| *** owalsh_ has quit IRC | 00:00 | |
| *** owalsh- has joined #openstack-infra | 00:00 | |
| *** owalsh has quit IRC | 00:01 | |
| *** owalsh- is now known as owalsh | 00:01 | |
| *** eharney has quit IRC | 00:01 | |
| corvus | how's this look? https://etherpad.openstack.org/p/8gFYuAwxXf | 00:03 |
|---|---|---|
| *** pahuang has quit IRC | 00:04 | |
| clarkb | corvus: lgtm | 00:05 |
| *** heyongli has quit IRC | 00:07 | |
| corvus | oh my, jobs are running on b8 now, but 058 hasn't finished yet. we have a race. | 00:07 |
| *** heyongli has joined #openstack-infra | 00:07 | |
| clarkb | py36 isn't a quick one either, my money is on b8 | 00:08 |
| corvus | yeah, zuul thinks so too | 00:09 |
| corvus | 15m vs 4m | 00:09 |
| *** slaweq has joined #openstack-infra | 00:11 | |
| *** dingyichen has joined #openstack-infra | 00:11 | |
| corvus | there goes b8 | 00:15 |
| *** slaweq has quit IRC | 00:15 | |
| clarkb | 5f has started jobs now too | 00:15 |
| *** jamesmcarthur has joined #openstack-infra | 00:15 | |
| *** yolanda_ has quit IRC | 00:16 | |
| corvus | w00t | 00:16 |
| corvus | woulda liked to see the bounce, but i reckon we sort of did after the reconfig | 00:16 |
| corvus | this is looking good enough i don't think we're going to do a paper bag revert, so i'll send the email now | 00:17 |
| *** heyongli has quit IRC | 00:17 | |
| *** heyongli has joined #openstack-infra | 00:17 | |
| *** jamesmcarthur has quit IRC | 00:20 | |
| *** pahuang has joined #openstack-infra | 00:21 | |
| openstackgerrit | Merged openstack-infra/zuul master: Change test prints to log.info https://review.openstack.org/554058 | 00:24 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul-jobs master: log-inventory: set inventory_file if needed https://review.openstack.org/578234 | 00:25 |
| corvus | looks like even 5f finished before 058 landed | 00:27 |
| *** heyongli has quit IRC | 00:27 | |
| corvus | the post pipeline queue is getting smaller very quickly | 00:27 |
| *** heyongli has joined #openstack-infra | 00:28 | |
| clarkb | seems to be doing its job | 00:28 |
| *** john_studarus_ has joined #openstack-infra | 00:31 | |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul master: executor: add inventory path to the zuul.executor job variable https://review.openstack.org/578235 | 00:36 |
| *** heyongli has quit IRC | 00:37 | |
| *** heyongli has joined #openstack-infra | 00:38 | |
| *** lifeless has quit IRC | 00:38 | |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul master: executor: add inventory_file path to the zuul.executor job variable https://review.openstack.org/578235 | 00:39 |
| johnsom | zuul is giving me a red box with "//api/status:" | 00:39 |
| openstackgerrit | Ian Wienand proposed openstack-infra/zuul master: Export inventory_file from zuul.executor https://review.openstack.org/578237 | 00:40 |
| clarkb | mordred: ^ is this fallout from your js updates? | 00:40 |
| clarkb | johnsom: confirmed it happens to me too | 00:40 |
| *** Shrews has quit IRC | 00:40 | |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul-jobs master: log-inventory: set inventory_file if needed https://review.openstack.org/578234 | 00:40 |
| clarkb | oh ya this will be fallout from those post jobs we have been watching | 00:40 |
| clarkb | since we deploy the built js webpack stuff | 00:41 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul-jobs master: log-inventory: remove inventory_file usage https://review.openstack.org/578238 | 00:41 |
| clarkb | the issue is that extra / I think. https://zuul.openstack.org/api/status works | 00:42 |
| clarkb | oh its literally GETing https://api/status | 00:43 |
| johnsom | grin | 00:43 |
| openstackgerrit | Paul Belanger proposed openstack-infra/zuul master: WIP: Support skip_child_jobs via zuul_return https://review.openstack.org/578230 | 00:44 |
| *** lifeless has joined #openstack-infra | 00:45 | |
| *** pcichy has quit IRC | 00:45 | |
| *** pcichy has joined #openstack-infra | 00:45 | |
| *** Shrews has joined #openstack-infra | 00:46 | |
| clarkb | ok you can build in config which I assume wedon't do for the generic builds | 00:46 |
| corvus | it's not supposed to require config | 00:46 |
| clarkb | params['source'] = url.searchParams.get('source_url') + '/' + 'status' is another option but we don't have a source_url parameter | 00:46 |
| clarkb | otherwise params['source'] = zuulService.getSourceUrl('status', tenant is how it would load the url I think | 00:47 |
| clarkb | return `${this.baseHref}/api/${filename}` is what we shoul dreturn from getSourceUrl in a whitelabled setup like ours | 00:47 |
| corvus | clarkb: i think this warrants an emergency revert unless you're working on an emergency fix? | 00:47 |
| clarkb | corvus: I was trying to figure out if I could fix it for emergency fix but my thoughts above are as far as I'v egotten | 00:48 |
| *** heyongli has quit IRC | 00:48 | |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul-jobs master: log-inventory: remove inventory_file usage https://review.openstack.org/578238 | 00:48 |
| clarkb | I'm ok with a revert unless maybe mordred or tristanc are more capable of emergency fixing | 00:48 |
| *** heyongli has joined #openstack-infra | 00:48 | |
| corvus | tristanC: ^ thoughts? | 00:48 |
| *** owalsh_ has joined #openstack-infra | 00:49 | |
| *** Dinesh_Bhor has joined #openstack-infra | 00:49 | |
| clarkb | this.baseHref = getBaseHrefFromPath(window.location.pathname) | 00:50 |
| *** owalsh- has joined #openstack-infra | 00:50 | |
| *** pcichy has quit IRC | 00:50 | |
| corvus | yeah, that should be the thing for us | 00:50 |
| *** pcichy has joined #openstack-infra | 00:50 | |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul master: Revert "Hide queue headers for empty queues when filtering" https://review.openstack.org/578241 | 00:52 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul master: Revert "Upgrade from angularjs (v1) to angular (v6)" https://review.openstack.org/578242 | 00:52 |
| clarkb | return path.split('/').slice(0, -1).join('/') + '/' is what that does | 00:52 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul master: executor: add inventory_file path to the zuul.executor job variable https://review.openstack.org/578235 | 00:52 |
| clarkb | I'm wondering if that only sees the path component and not the full url | 00:53 |
| *** owalsh has quit IRC | 00:53 | |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul master: WIP: Don't add implied branch matchers to project-pipeline variants https://review.openstack.org/577881 | 00:54 |
| *** owalsh_ has quit IRC | 00:54 | |
| clarkb | "The pathname property sets or returns the pathname of a URL." ya I think that is the bug | 00:54 |
| *** owalsh has joined #openstack-infra | 00:54 | |
| clarkb | or maybe we intended to do a relative get | 00:54 |
| *** owalsh- has quit IRC | 00:55 | |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul master: Revert "Upgrade from angularjs (v1) to angular (v6)" https://review.openstack.org/578242 | 00:55 |
| clarkb | jquery seems to want a url for getJSON | 00:56 |
| clarkb | I think that is the mixup here but definitely well beyond my knowledge to know for sure | 00:57 |
| *** heyongli has quit IRC | 00:58 | |
| *** heyongli has joined #openstack-infra | 00:58 | |
| *** Shrews has quit IRC | 01:00 | |
| ianw | fungi: thanks ... http://mirror.iad.rax.openstack.org/debian-security/ looks about how i'd expect finaly | 01:02 |
| *** Shrews has joined #openstack-infra | 01:02 | |
| fungi | excellent | 01:03 |
| fungi | at least it shouldn't take long to populate because it just accumulates packages between stable point releases | 01:03 |
| *** pcichy has quit IRC | 01:05 | |
| openstackgerrit | Clark Boylan proposed openstack-infra/zuul master: Use window.location.href for the href not pathname https://review.openstack.org/578243 | 01:06 |
| *** Shrews has quit IRC | 01:08 | |
| openstackgerrit | Paul Belanger proposed openstack-infra/zuul master: WIP: Support skip_child_jobs via zuul_return https://review.openstack.org/578230 | 01:08 |
| *** heyongli has quit IRC | 01:08 | |
| *** heyongli has joined #openstack-infra | 01:08 | |
| *** pcichy has joined #openstack-infra | 01:10 | |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul master: web: fix status page flickering https://review.openstack.org/578226 | 01:11 |
| tristanC | corvus: +1 for a revert, moreover I think we need to squash https://review.openstack.org/578226 | 01:12 |
| openstackgerrit | Paul Belanger proposed openstack-infra/zuul master: WIP: Support skip_child_jobs via zuul_return https://review.openstack.org/578230 | 01:12 |
| tristanC | alternatively, maybe we could stop puppet like we did when the first js refactor also broke the zuul.openstack.org webui? | 01:13 |
| tristanC | and restore the previous html files | 01:14 |
| *** Shrews has joined #openstack-infra | 01:14 | |
| clarkb | tristanC: I think the problem is it is broken for other users too so revert is most appropriate | 01:14 |
| corvus | tristanC: i don't understand what you mean about 578226 | 01:14 |
| clarkb | I have approved the revert and we can clean up the implementation and try again | 01:14 |
