| *** mriedem has joined #openstack-infra | 13:31 | |
| *** talluri has quit IRC | 13:31 | |
| *** gokrokve has joined #openstack-infra | 13:34 | |
| *** flaviof_zzz is now known as flaviof | 13:34 | |
| *** doug-fish has joined #openstack-infra | 13:34 | |
| *** annegent_ has quit IRC | 13:38 | |
| *** gokrokve has quit IRC | 13:39 | |
| *** bknudson has joined #openstack-infra | 13:42 | |
| anteaya | sdague: thanks | 13:49 |
| *** zhiyan_ has quit IRC | 13:50 | |
| *** eharney has joined #openstack-infra | 13:50 | |
| *** sweston has quit IRC | 13:53 | |
| *** gokrokve has joined #openstack-infra | 13:55 | |
| *** zz_gondoi is now known as gondoi | 13:55 | |
| anteaya | sdague: http://lists.openstack.org/pipermail/openstack-infra/2014-June/001407.html | 13:56 |
| *** sweston has joined #openstack-infra | 13:56 | |
| *** mkerrin has quit IRC | 13:56 | |
| *** pdmars has quit IRC | 13:57 | |
| *** mkerrin has joined #openstack-infra | 13:59 | |
| *** alaski is now known as lascii | 13:59 | |
| *** gokrokve has quit IRC | 14:00 | |
| *** pdmars has joined #openstack-infra | 14:00 | |
| *** ramashri has joined #openstack-infra | 14:02 | |
| *** mfer has joined #openstack-infra | 14:02 | |
| *** gokrokve has joined #openstack-infra | 14:02 | |
| krtaylor | anteaya, got your email, working on it now | 14:04 |
| *** comstud is now known as bearhands | 14:04 | |
| ArxCruz | anteaya: it's working now, our swift script isn't perfect yet. So, when it fails on stack.sh the log files isn't created :( | 14:05 |
| *** ramashri has quit IRC | 14:06 | |
| *** radez_g0n3 is now known as radez | 14:08 | |
| *** chianingwang has quit IRC | 14:09 | |
| *** ramashri has joined #openstack-infra | 14:10 | |
| *** chianingwang has joined #openstack-infra | 14:14 | |
| anteaya | krtaylor ArxCruz thanks for the quick response | 14:15 |
| anteaya | :( | 14:15 |
| *** zehicle_at_dell has quit IRC | 14:16 | |
| ArxCruz | anteaya: yeah, yesterday was holiday here, so I wasn't monitoring the jobs :/ | 14:16 |
| anteaya | is there a way that you can be alerted if the swift script fails on stack.sh? | 14:16 |
| anteaya | ah | 14:16 |
| anteaya | can you reply to the email thread as well please? | 14:16 |
| ArxCruz | anteaya: this week I will start to test the zuul swift upload | 14:16 |
| ArxCruz | anteaya: already did it | 14:16 |
| anteaya | ArxCruz: thanks | 14:16 |
| anteaya | ArxCruz: thank you | 14:16 |
| anteaya | I love working with you two | 14:16 |
| anteaya | :) | 14:16 |
| ArxCruz | :) | 14:17 |
| ArxCruz | <3 | 14:17 |
| *** annegent_ has joined #openstack-infra | 14:19 | |
| anteaya | krtaylor: you said you wanted to make some changes to the powerkvm wikipage | 14:19 |
| anteaya | have you had a chance to do that yet? | 14:20 |
| *** annegent_ has quit IRC | 14:20 | |
| *** Alexei_987 has quit IRC | 14:20 | |
| *** wenlock has joined #openstack-infra | 14:20 | |
| openstackgerrit | yolanda.robla proposed a change to openstack-infra/config: Only enabling rsyslog service if not chroot https://review.openstack.org/101562 | 14:22 |
| *** annegent_ has joined #openstack-infra | 14:24 | |
| *** rdopieralski has quit IRC | 14:24 | |
| *** Hal_ has quit IRC | 14:25 | |
| *** Abhishe__ has joined #openstack-infra | 14:25 | |
| *** david-lyle has joined #openstack-infra | 14:29 | |
| *** Abhishe__ has quit IRC | 14:30 | |
| *** dkehn_ has joined #openstack-infra | 14:31 | |
| *** dkehn has quit IRC | 14:31 | |
| *** dkehnx has quit IRC | 14:32 | |
| *** rcarrillocruz has quit IRC | 14:32 | |
| *** mrodden has joined #openstack-infra | 14:32 | |
| *** Hal_ has joined #openstack-infra | 14:32 | |
| *** yamahata has joined #openstack-infra | 14:33 | |
| *** dkehn has joined #openstack-infra | 14:33 | |
| *** lcostantino has quit IRC | 14:35 | |
| openstackgerrit | yolanda.robla proposed a change to openstack-infra/config: Only enabling rsyslog service if not chroot https://review.openstack.org/101562 | 14:35 |
| *** rwsu has joined #openstack-infra | 14:35 | |
| *** rcarrillocruz has joined #openstack-infra | 14:37 | |
| *** Hal_ has quit IRC | 14:37 | |
| *** nati_ueno has joined #openstack-infra | 14:37 | |
| *** Hal_ has joined #openstack-infra | 14:37 | |
| *** sabeen has joined #openstack-infra | 14:38 | |
| *** gokrokve has quit IRC | 14:38 | |
| mfer | do the trivial rebase checks still work? | 14:39 |
| mfer | I just ran into a case where that didn't happen and the code wasn't merged | 14:39 |
| morganfainberg | anteaya, if you have a few moments this morning could you take a gander at https://review.openstack.org/#/c/101406/ see if I'm missing anything (before these are looked at today) - i want to get any issues smushed early on | 14:39 |
| morganfainberg | anteaya, if not, no worries :) | 14:39 |
| *** YorikSar has quit IRC | 14:39 | |
| *** lcostantino has joined #openstack-infra | 14:40 | |
| *** radez is now known as radez_g0n3 | 14:42 | |
| yolanda | hi, having a problem with gate-config-puppet-apply-precise tests: http://logs.openstack.org/62/101562/1/check/gate-config-puppet-apply-precise/0fafe66/console.html | 14:42 |
| *** annegent_ has quit IRC | 14:42 | |
| *** adrian_otto has joined #openstack-infra | 14:43 | |
| *** dizquierdo has joined #openstack-infra | 14:43 | |
| krtaylor | anteaya, not yet, I'll look at that today | 14:44 |
| openstackgerrit | Thierry Carrez proposed a change to openstack-infra/release-tools: Add spec2bp.py script https://review.openstack.org/101566 | 14:45 |
| * anteaya clicks | 14:46 | |
| anteaya | krtaylor: thanks | 14:46 |
| *** dkehn_ is now known as dkehnx | 14:50 | |
| *** annegent_ has joined #openstack-infra | 14:50 | |
| *** cody-somerville has quit IRC | 14:51 | |
| anteaya | morganfainberg: why do you need refs/for/refs | 14:53 |
| anteaya | morganfainberg: I don't know as I have seen it before | 14:53 |
| *** cody-somerville has joined #openstack-infra | 14:54 | |
| morganfainberg | anteaya, keystoneclient has it? :P i figured it should have the same permissions as ksc across the board | 14:54 |
| *** habib_ has joined #openstack-infra | 14:54 | |
| morganfainberg | anteaya, i can remove it, but since this is meant to work just like keystoneclient i was hesitent to remove anything from the ACL file. | 14:54 |
| openstackgerrit | Sergey Skripnick proposed a change to openstack-infra/config: Add rally devstack integration to rally-cli job https://review.openstack.org/101570 | 14:55 |
| *** annegent_ has quit IRC | 14:55 | |
| *** mmaglana has joined #openstack-infra | 14:55 | |
| *** chandan_kumar has joined #openstack-infra | 14:55 | |
| *** atiwari has joined #openstack-infra | 14:57 | |
| anteaya | morganfainberg: just asking questions right now, don't do anything yet | 14:58 |
| *** radez_g0n3 is now known as radez | 14:58 | |
| anteaya | morganfainberg: I had just never seen it before, I don't know what it does | 14:58 |
| morganfainberg | anteaya, absolutely :) | 14:58 |
| anteaya | thanks | 14:58 |
| *** salv-orlando has joined #openstack-infra | 14:58 | |
| morganfainberg | anteaya, i think it's for feature branch related stuff. but honestly, i forget what refs/for/refs is | 14:59 |
| anteaya | morganfainberg: and how much do you use milestones on keystoneclient | 14:59 |
| *** Hal_ has quit IRC | 14:59 | |
| morganfainberg | anteaya, I don't think dolph uses milestone-proposed ever | 14:59 |
| *** annegent_ has joined #openstack-infra | 14:59 | |
| morganfainberg | we just tag a new release when ready | 14:59 |
| anteaya | tags I can see, but milestones as far as I know, are only of use for a repo that goes through our release cycle | 14:59 |
| anteaya | right | 15:00 |
| anteaya | so that seems kind of useless to have there | 15:00 |
| anteaya | keep the tags, lose the milestone proposed | 15:00 |
| anteaya | and I will figure out refs for refs, or try to | 15:00 |
| morganfainberg | anteaya, in theory it might be nice to have though, i could see a case where we don't want to increment the major version but need a patch (Sec fix) | 15:00 |
| morganfainberg | anteaya, especially since we're going to be releasing this as the first stable (1.0.0) version of the middleware | 15:01 |
| morganfainberg | anteaya, not really arguing we need it, just thinking outloud | 15:01 |
| anteaya | sure | 15:02 |
| anteaya | we could keep all the things if you want | 15:03 |
| anteaya | I'm just looking at folks who cargo cult other acl files | 15:03 |
| anteaya | they don't know what anything does and just jam everything into a file and hope for the best | 15:03 |
| openstackgerrit | Thierry Carrez proposed a change to openstack-infra/release-tools: Fix wait_for_tarball.py reporting whl artifacts https://review.openstack.org/101573 | 15:03 |
| anteaya | the best acl file is one that does exactly what you need, no more and no less | 15:03 |
| anteaya | and if you need a permission on a repo that you don't have, offer a patch to the acl file | 15:04 |
| anteaya | when you find you need that permission | 15:04 |
| morganfainberg | anteaya, yep. i'm fine with keep or leave | 15:05 |
| *** thedodd has joined #openstack-infra | 15:05 | |
| anteaya | morganfainberg: so have a chat with others in keystone and let me know what the consenses is | 15:05 |
| morganfainberg | anteaya, will do. | 15:05 |
| anteaya | thanks | 15:06 |
| ArxCruz | jhesketh: around? I will start to test zuul swift upload in my CI, may I bother you (a lot!) since you're wrote the patch? :D | 15:07 |
| anteaya | ArxCruz: he is in au | 15:07 |
| anteaya | so asleep right now | 15:07 |
| anteaya | or I imagine he is | 15:07 |
| *** habib_ has quit IRC | 15:07 | |
| * ArxCruz think should have a universal time for developers | 15:08 | |
| *** habib_ has joined #openstack-infra | 15:08 | |
| krtaylor | ArxCruz, it is all the time :) | 15:09 |
| ArxCruz | krtaylor: except for the Beer time :) | 15:09 |
| *** Longgeek has quit IRC | 15:09 | |
| *** W00die has joined #openstack-infra | 15:10 | |
| *** chianingwang has quit IRC | 15:11 | |
| sdague | beer time is the best time | 15:12 |
| *** homeless has joined #openstack-infra | 15:13 | |
| *** Hal_ has joined #openstack-infra | 15:13 | |
| mordred | morganfainberg: yup, no need for m-p | 15:13 |
| zaro | morning | 15:14 |
| mordred | zaro: morning! | 15:14 |
| *** adrian_otto has quit IRC | 15:15 | |
| morganfainberg | mordred, zaro, in gerrit any idea what refs/for/refs is? | 15:15 |
| mordred | yup | 15:16 |
| * zaro would really like to know too | 15:16 | |
| mordred | that's how things get submitted by git review | 15:16 |
| *** chianingwang has joined #openstack-infra | 15:17 | |
| mordred | when you upload changes, you're pushing to refs/for/master | 15:17 |
| *** lcostantino has quit IRC | 15:17 | |
| morganfainberg | mordred, so do i need pushMerge on refs/for/refs for keystonemiddleware? | 15:17 |
| mordred | now. refs/for/refs in the acls is deep magic ... | 15:18 |
| mordred | no | 15:18 |
| morganfainberg | yeah looking at this: https://review.openstack.org/#/c/101406/5/modules/openstack_project/files/gerrit/acls/openstack/keystonemiddleware.config | 15:18 |
| * mordred looks | 15:18 | |
| morganfainberg | getting all the bits in place so we can get rid of silly dependencies in keystoneclient | 15:18 |
| morganfainberg | :) | 15:18 |
| morganfainberg | mordred, that was originally a copy of keystoneclient's because i want to ensure i wasn't breaking "functionality" as the release / interaction is meant to be the same as ksc (for the most part) | 15:19 |
| zaro | mordred: what? refs/for/refs is for git review? or is it deep magic? | 15:19 |
| morganfainberg | but if refs/for/refs isn't needed, i'll get that removed as well as the m-p (unless it turns out we need it in the future) | 15:20 |
| mordred | morganfainberg: yah. you can lose it | 15:20 |
| anteaya | zaro: this is what i have found so far: https://gerrit-documentation.googlecode.com/svn/Documentation/2.3/access-control.html#category_push | 15:20 |
| morganfainberg | which case i'll add m-p back in then | 15:20 |
| *** johnthetubaguy is now known as zz_johnthetubagu | 15:21 | |
| mordred | zaro: both ... | 15:21 |
| zaro | morganfainberg: it's there by default when i install a fresh gerrit. so it must be for something. | 15:21 |
| morganfainberg | mordred, ok cool. | 15:21 |
| zaro | no way, i don't think you can remove refs/for/refs | 15:22 |
| clarkb | o/ | 15:23 |
| anteaya | hey clarkb | 15:23 |
| anteaya | healthy again yet? | 15:23 |
| clarkb | mordred: jeblair sdague everyone else: should we go ahead and trusty now? | 15:24 |
| clarkb | anteaya: yup for the most part | 15:24 |
| anteaya | good | 15:24 |