| clarkb | I'm going to make dinner now but will check in in a bit to see if revert has done what we need | 01:15 |
| tristanC | corvus: to reproduce the issue fix by 578226, use the zuul-build-dashboard output and click status, then job, then status then jobs repeatedely. Then un focus and focus again, you'll see a multiple status call happening at once. | 01:17 |
| corvus | tristanC: right, but are you saying we should make sure that is included when we re-do the v6 conversion? | 01:18 |
| *** heyongli has quit IRC | 01:18 | |
| tristanC | corvus: yes since this actually also affect single tenant setup | 01:19 |
| *** heyongli has joined #openstack-infra | 01:19 | |
| corvus | tristanC: i agree. seems like we should be able to get that in the un-revert. | 01:20 |
| *** pcichy has quit IRC | 01:20 | |
| *** yamahata has quit IRC | 01:22 | |
| *** jesslampe has joined #openstack-infra | 01:25 | |
| *** caphrim007 has joined #openstack-infra | 01:28 | |
| *** hongbin has joined #openstack-infra | 01:28 | |
| *** xinliang_ has joined #openstack-infra | 01:28 | |
| *** edwarnicke has left #openstack-infra | 01:29 | |
| *** heyongli has quit IRC | 01:29 | |
| *** heyongli has joined #openstack-infra | 01:29 | |
| *** jesslampe has quit IRC | 01:29 | |
| xinliang_ | Hi, anyone knows why the arm64 debian node fails to resize root partition? | 01:30 |
| xinliang_ | here is the growroot log: http://paste.openstack.org/show/724345/ | 01:30 |
| *** caphrim007 has quit IRC | 01:32 | |
| *** gongysh has joined #openstack-infra | 01:32 | |
| xinliang_ | ianw, do you know what's wrong? | 01:33 |
| clarkb | line 7 unsupported command is what Id try to find | 01:34 |
| *** caphrim007 has joined #openstack-infra | 01:34 | |
| fungi | | >>> line 7: unsupported command | 01:35 |
| *** caphrim007_ has joined #openstack-infra | 01:35 | |
| fungi | yeah, just spotted | 01:35 |
| clarkb | I want to say dib provides its own sfdisk script for this | 01:36 |
| ianw | ohhh, hrm, something happened last time we had a grow failure, what was it ... | 01:37 |
| Dinesh_Bhor | Hi All, I am trying to configure devstack multi compute node environment for functional testing on the gate for tacker: https://github.com/openstack/tacker/blob/master/.zuul.yaml | 01:38 |
| ianw | ahh, that was the dodgy label on the partition | 01:38 |
| openstackgerrit | Merged openstack-infra/zuul master: Revert "Upgrade from angularjs (v1) to angular (v6)" https://review.openstack.org/578242 | 01:38 |
| Dinesh_Bhor | Can anyone help to figure out what changes are required in these existing jobs? | 01:38 |
| *** ramishra has joined #openstack-infra | 01:38 | |
| *** heyongli has quit IRC | 01:39 | |
| *** heyongli has joined #openstack-infra | 01:39 | |
| *** caphrim007 has quit IRC | 01:39 | |
| *** dchen has joined #openstack-infra | 01:41 | |
| *** dchen has quit IRC | 01:41 | |
| *** dchen has joined #openstack-infra | 01:42 | |
| *** Shrews has quit IRC | 01:42 | |
| *** pahuang has quit IRC | 01:43 | |
| *** dingyichen has quit IRC | 01:43 | |
| *** Shrews has joined #openstack-infra | 01:44 | |
| *** heyongli has quit IRC | 01:49 | |
| *** heyongli has joined #openstack-infra | 01:49 | |
| *** pcichy has joined #openstack-infra | 01:50 | |
| *** gyee has quit IRC | 01:51 | |
| *** john_studarus_ has quit IRC | 01:51 | |
| *** pahuang has joined #openstack-infra | 01:56 | |
| *** eernst has joined #openstack-infra | 01:59 | |
| *** heyongli has quit IRC | 01:59 | |
| *** heyongli has joined #openstack-infra | 02:00 | |
| *** pcichy_ has joined #openstack-infra | 02:01 | |
| *** eernst has quit IRC | 02:01 | |
| ianw | the "unsupported command" appears to be -> /dev/sda1 : start 2048 size 16384 type=C12A7328-F81F-11D2-BA4B-00A0C93EC93B uuid=65D40C00-00A6-4965-9BC1-0FAC317B03D0 name=""ESP"" | 02:02 |
| *** pcichy has quit IRC | 02:04 | |
| *** pcichy has joined #openstack-infra | 02:05 | |
| ianw | xinliang_: hrrm, so GROWPART_RESIZER=sgdisk works. my appetite for chasing why sfdisk seemingly screws up the "type" value for gpt disks is ... low | 02:07 |
| *** pcichy_ has quit IRC | 02:08 | |
| *** pcichy has quit IRC | 02:09 | |
| *** bobh has quit IRC | 02:10 | |
| *** heyongli has quit IRC | 02:10 | |
| *** heyongli has joined #openstack-infra | 02:10 | |
| *** slaweq has joined #openstack-infra | 02:11 | |
| ianw | xinliang_: are you editing /usr/bin/growpart? | 02:12 |
| xinliang_ | ianw: sorry, not understand what you said.. | 02:12 |
| ianw | do you have that file open in vi on 103.20.251.146? | 02:13 |
| xinliang_ | yes, I have that file. I'm not edit that file | 02:13 |
| xinliang_ | I set -x to that file , this is the "growpart /dev/sda 3" output: http://paste.openstack.org/show/724348/ | 02:14 |
| ianw | sfdisk --unit=S --dump /dev/sda | 02:15 |
| ianw | http://paste.openstack.org/show/724349/ | 02:15 |
| ianw | those types look totally whack | 02:15 |
| *** slaweq has quit IRC | 02:16 | |
| *** gongysh has quit IRC | 02:16 | |
| xinliang_ | ianw: I'm not modify growpart, except adding set -x | 02:18 |
| ianw | yeah, sfdisk can't restore it's own dump | 02:18 |
| xinliang_ | ianw: so do you know how to fix this issue? | 02:19 |
| *** heyongli has quit IRC | 02:20 | |
| *** heyongli has joined #openstack-infra | 02:20 | |
| clarkb | Dinesh_Bhor: do you have specific concerns or problems? | 02:21 |
| ianw | xinliang_: well, i have some thoughts ... let me see if we can shoe-horn a later sfdisk on there and see if it works. that will tell us if there's a fixed bug | 02:23 |
| xinliang_ | ianw: ok, please | 02:23 |
| Dinesh_Bhor | clarkb: https://review.openstack.org/#/c/578170/ This is a trial patch but looks like it is failed to create the multi-node. | 02:23 |
| *** sthussey has quit IRC | 02:24 | |
| xinliang_ | ianw: so this problem maybe "'type' value for gpt disks is low", right? | 02:25 |
| *** tpatil has joined #openstack-infra | 02:25 | |
| clarkb | Dinesh_Bhor: the logs show it ran three nodes but that devstack failed on the two compute hosts | 02:27 |
| clarkb | Dinesh_Bhor: to start my recommendation is you do two nodes, controller and compute to simplify things | 02:28 |
| clarkb | then you can debug a single failing devstack run on the compute | 02:28 |
| *** heyongli has quit IRC | 02:30 | |
| clarkb | Dinesh_Bhor: http://logs.openstack.org/70/578170/2/check/tacker-functional-devstack-multinode/905f024/compute1/logs/devstacklog.txt.gz#_2018-06-27_01_02_40_172 is the error on the one compute host I'm assuming that is a config bug but may be a devstack issue | 02:30 |
| *** heyongli has joined #openstack-infra | 02:31 | |
| ianw | xinliang_: well it appears to put in a whole uuid thing, not a partition type? | 02:32 |
| clarkb | Dinesh_Bhor: appears to be attempting to use legacy neutron which mustve cleaned out the mutnauq code? | 02:33 |
| Dinesh_Bhor | clarkb: I see that function here: http://git.openstack.org/cgit/openstack-dev/devstack/tree/lib/neutron-legacy#n440 | 02:34 |
| *** Shrews has quit IRC | 02:34 | |
| clarkb | ya but isnt in lib/neutron but ut does attempt to source it earlier on | 02:35 |
| *** gongysh has joined #openstack-infra | 02:35 | |
| clarkb | if I had to guess a plugin enables q-something after that check so the function isnt loaded when it tries to run it | 02:35 |
| clarkb | q- is definitely not enabled in localrc for that node | 02:37 |
| *** psachin has joined #openstack-infra | 02:38 | |
| clarkb | but is enabled in localrc for the co troller which is why it is happy I expect | 02:38 |
| *** heyongli has quit IRC | 02:40 | |
| *** Shrews has joined #openstack-infra | 02:41 | |
| *** heyongli has joined #openstack-infra | 02:41 | |
| Dinesh_Bhor | clarkb: okay, Thanks for digging into it. I will check this part. | 02:41 |
| *** annp has joined #openstack-infra | 02:50 | |
| *** heyongli has quit IRC | 02:51 | |
| *** heyongli has joined #openstack-infra | 02:51 | |
| *** heyongli has quit IRC | 03:01 | |
| *** heyongli has joined #openstack-infra | 03:01 | |
| *** tpatil has quit IRC | 03:04 | |
| *** noama has joined #openstack-infra | 03:06 | |
| ianw | xinliang_: hmm, think i found the issue in the quoting ... | 03:07 |
| xinliang_ | really, great! | 03:08 |
| openstackgerrit | Xinliang Liu proposed openstack-infra/zuul-jobs master: Use debian-security mirror https://review.openstack.org/578259 | 03:09 |
| *** hongbin has quit IRC | 03:11 | |
| ianw | Partition name: '"ESP"' ... i think this is half our bug, and half util-linux's bug for not handling the quotes | 03:11 |
| *** heyongli has quit IRC | 03:11 | |
| *** heyongli has joined #openstack-infra | 03:12 | |
| *** roman_g has quit IRC | 03:18 | |
| prometheanfire | ´`'"ESP"'`´ | 03:19 |
| prometheanfire | just in case you needed some more | 03:19 |
| *** heyongli has quit IRC | 03:21 | |
| *** heyongli has joined #openstack-infra | 03:22 | |
| openstackgerrit | Merged openstack-dev/pbr master: Support python-requires metadata https://review.openstack.org/571795 | 03:23 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul master: executor: add inventory_file path to the zuul.executor job variable https://review.openstack.org/578235 | 03:29 |
| ianw | prometheanfire: i say we call our primary partition "☃" and just screw everything up :) | 03:29 |
| prometheanfire | ianw: sgtm | 03:31 |
| prometheanfire | can you do that with gpt (unicode partition names)? | 03:31 |
| *** heyongli has quit IRC | 03:32 | |
| prometheanfire | ¨ | 03:32 |
| *** dklyle has quit IRC | 03:32 | |
| ianw | i'm guessing you *could*, but seeing as sfdisk can't handle a quote, i'm also guessing it would not end at all well :) | 03:32 |
| *** heyongli has joined #openstack-infra | 03:32 | |
| prometheanfire | forgot that one, but that may be an umlaut, altgr | 03:33 |
| prometheanfire | I always just script out parted | 03:33 |
| ianw | pabelanger: appreciate it if you could confirm on https://review.openstack.org/578259 which sets up debian-security | 03:33 |
| *** hongbin has joined #openstack-infra | 03:37 | |
| ianw | damn it, of course it goes right to heart of how we call things in dib :/ | 03:38 |
| *** heyongli has quit IRC | 03:42 | |
| *** heyongli has joined #openstack-infra | 03:42 | |
| *** hamzy has joined #openstack-infra | 03:47 | |
| *** lpetrut has joined #openstack-infra | 03:47 | |
| *** udesale has joined #openstack-infra | 03:50 | |
| *** noama has quit IRC | 03:50 | |
| *** agopi has quit IRC | 03:50 | |
| *** heyongli has quit IRC | 03:52 | |
| *** heyongli has joined #openstack-infra | 03:52 | |
| *** heyongli has quit IRC | 04:02 | |
| *** heyongli has joined #openstack-infra | 04:03 | |
| openstackgerrit | Ian Wienand proposed openstack/diskimage-builder master: Don't quote names with sgdisk https://review.openstack.org/578265 | 04:04 |
| *** bhavik1 has joined #openstack-infra | 04:05 | |
| *** gongysh has quit IRC | 04:06 | |
| *** larainema has quit IRC | 04:08 | |
| ianw | xinliang_: ^ and filed https://github.com/karelzak/util-linux/issues/656 | 04:08 |
| *** dklyle has joined #openstack-infra | 04:11 | |
| *** Shrews has quit IRC | 04:13 | |
| *** heyongli has quit IRC | 04:13 | |
| *** heyongli has joined #openstack-infra | 04:13 | |
| *** Shrews has joined #openstack-infra | 04:14 | |
| *** gongysh has joined #openstack-infra | 04:15 | |
| *** sree_ has joined #openstack-infra | 04:18 | |
| *** sree_ is now known as Guest13850 | 04:19 | |
| *** bhavik1 has quit IRC | 04:22 | |
| *** Bhujay has joined #openstack-infra | 04:22 | |
| *** dhajare has joined #openstack-infra | 04:23 | |
| *** heyongli has quit IRC | 04:23 | |
| *** heyongli has joined #openstack-infra | 04:23 | |
| *** hongbin has quit IRC | 04:24 | |
| *** pahuang has quit IRC | 04:24 | |
| *** Bhujay has quit IRC | 04:24 | |
| *** Bhujay has joined #openstack-infra | 04:25 | |
| *** Bhujay has quit IRC | 04:26 | |
| openstackgerrit | Merged openstack-infra/zuul master: Don't reconfigure tenant on unprotected branch pushes https://review.openstack.org/576538 | 04:26 |
| openstackgerrit | Merged openstack-infra/zuul master: Fix branch parsing in _event_pull https://review.openstack.org/577371 | 04:26 |
| *** Bhujay has joined #openstack-infra | 04:27 | |
| *** ykarel has joined #openstack-infra | 04:27 | |
| *** viks_ has joined #openstack-infra | 04:27 | |
| viks_ | hi, i'm trying to run tempest only for keystone and swift using rally. But it does not succeed. it's giving the following error: 'ERROR rally.task.context EndpointNotFound: public endpoint for image service not found' | 04:27 |
| viks_ | also the following error: | 04:27 |
| viks_ | " ERROR rally NoOptionError: No option 'img_dir' in section: 'scenario'" | 04:27 |
| viks_ | I do have only swift setup along with keystone. Not able to figure out why it is trying to check image service endpoint and related settings. Can anyone please help me out? | 04:27 |
| *** lpetrut has quit IRC | 04:27 | |
| *** heyongli has quit IRC | 04:33 | |
| *** heyongli has joined #openstack-infra | 04:33 | |
| *** jesslampe has joined #openstack-infra | 04:34 | |
| *** hwoarang has quit IRC | 04:40 | |
| *** pahuang has joined #openstack-infra | 04:41 | |
| *** psachin has quit IRC | 04:41 | |
| *** heyongli has quit IRC | 04:43 | |
| *** noama has joined #openstack-infra | 04:44 | |
| *** heyongli has joined #openstack-infra | 04:44 | |
| *** dklyle has quit IRC | 04:52 | |
| *** Shrews has quit IRC | 04:53 | |
| *** heyongli has quit IRC | 04:54 | |
| *** heyongli has joined #openstack-infra | 04:54 | |
| *** Shrews has joined #openstack-infra | 04:55 | |
| *** dklyle has joined #openstack-infra | 04:57 | |
| *** gongysh has quit IRC | 04:59 | |
| *** Bhujay has quit IRC | 05:00 | |
| *** Bhujay has joined #openstack-infra | 05:00 | |
| *** Bhujay has quit IRC | 05:01 | |
| *** Bhujay has joined #openstack-infra | 05:02 | |
| *** Bhujay has quit IRC | 05:03 | |
| *** heyongli has quit IRC | 05:04 | |
| *** heyongli has joined #openstack-infra | 05:04 | |
| *** iranzo has joined #openstack-infra | 05:13 | |
| *** heyongli has quit IRC | 05:14 | |
| *** heyongli has joined #openstack-infra | 05:15 | |
| *** dhajare has quit IRC | 05:15 | |
| *** tdasilva has quit IRC | 05:21 | |
| *** hwoarang has joined #openstack-infra | 05:23 | |
| *** janki has joined #openstack-infra | 05:24 | |
| *** heyongli has quit IRC | 05:24 | |
| *** heyongli has joined #openstack-infra | 05:25 | |
| *** tdasilva has joined #openstack-infra | 05:33 | |
| *** heyongli has quit IRC | 05:35 | |
| *** heyongli has joined #openstack-infra | 05:35 | |
| *** e0ne has joined #openstack-infra | 05:35 | |
| *** quiquell|off is now known as quiquell|rover | 05:36 | |
| *** gongysh has joined #openstack-infra | 05:36 | |
| *** e0ne has quit IRC | 05:36 | |
| *** heyongli has quit IRC | 05:45 | |
| *** heyongli has joined #openstack-infra | 05:45 | |
| *** Shrews has quit IRC | 05:46 | |
| *** yolanda_ has joined #openstack-infra | 05:47 | |
| *** Shrews has joined #openstack-infra | 05:48 | |
| *** dhajare has joined #openstack-infra | 05:51 | |
| *** pahuang has quit IRC | 05:53 | |
| *** heyongli has quit IRC | 05:55 | |
| *** heyongli has joined #openstack-infra | 05:55 | |
| *** heyongli has quit IRC | 06:05 | |
| *** hwoarang has quit IRC | 06:06 | |
| *** yolanda__ has joined #openstack-infra | 06:06 | |
| *** heyongli has joined #openstack-infra | 06:06 | |
| *** pahuang has joined #openstack-infra | 06:06 | |
| *** hwoarang has joined #openstack-infra | 06:06 | |
| xinliang_ | ianw: thanks | 06:06 |
| *** andreas_s has joined #openstack-infra | 06:08 | |
| *** yolanda_ has quit IRC | 06:09 | |
| AJaeger | ianw: ready to merge https://review.openstack.org/577045 ? Or do you wait for anything with the nodepool grafana graphs? | 06:09 |
| *** bhavik1 has joined #openstack-infra | 06:09 | |
| AJaeger | viks_: for tempest, ask on #openstack-qa, for rally there should be a channel as well... | 06:09 |
| *** yolanda__ has quit IRC | 06:14 | |
| *** heyongli has quit IRC | 06:16 | |
| *** heyongli has joined #openstack-infra | 06:16 | |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool master: openstack: decline request for labels un-available in current pool https://review.openstack.org/578288 | 06:19 |
| *** threestrands has quit IRC | 06:19 | |
| *** ccamacho has joined #openstack-infra | 06:22 | |
| *** lpetrut has joined #openstack-infra | 06:24 | |
| *** larainema has joined #openstack-infra | 06:24 | |
| *** heyongli has quit IRC | 06:26 | |
| *** heyongli has joined #openstack-infra | 06:26 | |
| *** Bhujay has joined #openstack-infra | 06:27 | |
| *** namnh has joined #openstack-infra | 06:28 | |
| *** hashar has joined #openstack-infra | 06:28 | |
| *** dmellado has joined #openstack-infra | 06:29 | |
| *** yolanda__ has joined #openstack-infra | 06:30 | |
| *** jcoufal has joined #openstack-infra | 06:32 | |
| *** jcoufal_ has joined #openstack-infra | 06:36 | |
| *** heyongli has quit IRC | 06:36 | |
| *** heyongli has joined #openstack-infra | 06:37 | |
| *** Bhujay has quit IRC | 06:38 | |
| *** jcoufal has quit IRC | 06:39 | |
| *** nicolasbock has joined #openstack-infra | 06:40 | |
| *** quiquell|rover is now known as quique|rover|afk | 06:42 | |
| viks_ | @Ajaeger ok.. thanks | 06:43 |
| *** pcaruana has joined #openstack-infra | 06:44 | |
| *** heyongli has quit IRC | 06:46 | |
| *** heyongli has joined #openstack-infra | 06:47 | |
| *** jesslampe has quit IRC | 06:48 | |
| *** yolanda__ has quit IRC | 06:50 | |
| *** egonzalez has joined #openstack-infra | 06:51 | |
| *** Hal has joined #openstack-infra | 06:52 | |
| *** Hal is now known as Guest44976 | 06:53 | |
| *** yolanda__ has joined #openstack-infra | 06:53 | |
| *** yolanda__ is now known as yolanda | 06:53 | |
| *** andreas_s has quit IRC | 06:53 | |
| *** alexchadin has joined #openstack-infra | 06:54 | |
| *** udesale has quit IRC | 06:54 | |
| *** udesale has joined #openstack-infra | 06:55 | |
| *** heyongli has quit IRC | 06:57 | |
| *** heyongli has joined #openstack-infra | 06:57 | |
| *** udesale has quit IRC | 06:59 | |
| *** udesale has joined #openstack-infra | 07:00 | |
| *** andreas_s has joined #openstack-infra | 07:01 | |
| *** pcaruana has quit IRC | 07:02 | |
| *** peereb has joined #openstack-infra | 07:03 | |
| *** slaweq has joined #openstack-infra | 07:03 | |
| *** dpawlik has joined #openstack-infra | 07:03 | |
| *** peereb has quit IRC | 07:04 | |
| *** peereb has joined #openstack-infra | 07:04 | |
| *** Guest13850 has quit IRC | 07:05 | |
| *** peereb has quit IRC | 07:05 | |