| anteaya | I saw something in backscroll about trusting affecting tripleo | 15:24 |
| anteaya | I think sdague commented on that | 15:25 |
| clarkb | ya I updated the patch to fix that | 15:25 |
| anteaya | kk | 15:25 |
| *** Woodie has joined #openstack-infra | 15:25 | |
| morganfainberg | zaro, hm. | 15:25 |
| morganfainberg | zaro, i'm happy to leave it in. | 15:25 |
| anteaya | I think that would be a poor idea | 15:26 |
| anteaya | since other projects will copy that behaviour | 15:26 |
| *** zz_johnthetubagu is now known as johnthetubaguy | 15:26 | |
| anteaya | let's find out more about what it does | 15:26 |
| morganfainberg | i think this is meant so we could review mergepushes | 15:26 |
| morganfainberg | which... nah | 15:26 |
| morganfainberg | this is adding pushMerge permission not anything else | 15:27 |
| anteaya | deep magic doesn't have enough details for me | 15:27 |
| morganfainberg | i bet we can remove this safely | 15:27 |
| sdague | clarkb: sure | 15:27 |
| zaro | morganfainberg: ohh, i just looked at the keystonemiddleware.config. in your case you probably don't need it because it will get inherited from the parent all-projects | 15:27 |
| morganfainberg | zaro, ++ yeah | 15:27 |
| anteaya | morganfainberg: like I said before, none of the other acl files I have looked at have it | 15:27 |
| *** johnthetubaguy is now known as zz_johnthetubagu | 15:27 | |
| *** habib_ has quit IRC | 15:27 | |
| morganfainberg | this might be vestifgial in the ksc one. | 15:28 |
| anteaya | zaro: yes, my read is that refs/for/refs needs to be somewhere, just not in individual acl files | 15:28 |
| anteaya | morganfainberg: I can see that | 15:28 |
| sdague | clarkb: I ran a few more experimental runs, we're hitting both hp cloud and rax, and all the trusty runs have passed so far | 15:28 |
| clarkb | sdague: cool | 15:28 |
| clarkb | mordred: I think I will go ahead and approve the first chane in my stack then | 15:28 |
| *** zz_johnthetubagu is now known as johnthetubaguy | 15:28 | |
| zaro | morganfainberg, anteaya : gerrit adds it to all-projects by default | 15:28 |
| *** Hal_ has quit IRC | 15:29 | |
| *** W00die has quit IRC | 15:29 | |
| clarkb | and done | 15:30 |
| anteaya | zaro: cool, so we don't need it in individual acl files | 15:30 |
| openstackgerrit | David Caro proposed a change to openstack-infra/jenkins-job-builder: Use yaml local tags to support including files https://review.openstack.org/48783 | 15:30 |
| openstackgerrit | David Caro proposed a change to openstack-infra/jenkins-job-builder: Added possibility to use non-existent keys https://review.openstack.org/101580 | 15:30 |
| anteaya | morganfainberg: can you offer a patch removing it from keystoneclient as well? | 15:30 |
| mordred | clarkb: ++ | 15:30 |
| morganfainberg | anteaya, will do | 15:30 |
| anteaya | morganfainberg: thanks | 15:30 |
| morganfainberg | anteaya, mordred, talking over with keystonefolks, i'm going to add in the tempest gate runs so we don't need another patch for the first projects converted over to this. there should be minimal changes between the creation of the repo and the initial releasing. | 15:31 |
| *** dizquierdo has quit IRC | 15:31 | |
| morganfainberg | or.. strike that last comment | 15:31 |
| morganfainberg | :P | 15:32 |
| morganfainberg | neeeeevverrrrrmind we'll add that as soon as we're ready to cut over. maybe an alpha release. | 15:32 |
| anteaya | I don't parse your second to last comment at all | 15:32 |
| anteaya | but that is okay | 15:32 |
| *** talluri has joined #openstack-infra | 15:32 | |
| *** adrian_otto has joined #openstack-infra | 15:32 | |
| anteaya | I will review the patch(es) | 15:32 |
| morganfainberg | anteaya, thanks for your help! | 15:33 |
| *** johnthetubaguy is now known as zz_johnthetubagu | 15:34 | |
| anteaya | np | 15:34 |
| clarkb | sdague: I even did my best to have an early start this morning :) | 15:34 |
| clarkb | sdague: I want to get this done, then I have to go buy brisket and beer makings | 15:34 |
| clarkb | clearly not selfish at all | 15:34 |
| sdague | heh | 15:35 |
| openstackgerrit | Morgan Fainberg proposed a change to openstack-infra/config: Add keystonemiddleware project https://review.openstack.org/101406 | 15:36 |
| sdague | yeh, I'd say start flushing things through. At some point here I'm going to break for lunch, but then this is primary focus once I get back. | 15:36 |
| rcarrillocruz | hmm | 15:36 |
| rcarrillocruz | guys , question about jeepyb/update_bug | 15:36 |
| clarkb | sdague: yup I will get through them one at a time as they apply | 15:36 |
| clarkb | should go reasonably quickly | 15:36 |
| sdague | cool | 15:36 |
| rcarrillocruz | when process_bugtask is called, it goes thru all the projects affected by the bug and return a collection of what's called 'bug_task' | 15:37 |
| rcarrillocruz | however, can i assume it will ALWAYS return 1 'bug_task' per 'bug'? | 15:37 |
| openstackgerrit | A change was merged to openstack-infra/config: Explicitly set node label for every zuul job. https://review.openstack.org/101283 | 15:37 |
| rcarrillocruz | the 'process_bugtask' does this check: | 15:37 |
| rcarrillocruz | if lp_task.bug_target_name == project: | 15:37 |
| clarkb | rcarrillocruz: no, because launchpad allows multiple project bugtasks per bug | 15:38 |
| rcarrillocruz | i.e. i don't expect a bug having several bug_task referencing the same project (let's say 'openstack-infra' or 'tripleo') | 15:38 |
| rcarrillocruz | what's the use case here? why a bug would show in the 'Affects' section more than one entry for the same project? | 15:39 |
| clarkb | for launchpad I believe they are unique proejcts for each task | 15:39 |
| clarkb | but in storyboard they won't be | 15:40 |
| rcarrillocruz | interesting | 15:40 |
| clarkb | the use case is we haev a bug that affects all of openstack. We need to fix it in each individual project. So we have tasks for each project | 15:40 |
| *** ominakov has quit IRC | 15:41 | |
| anteaya | morganfainberg: +1 | 15:41 |
| *** gokrokve has joined #openstack-infra | 15:41 | |
| *** e0ne_ has quit IRC | 15:41 | |
| morganfainberg | anteaya, tyvm! | 15:41 |
| anteaya | np | 15:41 |
| rcarrillocruz | i'm asking cos i'm refactoring the jeepyb/update_bug code to be issue tracker independent , and the methods of the base class I need them to cover these kind of things... | 15:41 |
| anteaya | I learned about refs/for/refs | 15:41 |
| clarkb | rcarrillocruz: ya | 15:41 |
| morganfainberg | anteaya, i get to smack a bunch of reviews with a minus -2 once this repo is made *even grin*, and very politely tell them to use the new repo cause thats where new code goes | 15:42 |
| anteaya | ah the power of the -2 | 15:45 |
| *** annegent_ has quit IRC | 15:45 | |
| *** yjiang5_away is now known as yjiang5 | 15:45 | |
| *** changbl has joined #openstack-infra | 15:46 | |
| openstackgerrit | Dean Troyer proposed a change to openstack/requirements: Bump up minimum cliff version to 1.6.0 https://review.openstack.org/101588 | 15:48 |
| *** annegent_ has joined #openstack-infra | 15:49 | |
| *** freyes has quit IRC | 15:50 | |
| anteaya | dtroyer: I know clarkb is busy and fungi is afk and jeblair isn't here yet, and mordred - well mordred is travelling, but can we get a log of what you need to see in a server to transfer the devstack.org documentation under infra? | 15:51 |
| anteaya | dtroyer: if you are around, that is | 15:51 |
| sdague | clarkb: ok, I'm headed out to lunch now. Will check in once I get back to make sure nothing's super on fire | 15:53 |
| clarkb | ok | 15:53 |
| sdague | but I expect this to go quite smooth, at least on the integrated side, based on what I've seen so far | 15:53 |
| clarkb | I am still waiting on the first puppet run zuul | 15:54 |
| *** changbl has quit IRC | 15:54 | |
| clarkb | once thatis done I will update jenkins then we can move tempest | 15:54 |
| clarkb | ya me too | 15:54 |
| jeblair | clarkb: morning; need action from me? | 15:54 |
| *** skolekonov has quit IRC | 15:54 | |
| clarkb | jeblair: nothing yet. I approved and merged the first in my stack | 15:54 |
| *** zz_johnthetubagu is now known as johnthetubaguy | 15:54 | |
| jeblair | k | 15:54 |
| clarkb | mordred and jhesketh both +2'd | 15:54 |
| clarkb | jeblair: currently waiting on puppet to update the zuul layout and function | 15:54 |
| clarkb | nope looks like it just updated | 15:54 |
| clarkb | and a bunch of tempest jobs jst launched on what appaer to be the correct nodes | 15:55 |
| clarkb | jeblair: did you want to inspect things at this step before moving onto updating jenkins? | 15:56 |
| clarkb | I figured I would wait for py33 and py26 jobs to launch | 15:56 |
| openstackgerrit | Morgan Fainberg proposed a change to openstack-infra/config: Remove pushMerge from keystoneclient refs/for/refs acl https://review.openstack.org/101591 | 15:58 |
| morganfainberg | anteaya, ^ removing pushMerge for the refs/for/refs for ksc | 15:58 |
| *** mriedem has quit IRC | 15:58 | |
| morganfainberg | anteaya, as requested | 15:59 |
| clarkb | jeblair: the solum jobs that just launched lgtm | 15:59 |
| clarkb | I think I am ready to approve change 2 | 15:59 |
| *** gondoi is now known as zz_gondoi | 15:59 | |
| anteaya | morganfainberg: +1 | 16:00 |
| anteaya | thanks | 16:00 |
| jeblair | clarkb: i agree | 16:00 |
| clarkb | approved | 16:00 |
| *** annegent_ has quit IRC | 16:01 | |
| *** vponomaryov has quit IRC | 16:01 | |
| *** johnthetubaguy is now known as mr_oom-pah | 16:03 | |
| openstackgerrit | A change was merged to openstack-infra/config: Update jenkins jobs to optionally allow trusty https://review.openstack.org/101284 | 16:04 |
| *** mriedem has joined #openstack-infra | 16:05 | |
| dtroyer | anteaya: sure, is very simple: serve static HTML and a way to upload the files from a script. It's currently a GitHub gh-pages site, which is just a branch in the devstack repo. | 16:05 |
| openstackgerrit | Antoine Musso proposed a change to openstack-infra/zuul: cloner to easily clone dependent repositories https://review.openstack.org/70373 | 16:06 |
| anteaya | dtroyer: thanks | 16:07 |
| anteaya | dtroyer: at the summit you had mentioned something about preserving the vertical split layout | 16:07 |
| *** changbl has joined #openstack-infra | 16:07 | |
| anteaya | is the vertical split captured in the static HTML? | 16:07 |
| dtroyer | yup | 16:08 |
| anteaya | great | 16:08 |
| anteaya | so sounds like a server with apache on it | 16:08 |
| *** jlibosva has quit IRC | 16:08 | |
| anteaya | with an upload script, possibly some sort of puppet manifest | 16:08 |
| *** lukego has quit IRC | 16:11 | |
| *** lukego has joined #openstack-infra | 16:11 | |
| dtroyer | if I understand how the sphinx-generated docs work in the python projects, this should be very similar. about 80% of the static html is generated by a script that is the functional equivalent of sphinx | 16:12 |
| dtroyer | that's a bit of a step up though, it's all done manually today | 16:12 |
| anteaya | hmmm | 16:14 |
| anteaya | what about our jenkins jobs to generate sphnix docs | 16:14 |
| anteaya | oh sorry that is what you just said | 16:14 |
| *** arnaud__ has joined #openstack-infra | 16:14 | |
| anteaya | do you have a list of what steps you take manually? | 16:15 |
| *** jogo is now known as flashgordon | 16:15 | |
| anteaya | regarding refs/for/refs: http://paste.openstack.org/show/84585/ | 16:16 |
| anteaya | it looks to me like lines 2-6 need to keep it | 16:16 |
| *** rfolco has quit IRC | 16:17 | |
| anteaya | and the acl files on lines 7 to 14 should have it removed from their acl files | 16:17 |
| *** ildikov has quit IRC | 16:18 | |
| yolanda | here, question about the mid-cycle summit, is there any preferred hotel to stay? | 16:20 |
| *** talluri has quit IRC | 16:21 | |
| clarkb | yolanda: on the wiki page the maritim konferenz hotel was listed and it is an HP approved place so I picked that one | 16:21 |
| clarkb | but I don't think there was a room block set up | 16:21 |
| *** talluri has joined #openstack-infra | 16:21 | |
| yolanda | clarkb, thx, i'll take a look | 16:21 |
| dtroyer | anteaya: it's all generated by the tools/build_docs.sh script in the devstack repo. the glue we will need is getting those generated files to the web server, I'm presuming a process similar to what is used for the sphinx docs will work | 16:23 |
| anteaya | dtroyer: what part is manual right now? | 16:24 |
| anteaya | invoking build_docs? | 16:24 |
| clarkb | huh puppet has run on all of the jenkins nodes but the label expression for check-tempest-dsvm-full doesn't seem tp have updated. I will debug by hand now | 16:24 |
| dtroyer | yes | 16:24 |
| anteaya | ah | 16:24 |
| anteaya | puppet can do that | 16:24 |
| clarkb | oh the update is still running | 16:24 |
| *** lcheng has quit IRC | 16:24 | |
| anteaya | dtroyer: so we need a server with apache and a puppet manifest that invokes build_docs.sh | 16:25 |
| anteaya | dtroyer: can you think of anything else? | 16:25 |
| *** talluri has quit IRC | 16:26 | |
| dtroyer | at that level, no. it's all details from there | 16:27 |
| *** lcheng has joined #openstack-infra | 16:28 | |
| *** jpich has quit IRC | 16:31 | |
| *** BobBall is now known as BobBall_Away | 16:31 | |
| *** gyee has joined #openstack-infra | 16:31 | |
| *** sarob has joined #openstack-infra | 16:32 | |
| clarkb | jenkins02 had a bad jjb cache | 16:32 |
| clarkb | I cleaned that up and it is adding jobs now | 16:33 |
| *** ominakov has joined #openstack-infra | 16:34 | |
| *** ihrachyshka has quit IRC | 16:35 | |
| *** nati_ueno has quit IRC | 16:36 | |
| anteaya | dtroyer: awesome thank you | 16:36 |
| clarkb | we have a lot of jobs | 16:36 |
| *** sarob has quit IRC | 16:39 | |
| *** marcoemorais has joined #openstack-infra | 16:42 | |
| *** radez is now known as radez_g0n3 | 16:43 | |
| openstackgerrit | Aaron Greengrass proposed a change to openstack-infra/config: Convert static log cleanup to template, set retention, enable force. https://review.openstack.org/101603 | 16:44 |
| *** hashar has quit IRC | 16:44 | |
| * anteaya heads out for a walk | 16:44 | |
| openstackgerrit | Aaron Greengrass proposed a change to openstack-infra/config: Convert static log cleanup to template, enable gzip force. https://review.openstack.org/101603 | 16:44 |
| jeblair | clarkb: that reminds me, since we don't re-use workspaces anymore, i think we can consider reducing our use of jjb templates and just have a "gate-python27" job that uses the ZUUL_PROJECT parameter | 16:44 |
| jeblair | clarkb: that will reduce our job count | 16:44 |
| *** Ryan_Lane has joined #openstack-infra | 16:45 | |
| jeblair | which would be nice though i guess we won't win any contests at jenkins conferences | 16:45 |
| openstackgerrit | David Caro proposed a change to openstack-infra/jenkins-job-builder: Added parallelization options https://review.openstack.org/75514 | 16:46 |
| clarkb | jeblair: oh good point | 16:46 |
| *** annegent_ has joined #openstack-infra | 16:47 | |
| clarkb | in any case we shuold be mostly up to date on all jobs in a little bit here | 16:48 |
| *** ociuhandu has quit IRC | 16:50 | |
| *** yjiang5 is now known as yjiang5_afk | 16:50 | |
| *** praneshp has joined #openstack-infra | 16:52 | |
| *** annegent_ has quit IRC | 16:52 | |
| dprince | jeblair: is it possible to manually kill a zuul job https://review.openstack.org/#/c/101378/ | 16:53 |
| *** ociuhandu has joined #openstack-infra | 16:53 | |
| *** yjiang5_afk is now known as yjiang5 | 16:53 | |
| *** mestery has joined #openstack-infra | 16:53 | |
| *** salv-orlando has quit IRC | 16:54 | |
| clarkb | jenkins04 is the only jenkins that is still updating | 16:55 |
| clarkb | dprince: hrm | 16:55 |
| clarkb | dprince: I am going to manually stop the in progress job and see what happens | 16:56 |
| dprince | clarkb: cool | 16:56 |
| clarkb | did it work? | 16:56 |
| clarkb | looks like it | 16:56 |
| dprince | clarkb: yep, I think so | 16:57 |
| clarkb | and jenkins04 is done | 16:57 |
| dprince | clarkb: thanks | 16:57 |
| clarkb | going to spot check the label values across the servers | 16:57 |
| clarkb | jeblair: but then I am going to approve the change to make tempest run on trusty | 16:57 |
| clarkb | dprince: np | 16:57 |
| *** torgomatic has left #openstack-infra | 16:58 | |
| *** marcoemorais has quit IRC | 16:59 | |
| clarkb | ok jobs look good. | 16:59 |
| clarkb | trusty here we come | 16:59 |
| *** andreykurilin_ has joined #openstack-infra | 16:59 | |
| *** mestery has quit IRC | 16:59 | |
| jeblair | clarkb: neato! | 16:59 |
| *** marcoemorais has joined #openstack-infra | 17:00 | |
| *** mestery has joined #openstack-infra | 17:00 | |
| *** harlowja_away is now known as harlowja | 17:00 | |
| *** mr_oom-pah is now known as johnthetubaguy | 17:01 | |
| morganfainberg | clarkb, yay trusty! | 17:01 |
| *** ihrachyshka has joined #openstack-infra | 17:01 | |
| openstackgerrit | A change was merged to openstack-infra/config: Run all dsvm jobs from juno onwards on trusty https://review.openstack.org/101285 | 17:03 |
| *** jp_at_hp has quit IRC | 17:04 | |
| *** mestery has quit IRC | 17:04 | |
| clarkb | and now we wait for puppet | 17:04 |
| morganfainberg | clarkb, ooh that means i can fix the log format for keystone! | 17:04 |
| clarkb | oh? apache version difference? | 17:05 |
| morganfainberg | clarkb, yep! | 17:05 |
| morganfainberg | clarkb, can now specify log format for errorlog | 17:05 |
| clarkb | woot | 17:05 |
| morganfainberg | clarkb, glad this landed before i finished the devstack and grenade changes for getting keystone under mod_wsgi | 17:05 |
| morganfainberg | (hope to have those up today) | 17:05 |
| morganfainberg | today might be a big day for us keystone folks :) lots of good stuff | 17:06 |
| jesusaurus | where does test-matrix.py live? what project can I find that in? | 17:06 |
| jeblair | jesusaurus: devstack-gate | 17:06 |
| morganfainberg | jesusaurus, devstack-gate | 17:06 |
| clarkb | jesusaurus: devstack-gate | 17:06 |
| jesusaurus | thanks :) | 17:06 |
| *** homeless has quit IRC | 17:07 | |
| *** dguerri is now known as _dguerri | 17:08 | |
| *** ildikov has joined #openstack-infra | 17:09 | |
| *** Abhishek has joined #openstack-infra | 17:09 | |
| *** ihrachyshka has quit IRC | 17:15 | |
| *** yolanda has quit IRC | 17:17 | |
| *** nati_ueno has joined #openstack-infra | 17:17 | |
| *** _nadya_ has joined #openstack-infra | 17:18 | |
| *** Abhishek has quit IRC | 17:19 | |
| *** rcarrill` has joined #openstack-infra | 17:19 | |
| clarkb | ok we should be trustying from now on | 17:20 |
| *** rcarrillocruz has quit IRC | 17:20 | |
| clarkb | 97014,2 is our guinea pig | 17:20 |
| clarkb | hrm doesn't seem to be working | 17:21 |
| clarkb | we are building trusty nodes though so it seems like some demand is there | 17:22 |
| clarkb | and a couple switched to used | 17:22 |
| clarkb | its like trusty is used but not exclusively | 17:24 |
| clarkb | jeblair: ^ | 17:24 |
| *** eharney has quit IRC | 17:25 | |
| *** markmc has quit IRC | 17:26 | |
| clarkb | doing tail -f /var/log/zuul/debug.log | grep devstack-trusty shows that the jobs seem to be submitted correctly | 17:26 |
| *** ihrachyshka has joined #openstack-infra | 17:27 | |
| clarkb | but I am seeing trusty and precise mixed for changes | 17:27 |
| *** pdmars has quit IRC | 17:28 | |
| *** pdmars has joined #openstack-infra | 17:29 | |
| clarkb | it is almost like the jenkins side isn't honoring te node specification | 17:29 |
| sdague | clarkb: so far so good on the tests side? | 17:30 |
| clarkb | I am getting precise and trusty on the same jenkins server so not a mix jenkins server madness | 17:30 |
| clarkb | sdague: ya I haven't seen the tests complain yet | 17:30 |
| clarkb | but something is clearly not right in scheduling | 17:31 |
| clarkb | zaro: ^ | 17:31 |
| zaro | clarkb: did you apply a custom node label to trusty? have you check to see if the correct gearman functions are registered? | 17:34 |
| clarkb | zaro: yes and yes | 17:36 |
| clarkb | zaro: the jobs are running | 17:36 |
| clarkb | the problem is they do not appear to respect the label specific job name | 17:36 |
| *** changbl has quit IRC | 17:36 | |
| clarkb | zaro: so we are running build:jobname:label1 but its running on label2 | 17:37 |
| clarkb | zaro: the label on the jenkins side is label1 || label2 so that the jobs get registered | 17:37 |
| clarkb | then we want zuul to select the build node | 17:37 |
| clarkb | it could be an issue in zuul selecting the build node or on the jenkins side. I am trying to debug that now | 17:37 |
| clarkb | https://jenkins02.openstack.org/job/check-tempest-dsvm-neutron/20501/ should have run on devstack-trusty not devstack-precise for example | 17:40 |
| clarkb | I am going to use that as my debug job | 17:40 |
| openstackgerrit | Matthew Treinish proposed a change to openstack-infra/elastic-recheck: Add query for bug 1332512 https://review.openstack.org/101612 | 17:42 |
| uvirtbot | Launchpad bug 1332512 in tempest "Tempest ssh unit tests have races" [Undecided,In progress] https://launchpad.net/bugs/1332512 | 17:42 |
| *** e0ne has joined #openstack-infra | 17:43 | |
| *** mspreitz has joined #openstack-infra | 17:44 | |
| zaro | clarkb: i have a test few isloated jenkins master+gearman plugin and slave setup if you need. i have tested this functionality and it does run on the correct slave on my isolated test env. | 17:44 |
| openstackgerrit | Michael Krotscheck proposed a change to openstack-infra/storyboard-webclient: Display dates in timeago format https://review.openstack.org/96713 | 17:45 |
| zaro | clarkb: i didn't use zuul for the test, just my own gearman client. | 17:45 |
| *** ramashri has quit IRC | 17:45 | |
| clarkb | zaro: I think the gearman plugin is at fault | 17:45 |
| clarkb | http://paste.openstack.org/show/84587/ | 17:45 |
| clarkb | notice the job name uses a label but the worker in the handle is for a different label | 17:46 |
| clarkb | https://jenkins02.openstack.org/computer/devstack-precise-rax-ord-486287/? is that node | 17:46 |
| clarkb | the devstack-trusty label is not applied to it | 17:46 |
| *** johnthetubaguy is now known as zz_johnthetubagu | 17:47 | |
| mordred | jeblair: ++ to simplifying due to lack of workspace reuse | 17:47 |
| *** mriedem has quit IRC | 17:47 | |
| mordred | jesusaurus: also, we're probably not going to win prizes at jenkins conferences in the future anyway | 17:47 |
| clarkb | er sorry that is not the correct node | 17:47 |
| clarkb | let me find that node | 17:47 |
| mspreitz | Does anybody here use DevStack with flat (DHCP or not) nova networking? | 17:48 |
| *** pdmars has quit IRC | 17:48 | |
| clarkb | hrm that node doesn't exist anymore, but the log is pretty wrong | 17:48 |
| *** annegent_ has joined #openstack-infra | 17:48 | |
| *** yamahata has quit IRC | 17:48 | |
| mspreitz | vishy: are you there? | 17:49 |
| clarkb | so we probably need to consider rolling back the whole stack... I am looking to see if the bug is obvious in the gearman plugin | 17:49 |
| openstackgerrit | A change was merged to openstack-infra/storyboard-webclient: Sync name validation regex with API https://review.openstack.org/100458 | 17:50 |
| sdague | clarkb: what's happening? it's just not allocating the right class of nodes? | 17:51 |
| clarkb | sdague: the jenkins gearman plugin is either registering precise nodes as able to run trusty jobs | 17:51 |
| clarkb | sdague: or it is giving trusty jobs to precise nodes regardless | 17:51 |
| *** changbl has joined #openstack-infra | 17:51 | |
| clarkb | but from that log entry it apepars that zuul did everything properly and jenkins came back with a funny answer | 17:52 |
| clarkb | actually we don't need to rollback the first change in the stack | 17:52 |
| *** salv-orlando has joined #openstack-infra | 17:54 | |
| *** _nadya_ has quit IRC | 17:55 | |
| *** pdmars has joined #openstack-infra | 17:56 | |
| *** ramashri has joined #openstack-infra | 17:56 | |
| *** otherwiseguy has joined #openstack-infra | 17:56 | |
| *** Abhishe__ has joined #openstack-infra | 17:56 | |
| *** annegent_ has quit IRC | 17:56 | |
| *** ominakov has quit IRC | 17:57 | |
| clarkb | zaro: https://git.openstack.org/cgit/openstack-infra/gearman-plugin/tree/src/main/java/hudson/plugins/gearman/ExecutorWorkerThread.java#n187 could that be registering both lables on nodes with only one of the labels? | 17:58 |
| clarkb | zaro: I think that is the bug | 17:59 |
| clarkb | if it is true once it is true every time | 18:00 |
| clarkb | and each projectLabel is added | 18:00 |
| *** e0ne has quit IRC | 18:00 | |
| *** _nadya_ has joined #openstack-infra | 18:01 | |
| jeblair | clarkb: agreed | 18:01 |
| *** e0ne has joined #openstack-infra | 18:01 | |
| clarkb | I am makign some revert commits now | 18:01 |
| pleia2 | good morning! finally home, if terribly jetlagged :) | 18:02 |
| *** marcoemorais has quit IRC | 18:02 | |
| *** marcoemorais has joined #openstack-infra | 18:02 | |
| *** eharney has joined #openstack-infra | 18:03 | |
| *** ociuhandu has quit IRC | 18:04 | |