| *** peereb has joined #openstack-infra | 07:06 | |
| *** andreas_s has quit IRC | 07:07 | |
| *** heyongli has quit IRC | 07:07 | |
| *** heyongli has joined #openstack-infra | 07:07 | |
| *** tesseract has joined #openstack-infra | 07:07 | |
| *** quique|rover|afk is now known as quiquell|rover | 07:08 | |
| *** andreas_s has joined #openstack-infra | 07:11 | |
| *** lpetrut has quit IRC | 07:12 | |
| *** heyongli has quit IRC | 07:17 | |
| *** heyongli has joined #openstack-infra | 07:18 | |
| *** andreas_s has quit IRC | 07:18 | |
| *** rpittau has quit IRC | 07:19 | |
| *** andreas_s has joined #openstack-infra | 07:19 | |
| quiquell|rover | Hello | 07:19 |
| quiquell|rover | Is there a way to know the cloud provider for a zuul's build ? | 07:19 |
| *** rpittau has joined #openstack-infra | 07:21 | |
| *** andreas_s has quit IRC | 07:21 | |
| *** jaosorior has quit IRC | 07:23 | |
| *** andreas_s has joined #openstack-infra | 07:25 | |
| AJaeger | quiquell|rover: sure - check zuul-info/inventory.yaml in your logs, e.g. http://logs.openstack.org/88/578288/1/check/tox-pep8/bf27ab6/zuul-info/inventory.yaml | 07:25 |
| AJaeger | quiquell|rover: that one run in Rackspace ("rax") in the "rax-ord" data center | 07:26 |
| quiquell|rover | AJaeger: Also found t he job-output.json.gz has the nodename | 07:26 |
| AJaeger | quiquell|rover: yeah, it's in many places... | 07:26 |
| quiquell|rover | AJaeger: Thanks a lot, inventory is the place to go | 07:27 |
| quiquell|rover | AJaeger: I see that zuuls /builds API has node_name: null, do you know why ? | 07:27 |
| *** heyongli has quit IRC | 07:27 | |
| *** heyongli has joined #openstack-infra | 07:28 | |
| *** shardy has quit IRC | 07:28 | |
| *** zoli is now known as zoli|wfh | 07:28 | |
| *** zoli|wfh is now known as zoli | 07:28 | |
| *** Shrews has quit IRC | 07:29 | |
| *** shardy has joined #openstack-infra | 07:31 | |
| *** Shrews has joined #openstack-infra | 07:31 | |
| *** florianf|afk is now known as florianf | 07:31 | |
| AJaeger | quiquell|rover: can't help with that | 07:34 |
| quiquell|rover | AJaeger: We always have a primary and secondarfy ? | 07:35 |
| *** alexchadin has quit IRC | 07:35 | |
| quiquell|rover | AJaeger: Nah forget about last question | 07:35 |
| AJaeger | ;) | 07:35 |
| *** heyongli has quit IRC | 07:38 | |
| *** heyongli has joined #openstack-infra | 07:38 | |
| *** tosky has joined #openstack-infra | 07:40 | |
| *** amoralej|off is now known as amoralej | 07:42 | |
| *** andreas_s has quit IRC | 07:43 | |
| *** alexchadin has joined #openstack-infra | 07:43 | |
| yamamoto | build-openstack-sphinx-docs doesn't use tox-siblings? | 07:45 |
| *** jpena|off is now known as jpena | 07:46 | |
| yamamoto | and thus released versions from pypi are used for dependencies? (vs master) | 07:46 |
| *** andreas_s has joined #openstack-infra | 07:47 | |
| *** heyongli has quit IRC | 07:48 | |
| *** heyongli has joined #openstack-infra | 07:48 | |
| *** rcernin has quit IRC | 07:51 | |
| *** e0ne has joined #openstack-infra | 07:51 | |
| *** andreas_s has quit IRC | 07:51 | |
| *** andreas_s has joined #openstack-infra | 07:52 | |
| *** rcernin has joined #openstack-infra | 07:52 | |
| *** noama has quit IRC | 07:53 | |
| *** andreas_s has quit IRC | 07:56 | |
| *** pcaruana has joined #openstack-infra | 07:57 | |
| *** heyongli has quit IRC | 07:58 | |
| *** heyongli has joined #openstack-infra | 07:59 | |
| *** jpich has joined #openstack-infra | 08:02 | |
| *** heyongli has quit IRC | 08:08 | |
| *** heyongli has joined #openstack-infra | 08:09 | |
| openstackgerrit | Ian Wienand proposed openstack/diskimage-builder master: Don't quote names with sgdisk https://review.openstack.org/578265 | 08:10 |
| *** andreas_s_ has joined #openstack-infra | 08:15 | |
| *** ykarel is now known as ykarel|lunch | 08:16 | |
| *** s-shiono has quit IRC | 08:17 | |
| *** heyongli has quit IRC | 08:19 | |
| *** heyongli has joined #openstack-infra | 08:19 | |
| *** slaweq has quit IRC | 08:20 | |
| *** ykarel|lunch has quit IRC | 08:20 | |
| *** ykarel|lunch has joined #openstack-infra | 08:21 | |
| *** lpetrut has joined #openstack-infra | 08:21 | |
| *** andreas_s_ has quit IRC | 08:21 | |
| *** andreas_s has joined #openstack-infra | 08:22 | |
| *** dtantsur|afk is now known as dtantsur | 08:22 | |
| *** ccamacho has quit IRC | 08:27 | |
| *** ccamacho has joined #openstack-infra | 08:27 | |
| *** heyongli has quit IRC | 08:29 | |
| *** bhavik1 has quit IRC | 08:29 | |
| *** heyongli has joined #openstack-infra | 08:29 | |
| *** andreas_s has quit IRC | 08:30 | |
| *** andreas_s has joined #openstack-infra | 08:30 | |
| *** jcoufal_ has quit IRC | 08:32 | |
| *** andreas_s has quit IRC | 08:36 | |
| *** andreas_s has joined #openstack-infra | 08:36 | |
| *** slaweq has joined #openstack-infra | 08:36 | |
| *** electrofelix has joined #openstack-infra | 08:38 | |
| *** heyongli has quit IRC | 08:39 | |
| *** heyongli has joined #openstack-infra | 08:39 | |
| *** andreas_s has quit IRC | 08:40 | |
| *** andreas_s has joined #openstack-infra | 08:43 | |
| *** ykarel|lunch is now known as ykarel | 08:45 | |
| *** e0ne has quit IRC | 08:46 | |
| *** rcernin has quit IRC | 08:47 | |
| *** shardy has quit IRC | 08:48 | |
| openstackgerrit | YAMAMOTO Takashi proposed openstack-infra/irc-meetings master: Cancel Tap as a service meeting https://review.openstack.org/578328 | 08:49 |
| *** heyongli has quit IRC | 08:49 | |
| *** roman_g has joined #openstack-infra | 08:50 | |
| *** heyongli has joined #openstack-infra | 08:50 | |
| *** derekh has joined #openstack-infra | 08:52 | |
| *** shardy has joined #openstack-infra | 08:52 | |
| *** d0ugal has quit IRC | 08:54 | |
| *** jamesmcarthur has joined #openstack-infra | 08:55 | |
| *** alexchadin has quit IRC | 08:59 | |
| *** heyongli has quit IRC | 09:00 | |
| *** jamesmcarthur has quit IRC | 09:00 | |
| *** heyongli has joined #openstack-infra | 09:00 | |
| *** gfidente has joined #openstack-infra | 09:03 | |
| *** gfidente has joined #openstack-infra | 09:03 | |
| *** dchen has quit IRC | 09:04 | |
| quiquell|rover | What projects are included at codesearch ? | 09:08 |
| quiquell|rover | I am missing tripleo-upgrades | 09:09 |
| *** heyongli has quit IRC | 09:10 | |
| *** heyongli has joined #openstack-infra | 09:10 | |
| openstackgerrit | Oliver Walsh proposed openstack/diskimage-builder master: WIP: Add DIB element to install NVIDIA GPU drivers https://review.openstack.org/573223 | 09:13 |
| quiquell|rover | Ahh is there | 09:14 |
| *** Dinesh_Bhor has quit IRC | 09:14 | |
| *** ianychoi has quit IRC | 09:14 | |
| *** ianychoi has joined #openstack-infra | 09:15 | |
| *** heyongli has quit IRC | 09:20 | |
| *** heyongli has joined #openstack-infra | 09:20 | |
| *** witek has joined #openstack-infra | 09:25 | |
| witek | hello, could someone please help me with this change https://review.openstack.org/560849 ? | 09:27 |
| witek | it cannot get merged for some reason | 09:27 |
| witek | the gate pipeline passes passes | 09:27 |
| witek | and the status changes to: `Ready to Submit` | 09:28 |
| *** heyongli has quit IRC | 09:30 | |
| *** heyongli has joined #openstack-infra | 09:31 | |
| *** ykarel has quit IRC | 09:35 | |
| *** Bhujay has joined #openstack-infra | 09:38 | |
| *** ykarel has joined #openstack-infra | 09:39 | |
| *** heyongli has quit IRC | 09:41 | |
| *** heyongli has joined #openstack-infra | 09:41 | |
| *** alexchadin has joined #openstack-infra | 09:43 | |
| *** threestrands has joined #openstack-infra | 09:45 | |
| *** threestrands has quit IRC | 09:46 | |
| *** threestrands has joined #openstack-infra | 09:47 | |
| frickler | witek: it needs a manual rebase because it is based on an old patchset of https://review.openstack.org/#/c/563087/ (PS8, while PS9 got merged) | 09:51 |
| *** heyongli has quit IRC | 09:51 | |
| *** heyongli has joined #openstack-infra | 09:51 | |
| openstackgerrit | Merged openstack/diskimage-builder master: Add iscsi-boot element for CentOS images https://review.openstack.org/542708 | 09:54 |
| frickler | infra-root: ^^ this issue came up multiple times now, makes me wonder whether we have something lacking in the way gerrit+zuul handle mergeability checking now | 09:54 |
| frickler | or maybe we just need to document this situation better | 09:54 |
| *** heyongli has quit IRC | 10:01 | |
| *** heyongli has joined #openstack-infra | 10:01 | |
| *** jcoufal has joined #openstack-infra | 10:05 | |
| *** heyongli has quit IRC | 10:11 | |
| *** heyongli has joined #openstack-infra | 10:12 | |
| *** jaosorior has joined #openstack-infra | 10:18 | |
| *** heyongli has quit IRC | 10:22 | |
| *** heyongli has joined #openstack-infra | 10:22 | |
| *** zoli is now known as zoli|lunch | 10:29 | |
| *** pcaruana has quit IRC | 10:31 | |
| *** heyongli has quit IRC | 10:32 | |
| *** heyongli has joined #openstack-infra | 10:32 | |
| tojuvone | Hi, I just start working with openstack/fenix after creating it. To proceed with reviews I think I should be added to fenix-core. Can somebody help on this? | 10:33 |
| *** namnh has quit IRC | 10:40 | |
| *** e0ne has joined #openstack-infra | 10:41 | |
| *** heyongli has quit IRC | 10:42 | |
| *** heyongli has joined #openstack-infra | 10:42 | |
| *** shardy has quit IRC | 10:44 | |
| witek | frickler: thanks | 10:47 |
| *** alexchadin has quit IRC | 10:49 | |
| *** heyongli has quit IRC | 10:52 | |
| *** heyongli has joined #openstack-infra | 10:53 | |
| *** heyongli has quit IRC | 11:03 | |
| *** heyongli has joined #openstack-infra | 11:03 | |
| *** alexchadin has joined #openstack-infra | 11:10 | |
| *** heyongli has quit IRC | 11:13 | |
| *** heyongli has joined #openstack-infra | 11:13 | |
| *** tpsilva has joined #openstack-infra | 11:14 | |
| frickler | tojuvone: done | 11:15 |
| *** jpena is now known as jpena|lunch | 11:15 | |
| *** heyongli has quit IRC | 11:23 | |
| *** heyongli has joined #openstack-infra | 11:24 | |
| *** udesale has quit IRC | 11:24 | |
| openstackgerrit | Paul Belanger proposed openstack-infra/zuul master: Add min_avail_hdd governor for zuul-executor https://review.openstack.org/577948 | 11:26 |
| *** zoli|lunch is now known as zoli | 11:26 | |
| *** zoli is now known as zoli|wfh | 11:26 | |
| *** zoli|wfh is now known as zoli | 11:26 | |
| *** boden has joined #openstack-infra | 11:29 | |
| *** e0ne has quit IRC | 11:29 | |
| *** dpawlik has quit IRC | 11:33 | |
| *** heyongli has quit IRC | 11:33 | |
| *** heyongli has joined #openstack-infra | 11:34 | |
| *** ldnunes has joined #openstack-infra | 11:35 | |
| *** dpawlik has joined #openstack-infra | 11:36 | |
| *** amoralej is now known as amoralej|lunch | 11:37 | |
| tojuvone | frickler, thanks | 11:38 |
| *** xinliang_ has quit IRC | 11:38 | |
| *** e0ne has joined #openstack-infra | 11:42 | |
| *** heyongli has quit IRC | 11:44 | |
| *** heyongli has joined #openstack-infra | 11:44 | |
| openstackgerrit | Merged openstack-infra/zuul-jobs master: Use debian-security mirror https://review.openstack.org/578259 | 11:47 |
| *** Shrews has quit IRC | 11:48 | |
| *** Shrews has joined #openstack-infra | 11:50 | |
| *** heyongli has quit IRC | 11:54 | |
| *** heyongli has joined #openstack-infra | 11:54 | |
| *** alexchadin has quit IRC | 12:02 | |
| openstackgerrit | Paul Belanger proposed openstack-infra/zuul master: Add zuul.child_jobs in ansible inventory file https://review.openstack.org/578181 | 12:04 |
| openstackgerrit | Paul Belanger proposed openstack-infra/zuul master: WIP: Support skip_child_jobs via zuul_return https://review.openstack.org/578230 | 12:04 |
| *** heyongli has quit IRC | 12:04 | |
| *** heyongli has joined #openstack-infra | 12:04 | |
| *** dtantsur is now known as dtantsur|brb | 12:06 | |
| *** jpena|lunch is now known as jpena | 12:08 | |
| *** ttx has quit IRC | 12:09 | |
| *** jcoufal has quit IRC | 12:11 | |
| *** rh-jelabarre has joined #openstack-infra | 12:11 | |
| *** alexchadin has joined #openstack-infra | 12:14 | |
| *** heyongli has quit IRC | 12:14 | |
| *** heyongli has joined #openstack-infra | 12:15 | |
| *** shardy has joined #openstack-infra | 12:17 | |
| *** ispp has quit IRC | 12:20 | |
| *** ttx has joined #openstack-infra | 12:22 | |
| *** heyongli has quit IRC | 12:25 | |
| *** heyongli has joined #openstack-infra | 12:25 | |
| *** ykarel has quit IRC | 12:25 | |
| *** rlandy has joined #openstack-infra | 12:29 | |
| *** trown|outtypewww is now known as trown | 12:30 | |
| *** kgiusti has joined #openstack-infra | 12:34 | |
| *** zul has joined #openstack-infra | 12:35 | |
| *** heyongli has quit IRC | 12:35 | |
| *** heyongli has joined #openstack-infra | 12:35 | |
| *** gongysh has quit IRC | 12:37 | |
| *** ykarel has joined #openstack-infra | 12:39 | |
| *** mriedem has joined #openstack-infra | 12:39 | |
| *** ykarel has quit IRC | 12:39 | |
| *** ykarel has joined #openstack-infra | 12:40 | |
| *** auristor has quit IRC | 12:40 | |
| *** auristor has joined #openstack-infra | 12:42 | |
| *** zoli is now known as zoli|afk | 12:42 | |
| *** zul has quit IRC | 12:44 | |
| *** heyongli has quit IRC | 12:45 | |
| *** heyongli has joined #openstack-infra | 12:45 | |
| *** auristor has quit IRC | 12:46 | |
| *** myoung|off is now known as myoung | 12:47 | |
| *** emerson has quit IRC | 12:47 | |
| *** auristor has joined #openstack-infra | 12:47 | |
| *** jcoufal has joined #openstack-infra | 12:50 | |
| *** dhajare has quit IRC | 12:54 | |
| *** heyongli has quit IRC | 12:55 | |
| *** heyongli has joined #openstack-infra | 12:56 | |
| *** edmondsw has joined #openstack-infra | 12:57 | |
| *** edmondsw has quit IRC | 12:57 | |
| *** edmondsw has joined #openstack-infra | 12:57 | |
| *** hashar is now known as idoine | 12:59 | |
| *** panda|off is now known as panda | 13:00 | |
| *** zul has joined #openstack-infra | 13:00 | |
| *** Shrews has quit IRC | 13:00 | |
| *** quiquell|rover is now known as quique|rover|lch | 13:01 | |
| *** dpawlik has quit IRC | 13:03 | |
| *** heyongli has quit IRC | 13:06 | |
| *** heyongli has joined #openstack-infra | 13:06 | |
| *** dpawlik has joined #openstack-infra | 13:06 | |
| *** Shrews has joined #openstack-infra | 13:07 | |
| *** agopi has joined #openstack-infra | 13:07 | |
| tosky | if a job is defined for a certain repository in-tree, and I want to move its definition to project-config (in order to use a global template), should I remove the job first from the repository, or will it be merged and I can remove it later? | 13:09 |
| *** dhajare has joined #openstack-infra | 13:10 | |
| *** shardy has quit IRC | 13:14 | |
| *** heyongli has quit IRC | 13:16 | |
| *** heyongli has joined #openstack-infra | 13:16 | |
| *** ykarel is now known as ykarel|afk | 13:17 | |
| *** amoralej|lunch is now known as amoralej | 13:18 | |
| *** ykarel|afk has quit IRC | 13:23 | |
| *** e0ne has quit IRC | 13:26 | |
| fungi | yamamoto: i believe build-openstack-sphinx-docs doesn't use tox at all (and follows the new pti guidelines instead), so tox-siblings wouldn't do much good | 13:26 |
| *** heyongli has quit IRC | 13:26 | |
| *** heyongli has joined #openstack-infra | 13:26 | |
| fungi | quique|rover|lch: master branches of all non-retired repositories should be included in codesearch, so if it's not there i'd consider that a bug. i can double-check in a bit | 13:27 |
| fungi | frickler: zuul switched to relying on gerrit's relatively new ability to determine mergeability of a change, but gerrit uses very subtle clues in its ui for that unfortunately (compared to the heavy-handed commenting zuul used to do, since that was basically the only signalling option available to it) | 13:29 |
| *** alexchadin has quit IRC | 13:29 | |
| fungi | tosky: job names are globally-unique and zuul will enforce that, so if you try to add a new job to project-config with the same name as a job in your repo, it will reject that change | 13:30 |
| fungi | thus, needs to either be removed from your repo first or renamed somewhere | 13:30 |
| tosky | fungi: that means that I should define my local version of the job with a different name, for example extending the original job? | 13:30 |
| tosky | so that I can test while waiting for the project-config change to be merged | 13:30 |
| openstackgerrit | Thierry Carrez proposed openstack-infra/project-config master: Remove direct tagging from manila repositories https://review.openstack.org/578381 | 13:31 |
| yamamoto | fungi: so, if something is broken due to a problem in siblings, i need to wait for it to be released on pypi with a fix? | 13:32 |
| *** zoli|afk is now known as zoli|wfh | 13:32 | |
| fungi | tosky: yes, that should work | 13:32 |
| tosky | fungi: thanks | 13:32 |
| *** zoli|wfh is now known as zoli | 13:32 | |
| fungi | yamamoto: it sounds like an unusual scenario to me... your documentation builds need to rely on unreleased source code from another project? | 13:33 |
| yamamoto | fungi: sort of. for autodoc. | 13:34 |
| fungi | yamamoto: i believe the docs builds will still use global upper-constraints.txt so if a recent release of some dependency is broken you may want to talk to the requirements team about excluding that newer version | 13:34 |
| *** alexchadin has joined #openstack-infra | 13:34 | |
| *** dtantsur|brb is now known as dtantsur | 13:35 | |
| *** quique|rover|lch is now known as quiquell|rover | 13:35 | |
| *** ihar has joined #openstack-infra | 13:35 | |
| yamamoto | fungi: i guess excluding a version of neutron just for sphinx warnings is not acceptable | 13:36 |
| *** heyongli has quit IRC | 13:36 | |
| *** heyongli has joined #openstack-infra | 13:37 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul master: Use window.location.href for the href not pathname https://review.openstack.org/578243 | 13:37 |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul master: web: fix status page flickering https://review.openstack.org/578226 | 13:37 |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul master: Revert "Revert "Upgrade from angularjs (v1) to angular (v6)"" https://review.openstack.org/578384 | 13:37 |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul master: Consolidate navigation functions https://review.openstack.org/578385 | 13:37 |
| *** e0ne has joined #openstack-infra | 13:38 | |
| fungi | yamamoto: oh, so your project is importing neutron during the sphinx build? | 13:39 |
| yamamoto | fungi: yes | 13:39 |
| fungi | and it's installing neutron from pypi? | 13:40 |