| openstackgerrit | Clark Boylan proposed a change to openstack-infra/config: Revert "Run all dsvm jobs from juno onwards on trusty" https://review.openstack.org/101615 | 18:05 |
| openstackgerrit | Clark Boylan proposed a change to openstack-infra/config: Revert "Update jenkins jobs to optionally allow trusty" https://review.openstack.org/101616 | 18:05 |
| clarkb | jeblair: ^ | 18:05 |
| clarkb | in this case I think it is releatively safe to merge them together as zuul will update before the jenkins masters | 18:05 |
| *** e0ne has quit IRC | 18:06 | |
| jeblair | clarkb, zaro: i'm going to mock up a gearman plugin change real quick | 18:07 |
| *** e0ne has joined #openstack-infra | 18:07 | |
| clarkb | ok | 18:07 |
| zaro | clarkb: hmm. actually i thought that's desired functionality. | 18:07 |
| morganfainberg | pleia2, welcome home! :) | 18:07 |
| clarkb | I was going to look at that next, but realized i would have to dig into jenkins apis | 18:07 |
| pleia2 | morganfainberg: thanks :) | 18:07 |
| clarkb | zaro: what is desired? | 18:07 |
| zaro | clarkb: i thought current functionality is by design? | 18:08 |
| clarkb | zaro: the issue here is say I haev node1 with label1 and node2 with label2. Now I add project with label expression label1 || label2 | 18:08 |
| clarkb | zaro: if zuul says run project on label2 only it should only run on nodes with label2 | 18:09 |
| clarkb | this means only node2 and never node1 | 18:09 |
| *** gcha has quit IRC | 18:09 | |
| clarkb | zaro: if zuul says run project I don't care what label, then either is fine | 18:09 |
| *** _nadya_ has quit IRC | 18:09 | |
| clarkb | zaro: the bug is that gearman-plugin will register both labels in gearman for both nodes | 18:10 |
| clarkb | zaro: when it should only register the labels belonging to each node | 18:10 |
| zaro | clarkb: yeah, i agree. | 18:12 |
| *** Woodie has quit IRC | 18:12 | |
| *** blamar has joined #openstack-infra | 18:12 | |
| andreaf_ | clarkb: any chance you could have a look at https://review.openstack.org/#/c/98365/ again? It has now a +2 already and it's implemented they way you recommended | 18:12 |
| clarkb | andreaf_: sure | 18:13 |
| andreaf_ | clarkb: great thanks :) | 18:13 |
| *** marcoemorais has quit IRC | 18:13 | |
| *** dkehn_ has joined #openstack-infra | 18:13 | |
| *** ociuhandu has joined #openstack-infra | 18:14 | |
| *** dkehn_ has quit IRC | 18:14 | |
| *** dkehn_ has joined #openstack-infra | 18:15 | |
| *** dkehn has quit IRC | 18:15 | |
| *** dkehnx has quit IRC | 18:15 | |
| clarkb | +2 | 18:16 |
| clarkb | and approved | 18:16 |
| clarkb | I checked the logs for some of the test jobs too and it seems to have coninuted to pick the right branches adn services | 18:16 |
| mordred | andreaf_: ++ nice job | 18:17 |
| *** dkehn has joined #openstack-infra | 18:17 | |
| *** _nadya_ has joined #openstack-infra | 18:18 | |
| clarkb | andreaf_: note we are currently in a weird place as far as slaves go. That change may fail tests. We will just have to kick it again later when things are back to normal | 18:19 |
| andreaf_ | clarkb, mordred: thanks | 18:19 |
| *** W00die has joined #openstack-infra | 18:23 | |
| *** annegent_ has joined #openstack-infra | 18:24 | |
| *** CaptTofu_ has quit IRC | 18:24 | |
| *** mugsie has quit IRC | 18:24 | |
| *** CaptTofu_ has joined #openstack-infra | 18:24 | |
| *** pelix has quit IRC | 18:24 | |
| clarkb | jeblair: mordred: I am going to step away for a few minutes, but feel free to approve my changes if you are happy with them | 18:25 |
| clarkb | I will let jeblair decide on that based on his gearman spelunking | 18:26 |
| *** krotscheck has quit IRC | 18:28 | |
| *** lcheng has quit IRC | 18:29 | |
| *** reed has joined #openstack-infra | 18:29 | |
| *** CaptTofu_ has quit IRC | 18:29 | |
| *** krotscheck has joined #openstack-infra | 18:31 | |
| *** lcheng has joined #openstack-infra | 18:32 | |
| *** krotscheck has quit IRC | 18:32 | |
| *** ramashri has quit IRC | 18:33 | |
| *** ominakov has joined #openstack-infra | 18:33 | |
| *** marcoemorais has joined #openstack-infra | 18:34 | |
| *** Abhishe__ is now known as abhi_ | 18:34 | |
| openstackgerrit | Matthew Treinish proposed a change to openstack-infra/elastic-recheck: Add query for bug 1332512 https://review.openstack.org/101612 | 18:35 |
| uvirtbot | Launchpad bug 1332512 in tempest "Tempest ssh unit tests have races" [Undecided,In progress] https://launchpad.net/bugs/1332512 | 18:35 |
| *** abhi_ is now known as abhi_cs | 18:35 | |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/gearman-plugin: Fix project-node registration https://review.openstack.org/101621 | 18:35 |
| *** krotscheck has joined #openstack-infra | 18:36 | |
| jeblair | clarkb, zaro: that is untested as java code, however, I did run this groovy script to make sure that the atoms behaved the way i expected: http://paste.openstack.org/show/84589/ | 18:36 |
| *** ramashri has joined #openstack-infra | 18:37 | |
| jeblair | clarkb, zaro: with output: http://paste.openstack.org/show/84590/ | 18:37 |
| *** markmcclain has joined #openstack-infra | 18:38 | |
| openstackgerrit | A change was merged to openstack-infra/elastic-recheck: Add query for bug 1332512 https://review.openstack.org/101612 | 18:38 |
| uvirtbot | Launchpad bug 1332512 in tempest "Tempest ssh unit tests have races" [Undecided,In progress] https://launchpad.net/bugs/1332512 | 18:38 |
| anteaya | pleia2: welcome home | 18:39 |
| pleia2 | anteaya: thanks! | 18:39 |
| pleia2 | a week of inferior internet access, so much email | 18:40 |
| *** mestery has joined #openstack-infra | 18:40 | |
| *** changbl has quit IRC | 18:41 | |
| *** radez_g0n3 is now known as radez | 18:45 | |
| *** ihrachyshka has quit IRC | 18:45 | |
| *** tkelsey has quit IRC | 18:46 | |
| *** devananda has quit IRC | 18:47 | |
| clarkb | jeblair it failed the java testing | 18:48 |
| *** gokrokve_ has joined #openstack-infra | 18:49 | |
| anteaya | krtaylor: what is power kvm? | 18:50 |
| *** mestery has quit IRC | 18:50 | |
| anteaya | krtaylor: is there a hardware component? or just proprietary code? | 18:51 |
| clarkb | jeblair but I like the change /me looks at failure | 18:52 |
| *** gokrokve has quit IRC | 18:52 | |
| *** dkehn__ has joined #openstack-infra | 18:53 | |
| jeblair | clarkb: oh i think it's the import for Set | 18:53 |
| jeblair | nope | 18:53 |
| jeblair | labelatom? | 18:53 |
| *** dkehn__ is now known as dkehnx | 18:53 | |
| krtaylor | anteaya, PowerKVM is a system based on Power8 hardware | 18:54 |
| *** dkehn_ has quit IRC | 18:54 | |
| *** dkehn has quit IRC | 18:54 | |
| krtaylor | anteaya, I just finished the IBMPowerKVMCI edits, and added us to the Third Party Table | 18:54 |
| *** ildikov has quit IRC | 18:54 | |
| jeblair | clarkb: i love that in the logs we don't actually see which symbol it couldn't find. ;) | 18:55 |
| *** devananda has joined #openstack-infra | 18:55 | |
| krtaylor | btw, the table itself has a typo, was it intentional? ThirdPartySytemTableEntry | 18:55 |
| openstackgerrit | James E. Blair proposed a change to openstack-infra/gearman-plugin: Fix project-node registration https://review.openstack.org/101621 | 18:55 |
| clarkb | jeblair ya | 18:56 |
| *** changbl has joined #openstack-infra | 18:56 | |
| *** CaptTofu_ has joined #openstack-infra | 18:57 | |
| *** harlowja is now known as harlowja_away | 18:57 | |
| anteaya | krtaylor: yes, this prompted my question | 18:57 |
| anteaya | krtaylor: there is an assumption that I know what power kvm is, I don't | 18:58 |
| *** dkehn has joined #openstack-infra | 18:58 | |
| anteaya | typos are my fault, not intentional | 18:58 |
| clarkb | jeblair still failing. LabelAtom maybe? | 18:58 |
| anteaya | let me look at it | 18:58 |
| clarkb | oh no its there | 18:58 |
| openstackgerrit | Elizabeth K. Joseph proposed a change to openstack-infra/publications: Minor updates, new gerrit and zuul screenshots https://review.openstack.org/99173 | 18:59 |
| pleia2 | ^^ had one tiny tweak before my talk last week, but I'm done meddling now :) | 18:59 |
| *** dkehnx has quit IRC | 19:00 | |
| krtaylor | anteaya, the ThirdPartySystems page is live now? or still coming soon? | 19:00 |
| *** _nadya_ has quit IRC | 19:01 | |
| pleia2 | https://www.flickr.com/photos/vedran-art-photos/14450415442/in/pool-dc2014/ woo, pancake slide | 19:01 |
| *** harlowja_away is now known as harlowja | 19:01 | |
| *** dkehn_ has joined #openstack-infra | 19:01 | |
| krtaylor | anteaya, anything I can do to help with it? | 19:01 |
| *** zehicle_at_dell has joined #openstack-infra | 19:01 | |
| clarkb | name has protected access in Label | 19:02 |
| anteaya | krtaylor: it is a work in progress | 19:03 |
| clarkb | jeblair: is there a tostring? | 19:03 |
| anteaya | krtaylor: keeping your page editied is a great start for now, thank you | 19:03 |
| clarkb | jeblair: there is a getName() | 19:03 |
| clarkb | which is an alias for getDisplayName() | 19:04 |
| *** mspreitz has quit IRC | 19:06 | |
| *** Sukhdev has joined #openstack-infra | 19:06 | |
| anteaya | krtaylor: you could review these: https://review.openstack.org/#/c/101227/ and https://review.openstack.org/#/c/101013/ | 19:10 |
| anteaya | that would help me | 19:10 |
| *** lukego has quit IRC | 19:11 | |
| *** lukego has joined #openstack-infra | 19:12 | |
| *** esker has joined #openstack-infra | 19:16 | |
| openstackgerrit | Clark Boylan proposed a change to openstack-infra/gearman-plugin: Fix project-node registration https://review.openstack.org/101621 | 19:16 |
| clarkb | jeblair: ^ that passed `mvn test` locally | 19:17 |
| clarkb | well I need to get ready for lunch and some errands | 19:18 |
| clarkb | I believe the reverts are raedy to go if we decide that is necessary | 19:18 |
| clarkb | zaro: can you test that change? I think earlier you mentioned you might have a test env setup | 19:25 |
| *** bhuvan has joined #openstack-infra | 19:26 | |
| zaro | clarkb: will take a look in a min. | 19:28 |
| *** _nadya_ has joined #openstack-infra | 19:31 | |
| *** salv-orlando has quit IRC | 19:33 | |
| krtaylor | anteaya, first done, looking at second now | 19:34 |
| *** abhi_cs has quit IRC | 19:35 | |
| *** pdmars has quit IRC | 19:38 | |
| *** pdmars has joined #openstack-infra | 19:38 | |
| *** annegent_ has quit IRC | 19:38 | |
| anteaya | krtaylor: thanks | 19:39 |
| krtaylor | anteaya, for the name change and request change - someting I have wondered for a while, do we require an individual representing the entity doing the third party testing to have a signed ICLA and gerrit acct? | 19:39 |
| krtaylor | I would think so, but I don't recall seeing that anywhere | 19:39 |
| anteaya | no | 19:40 |
| anteaya | since commenting doestn' require a cla | 19:40 |
| *** blamar has quit IRC | 19:40 | |
| anteaya | just submitting a patch | 19:40 |
| *** ihrachyshka has joined #openstack-infra | 19:40 | |
| anteaya | and third party ci systems do not submit patches | 19:40 |
| krtaylor | ok, I did not know that, good to know | 19:40 |
| anteaya | good question | 19:41 |
| *** mestery has joined #openstack-infra | 19:41 | |
| krtaylor | I understand the system account wouldnt submit patches, thats why I was wondering if we require an individual | 19:41 |
| anteaya | right | 19:41 |
| anteaya | yes, iCLA not required for third party ci systems | 19:42 |
| krtaylor | I can't imagine someone not having an individual account but I guess it isnt required | 19:42 |
| anteaya | it is not required | 19:42 |
| anteaya | and yes some don't have an individual account | 19:42 |
| anteaya | although we never envisioned that either | 19:42 |
| anteaya | hence some of the problems we have been faced with | 19:42 |
| krtaylor | sigh, yes | 19:43 |
| anteaya | and our seemingly slow process to deal with them | 19:43 |
| anteaya | since we are in the flood before we even have a chance to discuss it | 19:43 |
| anteaya | since people are doing stuff we don't think of preventing | 19:43 |
| krtaylor | I do like the one method to request accounts, that will help | 19:44 |
| anteaya | thanks | 19:44 |
| anteaya | I think so too | 19:44 |
| *** mestery_ has joined #openstack-infra | 19:44 | |
| *** mestery has quit IRC | 19:45 | |
| *** annegent_ has joined #openstack-infra | 19:45 | |
| *** mestery_ is now known as mestery | 19:45 | |
| anteaya | thanks for the reviews, krtaylor | 19:46 |
| krtaylor | np, btw, you have excellent commit messages, I wish more were that detailed | 19:46 |
| anteaya | thanks | 19:47 |
| anteaya | do feel free to steal anything good | 19:47 |
| anteaya | spread it around | 19:47 |
| krtaylor | hehheh | 19:47 |
| anteaya | it is after all, opensource|open source|open-source | 19:47 |
| krtaylor | hehheh, I did review the neutron CI pages too, but I cant remember who requested it | 19:48 |
| krtaylor | I guess I could go look at the meeting log | 19:48 |
| *** marcoemorais has quit IRC | 19:48 | |
| *** marcoemorais1 has joined #openstack-infra | 19:48 | |
| anteaya | you never go wrong when you look at teh logs | 19:48 |
| anteaya | and thanks for those reviews too | 19:49 |
| *** marcoemorais1 has quit IRC | 19:49 | |
| *** marcoemorais has joined #openstack-infra | 19:49 | |
| krtaylor | mestery, was that you requesting the review on the neutron third party testing pages? | 19:49 |
| mestery | krtaylor: Yes, reviews welcome and encouraged! :) | 19:50 |
| *** ildikov has joined #openstack-infra | 19:50 | |
| krtaylor | mestery, I just finished reading it all, looks really good | 19:50 |
| *** CaptTofu_ has quit IRC | 19:50 | |
| mestery | krtaylor: Thanks for the feedback! | 19:50 |
| krtaylor | that and the plugin/driver page | 19:50 |
| *** CaptTofu_ has joined #openstack-infra | 19:51 | |
| *** lcheng has quit IRC | 19:51 | |
| *** thedodd has quit IRC | 19:55 | |
| *** CaptTofu_ has quit IRC | 19:55 | |
| *** marcoemorais has quit IRC | 19:58 | |
| *** marcoemorais has joined #openstack-infra | 19:58 | |
| ryanpetrello | anybody know what “NOT_REGISTERED” in failing CI check job means? | 20:00 |
| ryanpetrello | https://review.openstack.org/#/c/94244/, never seen this one before, and to my knowledge, I haven’t changed any of pecan’s infra configs re: ceilometer | 20:00 |
| *** salv-orlando has joined #openstack-infra | 20:01 | |
| *** ildikov has quit IRC | 20:02 | |
| *** marcoemorais1 has joined #openstack-infra | 20:06 | |
| *** _nadya_ has quit IRC | 20:06 | |
| *** _nadya_ has joined #openstack-infra | 20:06 | |
| *** abhi_cs has joined #openstack-infra | 20:06 | |
| *** lcheng has joined #openstack-infra | 20:07 | |
| clarkb | I think that is a result of todays changes | 20:07 |
| clarkb | jeblair mordred should we go ahead and revert? | 20:08 |
| zaro | clarkb: i'm testing now. | 20:09 |
| *** dprince has quit IRC | 20:10 | |
| *** marcoemorais has quit IRC | 20:10 | |
| clarkb | thanks | 20:10 |
| jeblair | clarkb: up to you; i don't think i can do much babysitting today | 20:11 |
| *** homeless has joined #openstack-infra | 20:11 | |
| *** ildikov has joined #openstack-infra | 20:12 | |
| *** mmaglana has quit IRC | 20:13 | |
| clarkb | ok I will probably do that after lunch | 20:13 |
| *** abhi_cs has quit IRC | 20:14 | |
| *** _nadya_ has quit IRC | 20:14 | |
| *** abhi_cs has joined #openstack-infra | 20:15 | |
| *** radez is now known as radez_g0n3 | 20:16 | |
| *** lcheng has quit IRC | 20:17 | |
| *** e0ne has quit IRC | 20:18 | |
| *** e0ne has joined #openstack-infra | 20:19 | |
| *** melwitt has joined #openstack-infra | 20:22 | |
| *** gokrokve_ has quit IRC | 20:26 | |
| *** thedodd has joined #openstack-infra | 20:26 | |
| *** annegent_ has quit IRC | 20:26 | |
| *** annegent_ has joined #openstack-infra | 20:27 | |
| *** mestery has quit IRC | 20:28 | |
| *** markmcclain has quit IRC | 20:28 | |
| *** lcheng has joined #openstack-infra | 20:29 | |
| *** maxbit has joined #openstack-infra | 20:33 | |
| *** thedodd has quit IRC | 20:37 | |
| reed | sdague, pointers to funny reviews? | 20:37 |
| sdague | reed: not off the top of my head, but I'd keep my eye out for them if we were doing them | 20:38 |
| clarkb | I approved first revert | 20:38 |
| reed | sdague, thanks :) | 20:38 |
| *** thedodd has joined #openstack-infra | 20:38 | |
| sdague | clarkb: ok, so what's the crux of the issue? for those of us tl;dr | 20:38 |
| sdague | just so I edumacate myself | 20:39 |
| clarkb | sdague: jenkins gearman plugin is registering jobs on nodes it shouldnt | 20:39 |
| sdague | oh, fun | 20:39 |
| clarkb | our jobs have the form of name:label | 20:39 |
| *** lcheng has quit IRC | 20:39 | |
| *** markmcclain has joined #openstack-infra | 20:39 | |
| clarkb | so we should get tempest:trusty on trust nodes only | 20:40 |
| clarkb | but its registering that job on precise too | 20:40 |
| *** wenlock has quit IRC | 20:40 | |
| clarkb | so zuul says only run on trusty and gets precise | 20:40 |
| *** zzelle has joined #openstack-infra | 20:40 | |
| openstackgerrit | A change was merged to openstack-infra/config: Revert "Run all dsvm jobs from juno onwards on trusty" https://review.openstack.org/101615 | 20:40 |
| jeblair | sdague: potential fix is here if you want to review: https://review.openstack.org/#/c/101621/3 | 20:41 |
| jeblair | i think zaro has a test env handy and will test it | 20:41 |
| sdague | oh, man, java. It's been a while, I'll do my best | 20:41 |
| *** esker has quit IRC | 20:43 | |
| sdague | yeh, I don't have enough context here to provide you with anything useful in reviewing that code | 20:43 |
| jeblair | heh; hopefully static typing will catch all the bugs then! :) | 20:43 |
| *** _nadya_ has joined #openstack-infra | 20:45 | |
| clarkb | ok going to approve second revert now | 20:46 |
| clarkb | this should separste the application of them by a puppet run | 20:46 |
| *** zhiyan_ has joined #openstack-infra | 20:46 | |
| zaro | clarkb, jeblair : ok, i've verified that the change works as expected for || labels. will continue to test the other conditions (&&, !, etc..) | 20:49 |
| *** CaptTofu_ has joined #openstack-infra | 20:50 | |
| *** lcheng has joined #openstack-infra | 20:51 | |
| openstackgerrit | A change was merged to openstack-infra/config: Revert "Update jenkins jobs to optionally allow trusty" https://review.openstack.org/101616 | 20:53 |
| *** markmcclain has quit IRC | 20:54 | |
| *** weshay has quit IRC | 20:54 | |
| *** jareking has joined #openstack-infra | 20:54 | |
| *** CaptTofu_ has quit IRC | 20:55 | |
| *** wenlock has joined #openstack-infra | 20:56 | |
| *** gokrokve has joined #openstack-infra | 20:57 | |
| clarkb | zaro: great so maybe we can roll that into the other plugin upgrade | 20:57 |
| clarkb | and crank it out early next week | 20:57 |
| clarkb | (this day is already crazy for me) | 20:57 |
| *** gokrokve_ has joined #openstack-infra | 20:59 | |
| *** melwitt has quit IRC | 20:59 | |
| *** melwitt has joined #openstack-infra | 20:59 | |
| openstackgerrit | Darragh Bailey proposed a change to openstack-infra/jenkins-job-builder: Make 'keep_descriptions' fully optional https://review.openstack.org/100098 | 20:59 |
| *** melwitt has quit IRC | 21:00 | |
| *** melwitt has joined #openstack-infra | 21:01 | |
| *** lcheng has quit IRC | 21:01 | |
| *** melwitt has joined #openstack-infra | 21:01 | |
| *** melwitt has quit IRC | 21:01 | |
| *** gokrokve has quit IRC | 21:02 | |
| *** melwitt has joined #openstack-infra | 21:02 | |
| *** melwitt has quit IRC | 21:02 | |
| *** melwitt has joined #openstack-infra | 21:02 | |
| *** gokrokve_ has quit IRC | 21:04 | |
| *** mfer has quit IRC | 21:04 | |
| *** Sukhdev has quit IRC | 21:06 | |
| *** marcoemorais1 has quit IRC | 21:07 | |
| *** marcoemorais has joined #openstack-infra | 21:07 | |
| sdague | clarkb: also, it looks like trusty nodes got sent to icehouse at least in the tempest case | 21:08 |
| sdague | http://logs.openstack.org/11/101511/2/check/check-grenade-dsvm-icehouse/aacd3fc/console.html | 21:08 |
| *** jamielennox|away has quit IRC | 21:08 | |
| *** pelix has joined #openstack-infra | 21:08 | |
| sdague | clarkb: yeh, and failed there because the nodepool images didn't have the old cirros | 21:09 |
| clarkb | yup | 21:10 |
| clarkb | same bug | 21:10 |
| *** marcoemorais has quit IRC | 21:10 | |
| clarkb | should be better in a short bit | 21:11 |
| *** marcoemorais has joined #openstack-infra | 21:11 | |
| *** marcoemorais has quit IRC | 21:11 | |
| clarkb | jjb needs to update all the jobs now | 21:11 |
| anteaya | what bug are we using for these failures? | 21:11 |
| anteaya | have we a number yet? | 21:11 |
| *** Sukhdev has joined #openstack-infra | 21:12 | |
| *** marcoemorais has joined #openstack-infra | 21:12 | |
| sdague | anteaya: I'm not sure that we do | 21:12 |
| sdague | ER found a different fail on one of the jobs, so I rechecked with that | 21:12 |
| sdague | given that tempest runs 27 jobs in check.... it is very sensitive to instability | 21:12 |
| clarkb | I didnt submit one but can here shortly | 21:12 |
| clarkb | just picked up a 14.6 lb brisket \o/ | 21:13 |
| anteaya | yay for you | 21:13 |
| sdague | clarkb: cool, is it always the cirros issue? I'll work on an ER fingerprint | 21:13 |
| *** lcheng has joined #openstack-infra | 21:13 | |
| clarkb | maybe? | 21:13 |
| sdague | clarkb: jealous | 21:13 |
| clarkb | it could be anything related to trusty being different | 21:13 |
| anteaya | I like to be armed with a bug number, like a walmart greeter | 21:14 |
| anteaya | then I can hand it out | 21:14 |
| sdague | oh, the grenade icehouse is actually explicit fail | 21:14 |
| anteaya | when the herd arrives | 21:14 |
| sdague | WARNING: this script has not been tested on trusty | 21:14 |
| sdague | because old == havana | 21:14 |
| anteaya | ha ha ha | 21:14 |
| anteaya | nice | 21:14 |
| anteaya | so grenade must have precise for juno? | 21:15 |
| sdague | oh, and that patch was litterally the only thing that hit it | 21:15 |
| sdague | anteaya: no | 21:15 |
| sdague | it's check-grenade-dsvm-icehouse | 21:15 |
| sdague | which is havana => icehouse | 21:15 |
| sdague | so that has to be precise | 21:15 |
| anteaya | ah | 21:15 |
| sdague | but icehouse => juno is fine on trusty | 21:16 |
| sdague | that's a job we have for branchless tempest | 21:16 |
| anteaya | but that will be havana => icehouse for all of juno? | 21:16 |
| anteaya | oh | 21:16 |
| anteaya | so only for backports | 21:16 |
| sdague | well and making sure that tempest doesn't stop working on it | 21:16 |
| *** ArxCruz has quit IRC | 21:17 | |
| sdague | branchless tempest means we have to test tempest against a ton of things, because we're using a single source tree | 21:17 |
| anteaya | oh | 21:17 |
| *** SumitNaiksatam has quit IRC | 21:18 | |
| sdague | clarkb: actually, the icehouse pg fail was a normal race | 21:18 |
| sdague | it actually worked on trusty | 21:18 |
| pleia2 | it's a week late based on our typical schedule, but I'm thinking Bug Day for Monday (Tuesday-Thursday I'm less available) | 21:18 |
| *** abhi_cs has quit IRC | 21:19 | |
| anteaya | we should do it | 21:19 |
| openstackgerrit | Josh Gachnang proposed a change to openstack-infra/config: Add Ironic Python Agent builder https://review.openstack.org/101654 | 21:19 |
| sdague | as long as it doesn't impact the Trusty roll over | 21:19 |
| anteaya | sdague: do you think it might? | 21:19 |
| *** abhi_cs has joined #openstack-infra | 21:19 | |
| sdague | I don't know how consumed people get with it | 21:19 |
| sdague | but given that we're going to have to try again next week, I think it would be better to be monday attempt if possible | 21:20 |
| pleia2 | or maybe the next week, Jul 1 | 21:20 |
| sdague | a bunch of things flow out once trusty is in there | 21:20 |
| pleia2 | I'll email the list for thoughts, we seems a bit scattered today :) | 21:20 |
| sdague | including being able to test mongo in the gate for people that need that | 21:20 |
| anteaya | kk | 21:20 |
| anteaya | July 1 is Canada Day, so a holiday for me | 21:21 |
| sdague | and I think neutron is going to get more stable on it | 21:21 |
| anteaya | which I will probably actually take this time | 21:21 |
| pleia2 | anteaya: ah | 21:21 |
| anteaya | sdague: neutron will be more stable on trusty? | 21:21 |
| sdague | anteaya: yeh, openvswitch 2.0 | 21:21 |
| openstackgerrit | Josh Gachnang proposed a change to openstack-infra/config: Add Ironic Python Agent builder https://review.openstack.org/101654 | 21:21 |
| sdague | salv-orlando said it helped a bunch in his testing | 21:21 |
| anteaya | sdague: awesome | 21:22 |
| anteaya | trusty is the priority then, more stable neutron ftw | 21:22 |
| *** lcheng has quit IRC | 21:23 | |
| salv-orlando | sdague: mostyl because we did not hit the kernel bg that plagued us back injanuary | 21:23 |
| *** dstanek has joined #openstack-infra | 21:23 | |
| sdague | salv-orlando: ok, I thought you said it could do more load as well | 21:24 |
| *** ArxCruz has joined #openstack-infra | 21:24 | |
| *** mrodden has quit IRC | 21:24 | |
| *** ArxCruz has quit IRC | 21:24 | |
| sdague | regardless, it will be a newer kernel too :) | 21:24 |
| sdague | and new libvirt | 21:24 |
| clarkb | new libvirt and new mongodb | 21:24 |