| yamamoto | fungi: i think so, yes | 13:41 |
| *** priteau has joined #openstack-infra | 13:41 | |
| yamamoto | fungi: it used to use tox_install but these days it's using pypi i think | 13:41 |
| fungi | this might be something that needs to be worked around specially. i have a feeling the pti guidelines for documentation builds didn't take into account that you might want to import unreleased source code from one project while building documentation for another project | 13:42 |
| fungi | mordred: dhellmann: ^ an interesting case study | 13:43 |
| fungi | it's probably going to need its own separate job | 13:43 |
| mordred | fungi: I don't think it's just PTI guidelines | 13:43 |
| fungi | well, the typical build-sphinx workflow | 13:44 |
| dhellmann | ugh, I wish we could just release neutron to avoid all of this special handling | 13:44 |
| fungi | i mean, getting rtd to do something like that would be similarly challenging | 13:44 |
| mordred | conceptually I'd argue that a repo's documentatoin not being able to be built without unreleased source code from another repo is a bug that should be fixed - not one that should be accomodated - and the fact that tox_install scripts were subverting our requirements process before is not a reason for us to continue supporting the broken thing | 13:45 |
| fungi | yeah, i'm curious to know _why_ whatever project this is needs neutron installed just do do a docs build | 13:46 |
| mordred | if a project depends on another project, that dependency should be expressed in a requirements file somewhere. if that dependency is not expressed or is not expressable, that is a fundamental problem | 13:46 |
| mordred | and I agree with dhellmann - I really wish we could just release neutron and stop this special handling | 13:46 |
| dhellmann | mordred : you're arguing my position w.r.t. unit tests, too. I thought you used to "like" tox-siblings, etc. | 13:46 |
| fungi | i think the bigger challenge is that plugins/drivers are tightly coupled to the development cycle of what they're plugging into right now | 13:47 |
| *** heyongli has quit IRC | 13:47 | |
| mordred | I *do* like tox-siblings - but tox-siblings is intended to provide a mechanism for additional vetting of pre-released content | 13:47 |
| mordred | it should not be a primary method of expressing dependencies | 13:47 |
| dhellmann | yamamoto : perhaps you can start a mailing list thread about this so we can ask the neutron team to consider changing their release model? | 13:47 |
| *** heyongli has joined #openstack-infra | 13:47 | |
| fungi | so to say don't consume unreleased neutron basically means plugins and drivers get to start development for neutron rocky support when neutron starts working on stein | 13:47 |
| mordred | all of the things *should* work without tox-siblings being involved | 13:47 |
| yamamoto | it's not only about neutron. we have the same issue with os-vif. | 13:47 |
| yamamoto | (we have a plugin for os-vif) | 13:48 |
| mordred | I think those things needto consider a cycle-with-intermediary release model | 13:48 |
| * fungi was using neutron as an example in this case | 13:48 | |
| fungi | so basically neutron, horizon, et cetera should be released more like libraries because they're being treated like libraries by their plugins/drivers? | 13:49 |
| mordred | yes | 13:49 |
| fungi | makes sense | 13:49 |
| mordred | dhellmann: I feel like I still need to do a better job of explaining what tox-siblings is there to help with and what it's not intended to do | 13:50 |
| yamamoto | in case of neutron i guess neutron-lib is supposed to be an answer but an intermediate solution might be useful | 13:51 |
| mordred | yamamoto: yah. if neutron-lib was sufficient for the plugin cases that would be wonderful - but it seems we're not there yet, yes? | 13:51 |
| dhellmann | yamamoto : os-vif does regular releases, so if you're relying on unreleased features you should probably not be doing that and should be asking the os-vif owners for even more frequent releases | 13:53 |
| *** dpawlik has quit IRC | 13:53 | |
| *** john_studarus has quit IRC | 13:53 | |
| fungi | i wonder if tagging prereleases and publishing those to pypi would be an acceptable middle-ground | 13:53 |
| *** dpawlik has joined #openstack-infra | 13:54 | |
| mordred | why not just actual intermediary releases though? we already have service projects that do that - and we've got version numbering that should allow it to not be too crazy, yeah? or am I missing an issue (likely) | 13:55 |
| *** bobh has joined #openstack-infra | 13:55 | |
| fungi | that also seems fine if they're on board | 13:56 |
| *** heyongli has quit IRC | 13:57 | |
| *** heyongli has joined #openstack-infra | 13:57 | |
| *** shardy has joined #openstack-infra | 13:59 | |
| *** eharney has joined #openstack-infra | 14:00 | |
| openstackgerrit | Petr Kovar proposed openstack-infra/irc-meetings master: Turn doc team meeting into office hours https://review.openstack.org/578398 | 14:03 |
| *** udesale has joined #openstack-infra | 14:04 | |
| *** atarakt_ has joined #openstack-infra | 14:05 | |
| *** Shrews has quit IRC | 14:06 | |
| *** heyongli has quit IRC | 14:07 | |
| *** heyongli has joined #openstack-infra | 14:08 | |
| *** Shrews has joined #openstack-infra | 14:08 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul master: Consolidate navigation functions https://review.openstack.org/578385 | 14:10 |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul master: web: fix status page flickering https://review.openstack.org/578226 | 14:10 |
| fungi | quiquell|rover: i see tripleo-upgrade (singular, not upgrades plural) in the project selection list for the advanced search on codesearch.openstack.org | 14:12 |
| fungi | what were you trying which wasn't working? | 14:12 |
| fungi | a working example search: http://codesearch.openstack.org/?q=pbr&i=nope&files=setup.py&repos=tripleo-upgrade | 14:13 |
| *** Shrews has quit IRC | 14:17 | |
| *** heyongli has quit IRC | 14:17 | |
| AJaeger | yamamoto: build-openstack-sphinx-docs should build without siblings installed. Create doc/requirememts.txt for it | 14:18 |
| *** heyongli has joined #openstack-infra | 14:18 | |
| fungi | it sounds like the project in question needs to have autodoc do some discovery of neutron functions referenced from that project's source code | 14:19 |
| fungi | and so wants neutron installed | 14:19 |
| AJaeger | yamamoto: there're also autodoc tools that work without the imports AFAIK - mordred, stephenfin might be able to help | 14:20 |
| AJaeger | fungi: now read full backscroll ;/ | 14:20 |
| *** lpetrut has quit IRC | 14:22 | |
| *** shardy has quit IRC | 14:22 | |
| *** jamesmcarthur has joined #openstack-infra | 14:24 | |
| boden | hi… is it valid to use multiple “Depends-On” in a patch to pull in patches from mulitple repos during zuul jobs? | 14:24 |
| clarkb | boden: yes | 14:24 |
| fungi | boden: yes, completely fine | 14:24 |
| fungi | lots of changes do that | 14:24 |
| boden | clarkb fungi cool thx | 14:25 |
| *** dhajare has quit IRC | 14:25 | |
| fungi | boden: just beware of accidentally adding dependency loops. the more complex your change dependency tree grows, the easier it is to lose track and accidentally create a cyclic dependency. if zuul encounters a cycle anywhere in a dependent change set, it will ignore all the changes in that set until the loop is removed | 14:26 |
| boden | fungi ack.. I will try to keep an eye out for that | 14:26 |
| *** jamesmcarthur has quit IRC | 14:27 | |
| *** jamesmcarthur has joined #openstack-infra | 14:27 | |
| *** heyongli has quit IRC | 14:28 | |
| *** heyongli has joined #openstack-infra | 14:28 | |
| *** threestrands has quit IRC | 14:30 | |
| *** Shrews has joined #openstack-infra | 14:32 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul master: Consolidate navigation functions https://review.openstack.org/578385 | 14:32 |
| *** felipemonteiro has joined #openstack-infra | 14:33 | |
| *** felipemonteiro_ has joined #openstack-infra | 14:34 | |
| *** janki has quit IRC | 14:35 | |
| dhellmann | fungi : yeah, given pip's special needs when it comes to pre-releases I would prefer to just use regular releases. otherwise we're trading one special case for another. | 14:37 |
| *** felipemonteiro_ has quit IRC | 14:37 | |
| *** rh-jelabarre has quit IRC | 14:37 | |
| *** felipemonteiro_ has joined #openstack-infra | 14:37 | |
| *** heyongli has quit IRC | 14:38 | |
| *** heyongli has joined #openstack-infra | 14:38 | |
| *** felipemonteiro has quit IRC | 14:38 | |
| mordred | ++ | 14:39 |
| *** alexchadin has quit IRC | 14:39 | |
| fungi | sure, we'd basically have to put the prerelease versions in upper-constraints.txt, which should work but what would get complicated is generating that constraints list to make sure that happens | 14:40 |
| flaper87 | hey folks, I've been seeing some failures in a job for this patch that I'm not sure whether they are our fault or whether there's something happening with the CI system. https://review.openstack.org/#/c/574233/ | 14:41 |
| flaper87 | specifically we're seeing: tripleo-ci-centos-7-scenario009-multinode-oooq finger://ze04.openstack.org/d816f2f1cc62476b89b98fbbe1825bfa : POST_FAILURE in 2h 14m 42s (non-voting) | 14:41 |