| clarkb | and new apache | 21:24 |
| clarkb | there is a bit of goodness here | 21:24 |
| sdague | yeh | 21:24 |
| JayF | clarkb: https://review.openstack.org/#/c/101654/ this is JoshNang implementing what we talked about the other day | 21:25 |
| JayF | clarkb: your review would be appreciated | 21:25 |
| JayF | clarkb: or any suggestions as to how to better test it :) | 21:25 |
| JoshNang | I'm sure there's at least one thing wrong :) | 21:25 |
| sdague | new apache was my enemy 6 months ago. Changing the whole access stanza model is really good way to break all your websites | 21:25 |
| JayF | that looks much too easy, haha | 21:25 |
| JayF | I wonder if we need something in there to feed the URLs from the uploaded agent images into the gerrit comments | 21:26 |
| JayF | and if it should be voting or not | 21:26 |
| anteaya | sdague: :( | 21:27 |
| *** pdmars has quit IRC | 21:27 | |
| *** annegent_ has quit IRC | 21:27 | |
| clarkb | JayF: for the urls simplest thing may be to link from the console logs | 21:27 |
| clarkb | JayF: currently trying to sort out trusty stuff. We switched and it didn't quite work :( | 21:28 |
| clarkb | but we have a fix up unfortunately I doubt it will be implemented today | 21:28 |
| clarkb | the good news is that other than this one hiccup I think everything mostly worked | 21:28 |
| clarkb | so all in all not terrible | 21:28 |
| JayF | clarkb: What do you mean, link from the console logs? | 21:28 |
| JayF | clarkb: and otherwise how does the PR look? | 21:28 |
| clarkb | sdague: anteaya https://bugs.launchpad.net/openstack-ci/+bug/1332716 is a bug | 21:28 |
| uvirtbot | Launchpad bug 1332716 in openstack-ci "gearman plugin registers all labels for jobs on node with subset of labels" [Undecided,New] | 21:28 |
| clarkb | JayF: in the job itself say The image from this job can be found at : | 21:29 |
| clarkb | JayF: or you can just publish in a knwn location and document on eg the wiki | 21:29 |
| clarkb | JayF: but ya that looks good. You will need to add it to zuul as well | 21:30 |
| clarkb | I will comment on the change | 21:30 |
| JayF | Will the publishers: stuff add any information about, for isntance, what git hash it was built against | 21:30 |
| JayF | or is all that stuff just handled for you | 21:30 |
| anteaya | clarkb: thanks | 21:30 |
| Kiall | soren: heya - about? Hoping you can /join uvirtbot into #openstack-dns :) | 21:30 |
| clarkb | JayF: it will not | 21:31 |
| clarkb | JayF: you could add a step to the build that does md5sums and also puts the git sha in a file and copy those too | 21:31 |
| JoshNang | clarkb: also, does this only get run every time something is merged or on each change request? | 21:31 |
| clarkb | JoshNang: right now that change won't run the test ever :) so that is something we can think about and add in the appropriate way | 21:32 |
| clarkb | JoshNang: I think it should only publish whenever something is merged | 21:32 |
| clarkb | JoshNang: so you might want a second job that does the build every patch but only publishes post merge | 21:32 |
| *** lukego has quit IRC | 21:32 | |
| JayF | clarkb: we were actually hoping to have this publish for every merge req for the agent | 21:32 |
| clarkb | JoshNang: but I don't quite know how you intend on using the images so you guys probably have a better idea | 21:32 |
| clarkb | JayF: cool | 21:32 |
| clarkb | JayF: or wait | 21:32 |
| clarkb | you want published artifacts for unmerged code? | 21:33 |
| JayF | clarkb: basically an IPA developer shoudl be able to download the image created against a patch, drop it into devstack and run tests | 21:33 |
| clarkb | JayF: would they build the image locally? | 21:33 |
| JayF | and eventually we'd ramp up this test (once said devstack integration exists) | 21:33 |
| JayF | clarkb: they /could/ but I'd strongly prefer it be done by a builder | 21:33 |
| clarkb | and you grab known good images off of the publishing site? | 21:33 |
| clarkb | huh | 21:34 |
| JayF | clarkb: wdym by 'grab known good images'? Here's a workflow I'd envision: | 21:34 |
| clarkb | JayF: lets take the project sdist example | 21:34 |
| clarkb | we build sdists for every proposed patch but do not publish them | 21:34 |
| clarkb | when you merge code the sdist is built again and published | 21:34 |
| clarkb | so we have known good sdists at all times | 21:34 |
| JayF | is there any way for a developer to 'get at' those sdists? | 21:34 |
| clarkb | if a developer needs an sdist for something unmerged they build it locally | 21:35 |
| clarkb | using python setup.py sdist | 21:35 |
| *** lcheng has joined #openstack-infra | 21:35 | |
| JoshNang | JayF: i think as far as testing goes, a devstack job that builds it before attempting to deploy to it might make more sense | 21:35 |
| clarkb | that is sort of how I envision image publishing would work | 21:35 |
| clarkb | because otherwise you will end up with hundreds of useless and large images to sort through | 21:35 |
| JayF | clarkb: *I* do the local builds of these images often, it was just said that many developers would like to have an agent image already available | 21:36 |
| JayF | honestly building on every merge is orders of magnitude better than what we're doing now, so I'm OK just doing that | 21:36 |
| clarkb | JayF: right you would make known good images available | 21:36 |
| clarkb | JayF: hold on I have examples for you | 21:36 |
| *** ominakov has quit IRC | 21:36 | |
| clarkb | http://tarballs.openstack.org/nova/ | 21:37 |
| clarkb | that giant list is mostly for when sdists were built for everything | 21:37 |
| clarkb | but now we keep most recent known good for each branch | 21:37 |
| clarkb | its a lot easier as a human being to decide that is the one I want | 21:37 |
| zaro | clarkb, jeblair : ok. i've verified the rest of the scenarios for the gearman-plugin change. | 21:37 |
| zaro | looks good. | 21:38 |
| clarkb | zaro: awesome thank you | 21:38 |
| clarkb | I am going to approve that change | 21:38 |
| JayF | clarkb: these images are going to be notably larger than those | 21:38 |
| JayF | clarkb: will this pose a problem? | 21:38 |
| jeblair | zaro, clarkb: should we merge it, then tag a release and upgrade to that along with the timeout plugin? | 21:38 |
| clarkb | jeblair: yes I think so | 21:39 |
| clarkb | JayF: not if we stick to publishing known good per branch or similar | 21:39 |
| JayF | right now our images are ~162MB for the .cpio.gz and ~21MB for the kernel (probably smaller when built upstream, without our utilities built in) | 21:39 |
| clarkb | JayF: the way that job is currently set up it will only keep one of each | 21:39 |
| *** gokrokve has joined #openstack-infra | 21:39 | |
| clarkb | whcih should be ok | 21:39 |
| JayF | clarkb: I was going to change that :) | 21:39 |
| JayF | clarkb: to make it publish with the SHA1, and probably CoreOS version as well, in the tarball | 21:39 |
| JayF | and I'd be very OK with, say, keeping only the last "n" of them | 21:40 |
| clarkb | JayF: thats fine | 21:40 |
| clarkb | cool | 21:40 |
| clarkb | also fwiw we are trying to move towards swift for artifacts | 21:40 |
| clarkb | adn once we do this becomes even easier | 21:40 |
| *** sabeen has quit IRC | 21:40 | |
| clarkb | but currently we have the maximum number of maximum size cinder volumes attached to a VM in rax and are using all the disk | 21:40 |
| clarkb | :) | 21:40 |
| JayF | Heh. Our internal Jenkins that runs these uploads to our internal swift account :) | 21:41 |
| *** jareking has quit IRC | 21:41 | |
| clarkb | ya we have the jobs and everything but jenkins is munging the hmac | 21:41 |
| clarkb | :/ silly jenkins | 21:41 |
| JayF | clarkb: BTW, part of why it killed me you guys are letting me do this upstream is that I was originally going to use one of our OnMetal IO or Compute nodes to do the third party CI so I could play with my shiny new servers :) | 21:41 |
| *** amalagon has joined #openstack-infra | 21:41 | |
| clarkb | nice | 21:42 |
| JayF | clarkb: yeah, in our setup, I just have a configured swiftclient on the builder and do a swift upload | 21:42 |
| openstackgerrit | A change was merged to openstack-infra/gearman-plugin: Fix project-node registration https://review.openstack.org/101621 | 21:42 |
| clarkb | we may end up using the onMetal IO nodes for elasticsearch | 21:42 |
| sdague | JayF: well you could give us some of those for our logstash environment instead :) | 21:42 |
| JayF | sdague: *I* can't :) | 21:42 |
| JayF | well maybe technically but not organizationally | 21:42 |
| JayF | you'd have to ask russell_h ^ about that | 21:43 |
| *** lcheng has quit IRC | 21:44 | |
| clarkb | sdague: it looks like the reverts are mostly applied at this point | 21:45 |
| clarkb | do let me know if yo usee more weirdness but I Think we may be food for the weekend | 21:45 |
| sdague | clarkb: cool | 21:45 |
| clarkb | then we can upgrade gearman plugin | 21:45 |
| clarkb | and try it all again | 21:46 |
| sdague | sounds good. | 21:46 |
| clarkb | JayF: JoshNang fwiw your change won't be affected by the weirdness because you will be trusty only | 21:46 |
| *** miqui has quit IRC | 21:47 | |
| JayF | clarkb: so does that mean we would be good2go today if we got it done quickly enough with the right +2s and such | 21:47 |
| *** miqui has joined #openstack-infra | 21:47 | |
| sdague | well the fact that all the fails were just really badly matched jobs for the environment (like trying to do havana on trusty) I think we're going to be pretty good when it flips | 21:47 |
| *** miqui has quit IRC | 21:47 | |
| clarkb | JayF: yes I think so | 21:47 |
| clarkb | sdague: yup I am reasonably happy with the outcome | 21:48 |
| clarkb | sdague: like even when things were derpy it wans't the end of the world | 21:48 |
| sdague | yep | 21:48 |
| *** craigbr has joined #openstack-infra | 21:49 | |
| *** jareking has joined #openstack-infra | 21:50 | |
| *** changbl has quit IRC | 21:50 | |
| JayF | publishers: -scp: files: - target: can be a dir? | 21:50 |
| JayF | so I can do something like source: 'dir/*' target: 'dir/' | 21:50 |
| *** andreykurilin__ has joined #openstack-infra | 21:50 | |
| clarkb | yeah if yo udo that the source name will be copied | 21:51 |
| JayF | and then as long as my build step named the files properly (with git sha1 being built, and coreos version) the upload step is simple | 21:51 |
| JayF | perfect | 21:51 |
| clarkb | JayF: http://ci.openstack.org/jenkins-job-builder/ | 21:51 |
| clarkb | that should have details | 21:51 |
| *** _nadya_ has quit IRC | 21:51 | |
| JayF | to get the git sha1, in this setup, I'd just run a git log, just like I usually would in a script? | 21:51 |
| JayF | sorry to be skeptical, this is just way simpler than it seemed it shoudl be :) | 21:51 |
| clarkb | JayF: yup | 21:51 |
| clarkb | JayF: or | 21:51 |
| *** abhi_cs has quit IRC | 21:52 | |
| clarkb | JayF: it will be in the environment as $ZUUL_REF iirc | 21:52 |
| clarkb | let me double check | 21:52 |
| clarkb | since zuul has to tell the gerrit-git-prep which ref to use | 21:52 |
| *** andreykurilin_ has quit IRC | 21:52 | |
| clarkb | JayF: ZUUL_NEWREV | 21:53 |
| JayF | okay, so the only thing I'm missing now is once I publish this, how do I get folks the link? | 21:53 |
| JayF | just echo the expected url given what $ZUUL_REF is | 21:53 |
| JayF | er, ZUUL_NEWREV is in the script | 21:54 |
| JayF | then have devs look in console log? | 21:54 |
| *** abhi_cs has joined #openstack-infra | 21:54 | |
| clarkb | yup but it will also be a known location like the nova sdists | 21:54 |
| clarkb | so you can tell them http://tarballs.openstack.org/ipa/builds/ or whatever | 21:54 |
| *** ihrachyshka has quit IRC | 21:54 | |
| JayF | is there a compression format preference? I usually use xz nowadays, but can use gzip if you don't want the load :) | 21:55 |
| clarkb | we use gz | 21:55 |
| *** marcoemorais has quit IRC | 21:55 | |
| JayF | got it | 21:55 |
| clarkb | ist a good lowest common denominator | 21:55 |
| *** praneshp_ has joined #openstack-infra | 21:56 | |
| *** marcoemorais has joined #openstack-infra | 21:56 | |
| *** melwitt has quit IRC | 21:56 | |
| pleia2 | SergeyLukjanov: nice work with these 3rd party testing accounts :) | 21:56 |
| *** praneshp has quit IRC | 21:57 | |
| *** praneshp_ is now known as praneshp | 21:57 | |
| *** lcheng has joined #openstack-infra | 21:57 | |
| *** craigbr has quit IRC | 21:57 | |
| *** abhi_cs_ has joined #openstack-infra | 21:57 | |
| *** Ryan_Lane has quit IRC | 21:59 | |
| *** abhi_cs has quit IRC | 21:59 | |
| *** melwitt has joined #openstack-infra | 21:59 | |
| *** nati_ueno has quit IRC | 22:01 | |
| *** abhi_cs_ has quit IRC | 22:01 | |
| *** melwitt has quit IRC | 22:02 | |
| *** Abhishe__ has joined #openstack-infra | 22:03 | |
| *** melwitt has joined #openstack-infra | 22:03 | |
| *** melwitt has quit IRC | 22:03 | |
| *** melwitt has joined #openstack-infra | 22:03 | |
| *** andreaf_ has quit IRC | 22:04 | |
| *** e0ne has quit IRC | 22:05 | |
| *** e0ne has joined #openstack-infra | 22:05 | |
| *** melwitt1 has joined #openstack-infra | 22:06 | |
| openstackgerrit | Eoghan Glynn proposed a change to openstack-infra/config: Create openstack-gnocchi mailing list https://review.openstack.org/101662 | 22:07 |
| sdague | clarkb: ok, off for the weekend. enjoy your brisket :) | 22:07 |
| clarkb | thanks have a good one | 22:07 |
| *** lcheng has quit IRC | 22:07 | |
| clarkb | I need to run shortly too, going to head downtown | 22:07 |
| clarkb | but then I haev to wake up at ~6am for brisketing | 22:08 |
| openstackgerrit | Jay Faulkner proposed a change to openstack-infra/config: Add Ironic Python Agent builder https://review.openstack.org/101654 | 22:08 |
| JayF | clarkb: ^ can you look at that first? | 22:08 |
| clarkb | sure | 22:08 |
| anteaya | clarkb: happy brisketing | 22:08 |
| *** melwitt has quit IRC | 22:09 | |
| JayF | clarkb: fixing that line wrap weirdness now | 22:10 |
| clarkb | JayF: thanks | 22:10 |
| clarkb | JayF: but one more thing | 22:10 |
| clarkb | do we want to gate on this? | 22:10 |
| clarkb | and can we start with one known good image? | 22:10 |
| clarkb | to avoid eg nova's issue in the example | 22:10 |
| clarkb | if we need to modify that later we can | 22:11 |
| clarkb | but I don't want to jump out of the gate with spam | 22:11 |
| *** e0ne has quit IRC | 22:11 | |
| JayF | I would say for #1, probably so. If we're not able to build an image, the only thing that would cause that is someone breaking the build via a bad Dockerfile or Makefile, etc | 22:11 |
| *** e0ne has joined #openstack-infra | 22:11 | |
| clarkb | JayF: ok then you need a second job | 22:11 |
| JayF | however I understand if we don't want that to vote, so if you can lmk how to configure it to run after merge without blocking a gate | 22:11 |
| clarkb | the gate job shouldn't publish | 22:11 |
| JayF | then I'm OK with that | 22:11 |
| *** lcheng has joined #openstack-infra | 22:11 | |
| clarkb | a post job should publish | 22:12 |
| JayF | the 2nd question is not something I understand. You just want it to upload *one* file and just overwrite each build? | 22:12 |
| clarkb | JayF: yes | 22:12 |
| clarkb | this is what we do with sdists and seems to work well | 22:12 |
| JayF | I can do that, but honestly I don't like assuming 'known good' as a consumer of software, I like picking the exact SHA1 I use, but if you want to take it as baby steps, I get it | 22:13 |
| clarkb | so the issue is if you look at the nova sdists for example there is too much noise | 22:13 |
| clarkb | most people go I want whatever master is now | 22:14 |
| clarkb | or whatever stable/foo is | 22:14 |
| JayF | I doubt we'd have that level of churn in IPA vs Nova :) | 22:14 |
| JayF | but I'll happily implement it this way to get it working | 22:14 |
| clarkb | ok | 22:14 |
| JayF | so I should rename the job to post-ironig-python-agent-buildimage-coreos? | 22:14 |
| clarkb | usually we just drop the pipeline thing completely for post | 22:14 |
| clarkb | not sure why.. | 22:14 |
| clarkb | ironic-python-agent-buildimage-coreos would work | 22:15 |
| *** Ryan_Lane has joined #openstack-infra | 22:15 | |
| clarkb | then you can keep gate-blah blah that just does a build and makes suere you don't break it | 22:15 |
| clarkb | but omits the publish step | 22:15 |
| openstackgerrit | Jay Faulkner proposed a change to openstack-infra/config: Add Ironic Python Agent builder https://review.openstack.org/101654 | 22:15 |
| *** Sukhdev has quit IRC | 22:15 | |
| JayF | I'm very OK not gating on this, especially to start | 22:15 |
| clarkb | ok | 22:15 |
| JayF | ^ is updated per your comments | 22:16 |
| anteaya | why are google docs being called wiki pages: http://www.openstack.org/blog/2014/06/6102/ | 22:16 |
| JayF | although I'll want to work very quickly on getting more validation in that build if we're going to blindly overwrite the image each time | 22:16 |
| anteaya | and why are we being asked to use google docs? https://docs.google.com/a/openstack.org/spreadsheet/ccc?key=0ArUDleqDPUTAdHFNdGx0X1dyVUtLMWRYVU4wdGgtTUE&usp=sharing#gid=8 | 22:16 |
| openstackgerrit | A change was merged to openstack-infra/jenkins-job-builder: Consistency updates in documentation https://review.openstack.org/101394 | 22:16 |
| clarkb | JayF: yup sounds good, I think adding a gate job would be the next step | 22:16 |
| clarkb | JayF: your echo keeps the NEWREV var in the echo | 22:17 |
| JayF | clarkb: then that should be ready for review. What's the best way to test it once it's in? I'd prefer not doing a dummy commit but can if needed | 22:17 |
| JayF | gah | 22:17 |
| clarkb | dummy commit would be most direct way | 22:17 |
| openstackgerrit | Jay Faulkner proposed a change to openstack-infra/config: Add Ironic Python Agent builder https://review.openstack.org/101654 | 22:17 |
| clarkb | otherwise we have to construct a job on the gearman queue directly for eg the last change that merged | 22:17 |
| clarkb | that looks good +2 | 22:18 |
| JayF | Can I get another +2 and +A from anyone on ==> https://review.openstack.org/#/c/101654/5 | 22:18 |
| *** _nadya_ has joined #openstack-infra | 22:18 | |
| jeblair | i guess that would be me : | 22:19 |
| jeblair | JayF: done | 22:21 |
| anteaya | reed: when did flaper87|afk become marconi's ptl? and why isn't that change in the goverance repo? http://git.openstack.org/cgit/openstack/governance/tree/reference/programs.yaml#n317 | 22:22 |
| reed | anteaya, argh, I forgot to go doublecheck | 22:22 |
| reed | dang it | 22:22 |
| *** yamahata has joined #openstack-infra | 22:22 | |
| JayF | looks like gate-config-layout failed on patchset 3.... hmmmm | 22:22 |
| JayF | I hope it passes on 5 :) | 22:22 |
| *** e0ne has quit IRC | 22:23 | |
| *** e0ne has joined #openstack-infra | 22:23 | |
| anteaya | reed: why is a google doc being referenced as a wiki? | 22:24 |
| JayF | hmm https://review.openstack.org/#/c/101654/5 from gate-config-layout | 22:24 |
| reed | anteaya, the original points there | 22:24 |
| JayF | did we miss a spot? | 22:24 |
| anteaya | http://www.openstack.org/blog/2014/06/6102/ | 22:24 |
| anteaya | the original what? | 22:24 |
| anteaya | the Global Events Calendar wiki links to a google doc | 22:25 |
| clarkb | JayF make sure the job name matches in both files | 22:25 |
| reed | anteaya, exactly, the original message wrote by Allison, not by me | 22:26 |
| JayF | name: 'ironic-python-agent-buildimage-coreos' - ironic-python-agent-buildimage-coreos | 22:26 |
| JayF | aha | 22:26 |
| reed | I won't go and edit someone else's post, moving their stuff to other places only because I don't like them | 22:26 |
| JayF | clarkb: ^ should it not be quoted in the jenkins_job_builder? | 22:26 |
| anteaya | why are we advocating that openstack use google docs | 22:26 |
| clarkb | oh i know | 22:26 |
| clarkb | jayf: change job-template to job | 22:26 |
| *** e0ne has quit IRC | 22:26 | |
| anteaya | do you have suggestions for blog contributions that advocate using open source software | 22:27 |
| openstackgerrit | Darragh Bailey proposed a change to openstack-infra/jenkins-job-builder: Use yaml local tags to support including files https://review.openstack.org/48783 | 22:27 |
| openstackgerrit | Darragh Bailey proposed a change to openstack-infra/jenkins-job-builder: Support lazy resolving of include yaml tags https://review.openstack.org/63580 | 22:27 |
| *** jdahlin has joined #openstack-infra | 22:27 | |
| anteaya | and avoiding proprietary things? | 22:27 |
| reed | anteaya, I'm not, but feel free to email allison about it | 22:27 |
| openstackgerrit | Jay Faulkner proposed a change to openstack-infra/config: Add Ironic Python Agent builder https://review.openstack.org/101654 | 22:27 |
| clarkb | in this case you dont need a template and templates require instantiation | 22:27 |
| anteaya | I don't think that is my place | 22:27 |
| JayF | clarkb: jeblair ^ fixed the test failure (I hope) | 22:27 |
| jdahlin | hi, how do you guys get colored messages and custom links as gerrit comments from jenkins? | 22:27 |
| anteaya | jdahlin: what is the context for your question | 22:28 |
| reed | anteaya, there is limited time, I have to pick the battles that can be won, the battle against google docs cannot be won at the moment with free software | 22:28 |
| anteaya | are you seeing something and checking to see if others see the same | 22:28 |
| clarkb | +2 that should do it | 22:28 |
| jdahlin | anteaya: looking at the jenkins approved comment for https://review.openstack.org/#/c/101633/ | 22:28 |
| anteaya | reed: I'm really disappointed to hear you say that | 22:28 |
| jdahlin | anteaya: I can't figure out how to do that on my gerrit installation | 22:28 |
| clarkb | jgahlin its a commentlinks gerrit config | 22:29 |
| jdahlin | clarkb: ah! that helps, thanks | 22:29 |
| clarkb | you can find the magic in our puppet repo modules/openstack_project/manifests/gerrit.pp iirc | 22:29 |
| jdahlin | yup, got it | 22:30 |
| reed | anteaya, I'm more disappointed that the free sw community gave up on software as a service and focused on nitpicking stuff of limited interest, more | 22:30 |
| jdahlin | clarkb: do you know where the gerrit approve call is done as well? | 22:30 |
| jdahlin | (or gerrit review, as it is in newer versions) | 22:30 |
| JayF | jeblair: can I get your +2 again on https://review.openstack.org/101654 ? We had to fix a single test failure, and it's passed now | 22:31 |
| *** mestery has joined #openstack-infra | 22:31 | |
| *** ramashri has quit IRC | 22:31 | |
| *** mestery has quit IRC | 22:31 | |
| anteaya | reed how do you mean? | 22:32 |
| reed | anteaya, I'm glad we, as a community, are addressing another battle that seemed to be lost though, storyboard is making me very happy | 22:32 |
| *** mestery has joined #openstack-infra | 22:32 | |
| clarkb | jdahlin that happens in zuul | 22:32 |
| anteaya | what is the incentive to act, if we jsut throw up our hands and say "oh too hard" | 22:32 |
| clarkb | it uses the gerrit ssh api | 22:32 |
| JayF | We have hackday demos in ~30m and I'd love to be able to demo https://review.openstack.org/101654 working :D | 22:33 |
| clarkb | I can approve it | 22:33 |
| reed | anteaya, it would too long as a discussion ... next time we meet in person | 22:33 |
| JayF | clarkb++ | 22:33 |
| JoshNang | \o/ | 22:33 |
| clarkb | it was basically the same as what jeblair approved | 22:34 |
| *** andreykurilin__ has quit IRC | 22:34 | |
| *** ramashri has joined #openstack-infra | 22:34 | |
| anteaya | reed well it seems to me that we are just posing as an open source software organizition | 22:34 |
| anteaya | because anytime it isn't conveinet for us, we use proprietary things | 22:34 |
| anteaya | and encourage and expect others to do the same | 22:35 |
| reed | anteaya, there are some people without the appropriate training (habits) to use different tools... etherpads are objectively... primitive | 22:35 |
| anteaya | so we aren't open source software, we are open source (when we can) | 22:35 |
| *** thedodd has quit IRC | 22:36 | |
| anteaya | what is wrong with etherpads? | 22:36 |
| reed | anteaya, do you really want me to start ranting? :) | 22:36 |
| anteaya | if you don't like them, people take their cue from you | 22:36 |
| anteaya | you are the community manager | 22:36 |
| reed | I use them, I tend not to use gdocs | 22:36 |
| anteaya | people are going to do what you do | 22:36 |
| reed | unless I really have to (like when I have to interact with rich text documents, with annotations and stuff) | 22:36 |
| anteaya | you are the first point of contact for most new contributors | 22:37 |
| reed | anteaya, you're confusing the blog with me... Allison and the marketing team uses Gdocs, not me | 22:37 |
| anteaya | what you say and do affects the shape of the community a lot | 22:37 |
| anteaya | right | 22:37 |
| anteaya | but you are in a place to say something | 22:37 |
| anteaya | to have a standard and communicate it | 22:37 |
| reed | and I'm not advocating using Gdocs, I've never done that (to my memory) | 22:37 |