| *** Bhujay has quit IRC | 14:41 | |
| flaper87 | any idea how I can debug this? Is this our fault? have I missed some important email coming from you? | 14:41 |
| *** alexchadin has joined #openstack-infra | 14:41 | |
| * flaper87 checks os-dev again | 14:41 | |
| fungi | a lot of the time POST_FAILURE results are actually a secondary failure, and the job itself failed but the way it failed also caused something like log collection to fail too (perhaps because some expected logs weren't created) | 14:43 |
| openstackgerrit | Merged openstack-infra/irc-meetings master: Turn doc team meeting into office hours https://review.openstack.org/578398 | 14:43 |
| flaper87 | fungi: mmh, in that case, can we set a hold on these nodes again? Sorry for asking for this so much lately :/ | 14:44 |
| flaper87 | fungi: 574233 tripleo-ci-centos-7-scenario009-multinode-oooq | 14:44 |
| clarkb | flaper87: fungi usually I like to read the logs and identify any obvious time losses | 14:44 |
| clarkb | as a first step. Fairly often it is something obvious like download massive image or database migrations take 15 minutes now for some reason | 14:45 |
| *** evrardjp has quit IRC | 14:45 | |
| *** evrardjp has joined #openstack-infra | 14:45 | |
| flaper87 | These are the logs of the last run: https://paste.ee/p/qbcqv | 14:46 |
| flaper87 | this is the log* :P | 14:46 |
| *** sthussey has joined #openstack-infra | 14:46 | |
| corvus | flaper87: here are the logs: http://logs.openstack.org/33/574233/22/experimental/tripleo-ci-centos-7-scenario009-multinode-oooq/d816f2f/ | 14:46 |
| flaper87 | corvus: oh, nice! | 14:47 |
| flaper87 | thanks | 14:47 |
| corvus | looks like the console log is missing (as expected since that's the last thing that gets uploaded), but you have that in the pastebin | 14:47 |
| flaper87 | corvus: did you have to do something to collect them or was it as easy as building the URL yourself? | 14:47 |
| flaper87 | corvus: https://paste.ee/p/qbcqv this should be the console log | 14:48 |
| corvus | flaper87: found the url. started at: logs.openstack.org/33/574233/ | 14:48 |
| flaper87 | corvus: thanks! | 14:48 |
| flaper87 | I'll do that first if this happens again | 14:48 |
| *** heyongli has quit IRC | 14:48 | |
| *** heyongli has joined #openstack-infra | 14:49 | |
| fungi | yeah, looks like ara says the run phase did indeed fail | 14:49 |
| fungi | http://logs.openstack.org/33/574233/22/experimental/tripleo-ci-centos-7-scenario009-multinode-oooq/d816f2f/ara-report/ | 14:49 |
| *** hamzy has quit IRC | 14:49 | |
| corvus | but most of the post playbook ran, so it's very likely that just the console log upload failed for some reason | 14:49 |
| fungi | toci_quickstart.sh seems to have terminated early according to that | 14:50 |
| smcginnis | I'm looking at un-retiring the openstack/operations-guide. Its core group had been openstack-docs-core, but now we want core rights to be for a subset of the newly proposed operations sig. | 14:50 |
| smcginnis | How does one go about requesting a new gerrit group? | 14:50 |
| mriedem | gerrit dead? | 14:50 |
| fungi | smcginnis: the group will still be there | 14:50 |
| corvus | flaper87: does this happen a lot for this job? | 14:50 |
| smcginnis | mriedem: Works for me. | 14:50 |
| fungi | smcginnis: oh, changing the group... just mention the new group name in the acl and it will get created | 14:51 |
| mriedem | i'm getting: "The page you requested was not found, or you do not have permission to view this page." | 14:51 |
| mriedem | while saving a comment | 14:51 |
| smcginnis | fungi: OK, thanks. Then request to have initial members added to it? | 14:51 |
| flaper87 | corvus: no, it just started happening on Thur / Fri last week but, tbh, I didn't have much of a clue on how to get the logs and we tried some things before reaching out | 14:51 |
| mriedem | nvm, refreshed | 14:51 |
| clarkb | install the undercloud took almost half an hour, prepare for containerized deployment another 5 minutes. Deploy the overcloud another half hour, this actually failed. Tripleo collect logs then fails after 15 minutes and has a broken callback plugin. I don't think that timed out it was just slow and broken in a couple places and failed | 14:51 |
| smcginnis | mriedem: You didn't get the memo? | 14:51 |
| clarkb | flaper87: corvus ^ | 14:51 |
| mriedem | smcginnis: that i'm out? | 14:51 |
| mriedem | :( | 14:51 |
| smcginnis | mriedem: Accounting glitch has been fixed. | 14:52 |
| mriedem | alright, it's been fun folks | 14:52 |
| * mriedem leaves | 14:52 | |
| flaper87 | clarkb: corvus thanks a bunch, now we can debug this better | 14:52 |
| *** caphrim007_ has quit IRC | 14:52 | |
| smcginnis | mriedem: You don't need to leave, just move your stuff to the basement. | 14:52 |
| *** hongbin has joined #openstack-infra | 14:52 | |
| *** quiquell|rover is now known as quiquell|off | 14:52 | |
| *** hamzy has joined #openstack-infra | 14:52 | |
| clarkb | flaper87: the actual failure is the deploy overcloud failure. The collect logs being broken is why we don't have complete logs I think (though job-output.txt shouldn't be affected by that) | 14:52 |
| fungi | smcginnis: correct | 14:53 |
| *** caphrim007_ has joined #openstack-infra | 14:53 | |
| *** hamzy has quit IRC | 14:53 | |
| smcginnis | fungi: Perfect, thanks. | 14:53 |
| *** hemna_ has joined #openstack-infra | 14:54 | |
| *** hamzy has joined #openstack-infra | 14:54 | |
| smcginnis | fungi: There's an existing osops-core that I don't think is used anymore, but would require redoing most of its membership. | 14:57 |
| *** mriedem is now known as mriedem_afk | 14:57 | |
| smcginnis | fungi: It does have rights on a different set of repos, but I think they are dormant. | 14:57 |
| smcginnis | fungi: Would it be better to create a new group, or take over this existing one? | 14:58 |
| *** heyongli has quit IRC | 14:58 | |
| fungi | probably reuse that one | 14:59 |
| *** heyongli has joined #openstack-infra | 14:59 | |
| smcginnis | fungi: Would you be able to remove all but Melvin and add me? https://review.openstack.org/#/admin/groups/1076,members | 14:59 |
| clarkb | smcginnis: typically we try to have the groups self manage first | 15:00 |
| *** camunoz has joined #openstack-infra | 15:00 | |
| smcginnis | clarkb: Makes sense. So I should be able to speak to Melvin to have him modify it. | 15:00 |
| *** tonytan4ever has joined #openstack-infra | 15:00 | |
| clarkb | Yup | 15:00 |
| *** egonzalez has quit IRC | 15:02 | |
| *** evrardjp has quit IRC | 15:02 | |
| *** evrardjp has joined #openstack-infra | 15:03 | |
| *** zul has quit IRC | 15:05 | |
| *** Parvej has joined #openstack-infra | 15:06 | |
| *** janki has joined #openstack-infra | 15:08 | |
| *** heyongli has quit IRC | 15:09 | |
| *** rpioso|afk is now known as rpioso | 15:09 | |
| *** heyongli has joined #openstack-infra | 15:09 | |
| openstackgerrit | Sean McGinnis proposed openstack-infra/project-config master: Readd openstack/operations-guide repo https://review.openstack.org/578441 | 15:15 |
| *** lpetrut has joined #openstack-infra | 15:15 | |
| *** kien_ has joined #openstack-infra | 15:17 | |
| *** heyongli has quit IRC | 15:19 | |
| *** heyongli has joined #openstack-infra | 15:19 | |
| smcginnis | Apparently that repo needs an additional manual step? ^ | 15:22 |
| smcginnis | "All projects | 15:22 |
| smcginnis | referenced within a Zuul configuration must first be added to the main | 15:22 |
| smcginnis | configuration file by the Zuul administrator." | 15:22 |
| corvus | smcginnis: luckily, you're the zuul administrator :) | 15:23 |
| smcginnis | corvus: Oh, nifty! :) | 15:23 |
| corvus | smcginnis: https://docs.openstack.org/infra/manual/creators.html#add-project-to-zuul | 15:23 |
| smcginnis | Where does that need to be added? | 15:23 |
| smcginnis | Sorry, I should have read through there. | 15:24 |
| openstackgerrit | Sean McGinnis proposed openstack-infra/project-config master: Readd openstack/operations-guide repo https://review.openstack.org/578441 | 15:25 |
| *** jcoufal has quit IRC | 15:25 | |
| corvus | smcginnis: that'll need to be its own change that lands first before the rest | 15:27 |
| *** alexchadin has quit IRC | 15:27 | |
| smcginnis | Oh, OK. | 15:27 |
| *** heyongli has quit IRC | 15:29 | |
| *** heyongli has joined #openstack-infra | 15:30 | |
| openstackgerrit | Sean McGinnis proposed openstack-infra/project-config master: Readd openstack/operations-guide repo https://review.openstack.org/578441 | 15:30 |
| openstackgerrit | Sean McGinnis proposed openstack-infra/project-config master: Add openstack/operations-guide to zuul https://review.openstack.org/578449 | 15:30 |
| *** edmondsw has quit IRC | 15:33 | |
| *** peereb has quit IRC | 15:35 | |
| *** lpetrut has quit IRC | 15:38 | |
| *** lpetrut has joined #openstack-infra | 15:39 | |
| *** heyongli has quit IRC | 15:39 | |
| *** heyongli has joined #openstack-infra | 15:40 | |
| *** gyee has joined #openstack-infra | 15:42 | |
| *** dpawlik has quit IRC | 15:44 | |
| *** cicisong has joined #openstack-infra | 15:48 | |
| *** pbourke has quit IRC | 15:48 | |