| anteaya | are you helping people to learn that is not the expected way of doing things in open source? | 22:38 |
| anteaya | a lot of the thrid party folks have no idea | 22:38 |
| anteaya | so I have to tell them | 22:38 |
| anteaya | and give them information about open source | 22:38 |
| *** Sam_S has quit IRC | 22:38 | |
| anteaya | and expected ettiquette | 22:38 |
| reed | I don't manage the Foundation's marketing team and I don't have the authority to tell them what tools to use... I can ask them to interact with me using etherpads, and I do | 22:38 |
| anteaya | and yes, it takes a lot of my energy | 22:38 |
| openstackgerrit | A change was merged to openstack-infra/config: Add Ironic Python Agent builder https://review.openstack.org/101654 | 22:39 |
| anteaya | can you let them know that using proprietary software in an open source setting is discouraged? | 22:39 |
| JayF | clarkb: ^ will that take a restart or anything to take effect, or does it just go immediately? | 22:39 |
| reed | I sure can, and I tell them when I notice... in this specific case, I didn't notice | 22:40 |
| anteaya | thanks | 22:40 |
| anteaya | but you are aware of it now? | 22:40 |
| anteaya | can you talk to them about their google doc wiki? | 22:40 |
| anteaya | and how we have lots of space on wiki.openstack.org | 22:40 |
| reed | that's what I meant, the link to google docs | 22:41 |
| anteaya | right | 22:41 |
| *** morganfainberg is now known as morganfainberg_Z | 22:41 | |
| anteaya | can you ask them to move the information to wiki.openstack.org | 22:41 |
| clarkb | jayf takes a puppet cycles | 22:41 |
| anteaya | the link to the google doc can have a link to the wiki page at openstack.org | 22:42 |
| JayF | clarkb: is there a way for me to know when that happens? | 22:42 |
| reed | anteaya, ah, it's even a spreadsheet ... | 22:42 |
| anteaya | thanks for fixing the operator's newsletter, re: the ptl | 22:43 |
| anteaya | reed: it doesnt' need to be | 22:43 |
| reed | see, that's even more of a problem... | 22:43 |
| anteaya | it is a list of information | 22:43 |
| *** ArxCruz has joined #openstack-infra | 22:43 | |
| *** ArxCruz has quit IRC | 22:43 | |
| anteaya | it doesn't need to be a spreadsheet | 22:43 |
| *** david-lyle has quit IRC | 22:44 | |
| *** ramashri has quit IRC | 22:45 | |
| clarkb | JayF puppetboard.openstack.org and look for zuul there | 22:45 |
| clarkb | or on status.o.o/zuul at the bottom of page is a last config reloaded timestamp | 22:46 |
| *** david-lyle has joined #openstack-infra | 22:46 | |
| JayF | hah we missed the last puppet run by two minutes | 22:49 |
| reed | anteaya, so you know where my heart is: http://maffulli.net/2014/06/20/engage-in-technical-discussions-keeping-in-mind-the-openstack-promise/ | 22:50 |
| *** david-lyle has quit IRC | 22:50 | |
| reed | and where my priorities are | 22:50 |
| *** ArxCruz has joined #openstack-infra | 22:51 | |
| *** ArxCruz has quit IRC | 22:51 | |
| anteaya | reed: great | 22:52 |
| zaro | pelix: around? | 22:52 |
| anteaya | we are in agreement | 22:52 |
| anteaya | now to implementation | 22:52 |
| *** penguinRaider has quit IRC | 22:53 | |
| *** e0ne has joined #openstack-infra | 22:54 | |
| *** ramashri has joined #openstack-infra | 22:55 | |
| *** ianw has quit IRC | 22:58 | |
| *** melwitt1 has quit IRC | 22:58 | |
| openstackgerrit | Khai Do proposed a change to openstack-infra/jenkins-job-builder: tests for JJB command processing https://review.openstack.org/97908 | 22:59 |
| JayF | Hmm. https://review.openstack.org/#/c/101664/ was merged /after/ Zuul+puppet showed updated configs, but it doesn't appear to have run the post job | 22:59 |
| JayF | clarkb: ^ any suggestions on troubleshooting this? | 23:01 |
| *** CaptTofu_ has joined #openstack-infra | 23:02 | |
| clarkb | go to logs.openstack.org/first2charsofsha1/sha1 | 23:04 |
| clarkb | and see if the job was logged | 23:04 |
| JayF | clarkb: http://logs.openstack.org/b4/b43c8c4c4e419abcd412707ac46f827f69fedf90 is a 404 | 23:06 |
| JayF | clarkb: is that change-id and not git sha1? | 23:06 |
| *** otherwiseguy has quit IRC | 23:06 | |
| clarkb | its sha1 but of any resulting merge commits | 23:06 |
| *** e0ne has quit IRC | 23:07 | |
| clarkb | so check git log after merge | 23:07 |
| *** otherwiseguy has joined #openstack-infra | 23:07 | |
| JayF | the HEAD commit on my git log of ipa, post-merge, is b43c8c4c4e419abcd412707ac46f827f69fedf90 which is the exact commit that I pusehd up to gerrit | 23:07 |
| JayF | there doesn't appear to be a merge commit at all | 23:07 |
| clarkb | hrm. | 23:09 |
| JayF | http://git.openstack.org/cgit/openstack/ironic-python-agent/commit/ shows the same sha1 I see locally | 23:09 |
| JayF | and there's nothing on logs. about it | 23:09 |
| clarkb | currently commuting so hard to debug further. jeblair may have ideas if he is around | 23:09 |
| JayF | jeblair: around? | 23:10 |
| JayF | Anyone else who's willing to help me troubleshoot a new job on IPA? | 23:11 |
| *** Sukhdev has joined #openstack-infra | 23:11 | |
| JayF | clarkb: thanks for trying to help despite commuting. If you have time when you get back at a computer, we can keep going when you're done :) | 23:13 |
| *** hdd_ has joined #openstack-infra | 23:13 | |
| clarkb | ok | 23:14 |
| JayF | clarkb: are you sure it's not /last two/ of the sha1? That's the pattern it follows for gerrit change ids | 23:14 |
| *** reed has quit IRC | 23:14 | |
| JayF | hm that doesn't work either | 23:14 |
| *** nati_ueno has joined #openstack-infra | 23:15 | |
| *** Sukhdev has quit IRC | 23:18 | |
| clarkb | no its first two | 23:18 |
| *** mestery has quit IRC | 23:21 | |
| JayF | I'm merging in another change to see if it maybe Just Works(tm) this time :x | 23:21 |
| *** mbacchi has quit IRC | 23:24 | |
| *** annegent_ has joined #openstack-infra | 23:25 | |
| *** otherwiseguy has quit IRC | 23:26 | |
| *** annegent_ has quit IRC | 23:32 | |
| *** e0ne has joined #openstack-infra | 23:33 | |
| *** zzelle has quit IRC | 23:38 | |
| *** ArxCruz has joined #openstack-infra | 23:38 | |
| *** ArxCruz has quit IRC | 23:38 | |
| JayF | The new merge is in the same state as the original: no logs on logs.openstack.org, no post job ran whatsoever | 23:40 |
| *** lyxus_ has joined #openstack-infra | 23:41 | |
| lyxus_ | sdague: Hi Sean, can I pick your brain on https://review.openstack.org/#/c/101143/ | 23:41 |
| pelix | zaro: just about don't get the notifications at home when someone pings me | 23:42 |
| lyxus_ | sdague: let me know if I did something incorrectly or did not follow a standard | 23:43 |
| anteaya | lyxus_: he might be gone for the day | 23:44 |
| anteaya | I think I remember him saying something about being on dinner duty | 23:44 |
| *** nati_ueno has quit IRC | 23:44 | |
| lyxus_ | anteaya: well, It will wait until monday :) | 23:44 |
| anteaya | sure | 23:44 |
| anteaya | he's up early in the mornings | 23:45 |
| *** morganfainberg_L has joined #openstack-infra | 23:45 | |
| lyxus_ | anteaya: I was just wondering if I did something wrong, I didn’t have that much +1 :) | 23:45 |
| *** ArxCruz has joined #openstack-infra | 23:45 | |
| *** ArxCruz has quit IRC | 23:45 | |
| *** homeless has quit IRC | 23:46 | |
| lyxus_ | anteaya: have a great weekend and see you at the 3rd party meeting. | 23:46 |
| harlowja | qq guys, i am doing a presentation and wanted to include the number of zuul jobs that typically run per day (or commits per day), does anyone have a guestimate? | 23:46 |
| *** mrodden has joined #openstack-infra | 23:47 | |
| *** locke105 has quit IRC | 23:48 | |
| morganfainberg_L | harlowja, a lot ;) [not helpful, sorry] | 23:48 |
| harlowja | ;) | 23:49 |
| harlowja | 3 trillion! | 23:49 |
| fungi | harlowja: http://graphite.openstack.org/render/?from=-30days&height=180&until=now&width=334&bgcolor=ffffff&fgcolor=000000&target=alias(summarize(sumSeries(stats_counts.zuul.pipeline.*.all_jobs),%271d%27),%27All%20Jobs%27)&title=Zuul%20Jobs%20Launched%20(per%20Day)&_t=0.9751067187735285 | 23:49 |
| harlowja | nice | 23:49 |
| harlowja | thx | 23:49 |
| fungi | harlowja: similar tuning of the graph urls at the bottom of http://status.openstack.org/zuul/ will get you similar info for changes merged (off the gerrit events graph) | 23:49 |
| *** gokrokve has quit IRC | 23:50 | |
| harlowja | fungi much appreciated | 23:50 |
| JayF | fungi: do you have a moment to help me troubleshoot a new post-merge job for IPA? | 23:50 |
| fungi | JayF: if it really is a moment, sure (just checked in, still buried under move-in tasks) | 23:50 |
| fungi | JayF: got a url to the failure log? | 23:50 |
| fungi | or a project name and commit sha? | 23:51 |
| JayF | fungi: we just merged https://review.openstack.org/101654 -- and even though puppet has run on zuul and updated the configs | 23:51 |
| JayF | fungi: we've gotten two things merged into ipa since then, and neither ran the post job or has a log at logs.openstack.org/sh/sha1 | 23:51 |
| fungi | JayF: i see, merged about an hour ago | 23:52 |
| JayF | http://logs.openstack.org/8f/8f32811710da943af487670f2591b984a1a5556e should be an example of one that should exist, but doesn't | 23:52 |
| JayF | https://review.openstack.org/101664 + https://review.openstack.org/101667 were merged after the zuul status + puppet showed the merge happened | 23:52 |
| JayF | but I can't see any evidence it even attempted to run the post job | 23:52 |
| anteaya | lyxus_: yeah, I don't see that you have done anything wrong | 23:53 |
| morganfainberg_L | fungi, anteaya, what is the usual timeline on new projects being added? Have had people ask about the new keystone repo and want to be able to say "yeah they're working no it" :) | 23:53 |
| fungi | ooh, runs on bare-trusty nodes. i wonder how many of those we have yet | 23:53 |
| anteaya | it is just a matter of getting it in front of the right people when they have 30 seconds to look at it | 23:53 |
| anteaya | fungi is back!! | 23:53 |
| morganfainberg_L | anteaya, works for me, i'll plan to bug people early next week :) | 23:54 |
| fungi | anteaya: i made the mistake of glancing at irc and answering an easy question. broke my silence :/ | 23:54 |
| morganfainberg_L | anteaya, no real rush at this point | 23:54 |
| *** ArxCruz has joined #openstack-infra | 23:54 | |
| *** ArxCruz has quit IRC | 23:54 | |
| morganfainberg_L | fungi, i'll pretend i didn't see it, quick go idle again :) | 23:54 |
| anteaya | fungi: that will teach you | 23:54 |
| anteaya | fungi: I hope the new house is treating you well | 23:54 |
| *** wenlock has quit IRC | 23:54 | |
| JayF | fungi: If you have more important meatspace things to do, I get it :) I can wait for clarkb to get back around | 23:55 |
| anteaya | fungi: we had trusty nodes and then reverted | 23:55 |
| anteaya | something in the gearman zuul interaction got jobs destined for trusty on both trusty and presices | 23:56 |
| anteaya | so rolling back | 23:56 |
| fungi | JayF: i see bare trusty nodes assigned to a couple of jenkins masters and they have your job configured at least as of now from the jjb config, but don't seem to have tried to run it yet | 23:56 |
| anteaya | at least that is what _I_ think is happening | 23:56 |
| JayF | fungi: yeah, that's what it looked like to me. I was trying to find out if there was some setting in zuul I missed to make it run post jobs for my project, but came up short | 23:56 |
| *** david-lyle has joined #openstack-infra | 23:56 | |
| *** bknudson has quit IRC | 23:57 | |
| fungi | JayF: i'm not up on the trusty node status, but if what anteaya says is accurate then you probably need to wait for that to settle out before expecting jobs to run which require one | 23:57 |
| anteaya | JayF: this is a bug from the moving to trusty process, I do belive | 23:57 |
| anteaya | yes | 23:57 |
| anteaya | I think patience is the way forward | 23:57 |
| JayF | clark said it would work because we're running the job only on trusty | 23:58 |
| JayF | but if he's wrong, that's fine, and I'll just wait for the trusty stuff to roll back forward | 23:58 |
| fungi | anteaya: the new house is running me ragged. so many things were half-done and discovered after we moved in. so many errands to run. it's been nonstop all week. no idea how i thought i would find time to do openstack stuff this week. total loss | 23:58 |
| anteaya | :( | 23:59 |
| fungi | JayF: i'll take a quick peek in the zuul log to see if it acted on one of those merge events | 23:59 |
| anteaya | I'm sorry the endpoint is not in the state you expected | 23:59 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!