| *** heyongli has quit IRC | 15:50 | |
| *** heyongli has joined #openstack-infra | 15:50 | |
| *** mriedem_afk is now known as mriedem | 15:50 | |
| *** pbourke has joined #openstack-infra | 15:50 | |
| *** myoung is now known as myoung|biaf | 15:59 | |
| openstackgerrit | Artem Goncharov proposed openstack-infra/nodepool master: retire shade in favor of openstacksdk https://review.openstack.org/572829 | 16:00 |
| *** heyongli has quit IRC | 16:00 | |
| *** heyongli has joined #openstack-infra | 16:00 | |
| *** lpetrut has quit IRC | 16:02 | |
| *** andreas_s has quit IRC | 16:05 | |
| *** jesslampe has joined #openstack-infra | 16:06 | |
| *** myoung|biaf is now known as myoung | 16:06 | |
| *** idoine is now known as hashar | 16:07 | |
| *** ramishra has quit IRC | 16:08 | |
| *** jamesmcarthur has quit IRC | 16:08 | |
| *** edmondsw has joined #openstack-infra | 16:09 | |
| *** heyongli has quit IRC | 16:10 | |
| *** heyongli has joined #openstack-infra | 16:11 | |
| *** tobasco is now known as tobasco_afk | 16:12 | |
| *** e0ne has quit IRC | 16:13 | |
| *** atarakt_ has quit IRC | 16:14 | |
| *** efried is now known as efried_rollin | 16:17 | |
| *** gfidente has quit IRC | 16:20 | |
| *** lpetrut has joined #openstack-infra | 16:20 | |
| *** heyongli has quit IRC | 16:20 | |
| *** heyongli has joined #openstack-infra | 16:21 | |
| *** apetrich has quit IRC | 16:21 | |
| *** apetrich has joined #openstack-infra | 16:21 | |
| *** hashar is now known as hasharAway | 16:28 | |
| *** tosky has quit IRC | 16:29 | |
| *** tosky has joined #openstack-infra | 16:29 | |
| *** jpich has quit IRC | 16:29 | |
| *** florianf is now known as florianf|afk | 16:30 | |
| *** heyongli has quit IRC | 16:31 | |
| *** heyongli has joined #openstack-infra | 16:31 | |
| *** panda is now known as panda|bike | 16:33 | |
| *** udesale has quit IRC | 16:33 | |
| *** tesseract has quit IRC | 16:34 | |
| *** felipemonteiro__ has joined #openstack-infra | 16:36 | |
| *** dpawlik has joined #openstack-infra | 16:38 | |
| *** dhajare has joined #openstack-infra | 16:40 | |
| *** felipemonteiro_ has quit IRC | 16:40 | |
| *** heyongli has quit IRC | 16:41 | |
| *** heyongli has joined #openstack-infra | 16:41 | |
| openstackgerrit | Doug Hellmann proposed openstack-infra/project-config master: remove oslo.config job settings https://review.openstack.org/578478 | 16:42 |
| *** dpawlik has quit IRC | 16:43 | |
| *** edmondsw has quit IRC | 16:43 | |
| *** dpawlik has joined #openstack-infra | 16:44 | |
| *** edmondsw has joined #openstack-infra | 16:44 | |
| *** edmondsw has quit IRC | 16:44 | |
| *** rlandy is now known as rlandy|brb | 16:45 | |
| boden | hi. I’m trying to get the networking-midonet zuul v3 jobs working in https://review.openstack.org/#/c/578212/ and debugging the tempest failures seen therein.. what I see is that the neuron-dynamic-routing dependency is installed ver 12.0.1 (http://logs.openstack.org/12/578212/2/check/networking-midonet-tempest-aio-ml2/3b38f51/job-output.txt.gz#_2018-06-27_13_35_16_104997), but its supposed to be installed from master as per | 16:47 |
| boden | required-projects… feel like I’m missing something here.. Any ideas? | 16:47 |
| clarkb | boden: tempest runs its own installation in a virtualenv independent of the global install | 16:49 |
| boden | clarkb hmm. I wasn’t aware of that.. any ideas how I would address this issue? I’m a bit confused | 16:49 |
| fungi | yeah, devstack should install neuron-dynamic-routing from source, presumably, but tempest may not install it from source in its own tox env? | 16:49 |
| clarkb | fungi: exactly | 16:50 |
| clarkb | devstack installs a cloud using the specified sources. Then it installs tempest by running tox | 16:50 |
| fungi | why is tempest installing neuron-dynamic-routing? is there a tempest plugin in there which hasn't been extracted yet? | 16:50 |
| clarkb | I don't know if tempest has a mechanism for installing its deps from source based on job config, probably a question for the qa team | 16:50 |
| clarkb | fungi: that is my guess | 16:50 |
| fungi | the qa team will probably say "put your tempest plugin in its own package" ;) | 16:51 |
| *** derekh has quit IRC | 16:51 | |
| *** heyongli has quit IRC | 16:51 | |
| *** heyongli has joined #openstack-infra | 16:52 | |
| boden | fungi so this could be a direct result of midonet not having pulled their tempest “stuff” into its own repo like many others did? | 16:53 |
| fungi | boden: if that's the reason neuron-dynamic-routing needs to be installed in the tempest env, then i'd say yes | 16:54 |
| fungi | you're basically asking to install that one package multiple times in different contexts to use different parts of it. therein lies madness | 16:55 |
| fungi | boden: but as clarkb pointed out, this is better hashed out in #openstack-qa | 16:56 |
| dhellmann | I'm starting to move the zuul job settings for oslo.config from project-config to the file in the oslo.config repo. When I submit one of those patches, how do I know zuul is using the new config to test it? https://review.openstack.org/578481 | 16:58 |
| clarkb | dhellmann: in the logs dir zuul-infra/iventory file it will show you the job inheritance config path | 16:58 |
| dhellmann | excellent, thanks | 16:58 |
| clarkb | dhellmann: that path should have your oslo.config in it | 16:58 |
| fungi | yeal, tells you the repository, branch and filename i believe | 16:59 |
| *** ssbarnea[m] has joined #openstack-infra | 17:00 | |
| *** ssbarnea has left #openstack-infra | 17:00 | |
| *** kien_ has quit IRC | 17:00 | |
| clarkb | hrm weechat just disabled smartfilters on this channel | 17:00 |
| clarkb | oh TIL about alt + - | 17:01 |
| *** heyongli has quit IRC | 17:01 | |
| *** zoli is now known as zoli|gone | 17:02 | |
| *** heyongli has joined #openstack-infra | 17:02 | |
| *** zoli|gone is now known as zoli | 17:02 | |
| *** myoung is now known as myoung|lunch | 17:02 | |
| *** dtantsur is now known as dtantsur|afk | 17:04 | |
| fungi | heh | 17:05 |
| fungi | you'd never used the toggle (at least intentionally)? | 17:06 |
| clarkb | nope just alt + = | 17:06 |
| *** rlandy|brb is now known as rlandy | 17:07 | |
| fungi | oh, huh yeah i normally do alt+= | 17:07 |
| fungi | what's the difference? | 17:07 |
| fungi | they seem to do the same thing to me | 17:07 |
| *** dhajare has quit IRC | 17:08 | |
| * fungi finds the default keybinding docs | 17:08 | |
| fungi | oh, = is global, - is per buffer | 17:09 |
| fungi | https://weechat.org/files/doc/devel/weechat_user.en.html#key_bindings_buffers_windows | 17:10 |
| fungi | the other day i accidentally toggled "mouse mode" on and could no longer use my pointer to highlight and copy for pasting into another window | 17:11 |
| fungi | that took a bit of reading about a feature i don't want | 17:11 |
| fungi | now it's thoroughly disabled via configuration | 17:11 |
| *** Guest44976 has quit IRC | 17:11 | |
| clarkb | fungi: I toggle that one on accidently about once every other month | 17:12 |
| clarkb | I've learned /mouse disable and then /save | 17:12 |
| *** heyongli has quit IRC | 17:12 | |
| clarkb | but don't know what I do to toggle it on | 17:12 |
| *** heyongli has joined #openstack-infra | 17:12 | |
| fungi | alt+m i think is how it happened | 17:12 |
| *** jpena is now known as jpena|off | 17:12 | |
| fungi | but if you don't want it, you can /set weechat.look.mouse off | 17:12 |
| openstackgerrit | Doug Hellmann proposed openstack-infra/project-config master: remove oslo.config job settings https://review.openstack.org/578478 | 17:13 |
| *** trown is now known as trown|lunch | 17:16 | |
| *** felipemonteiro__ has quit IRC | 17:16 | |
| *** felipemonteiro_ has joined #openstack-infra | 17:16 | |
| *** efried_rollin is now known as efried | 17:19 | |
| *** iranzo has quit IRC | 17:21 | |
| *** heyongli has quit IRC | 17:22 | |
| *** jklare has quit IRC | 17:22 | |
| *** heyongli has joined #openstack-infra | 17:22 | |
| *** calbers has quit IRC | 17:23 | |
| *** eharney has quit IRC | 17:23 | |
| *** jklare has joined #openstack-infra | 17:23 | |
| *** calbers has joined #openstack-infra | 17:24 | |
| *** eharney has joined #openstack-infra | 17:27 | |
| *** amoralej is now known as amoralej|off | 17:29 | |
| efried | Is this the right place to get help on sphinx builds via zuul? | 17:31 |
| fungi | efried: it can be... what's up? | 17:32 |
| *** heyongli has quit IRC | 17:32 | |
| *** heyongli has joined #openstack-infra | 17:33 | |
| *** zul has joined #openstack-infra | 17:33 | |
| efried | fungi: The cyborg team is getting interested in putting a seqdiag into one of their specs. But in your typical one-step-forward-two-steps-back fashion, we've found out that their specs aren't actually *building* anywhere. | 17:33 |
| efried | at least as far as I can see. | 17:33 |
| efried | I thought I could make it happen with this: https://review.openstack.org/#/c/578455/ | 17:33 |
| efried | ...which works for me locally - i.e. builds HTML files from the specs as one would expect | 17:34 |
| efried | ...but which doesn't seem to be having any effect in the zuul build: http://logs.openstack.org/55/578455/1/check/build-openstack-sphinx-docs/38f2b5d/html/specs | 17:34 |
| efried | I've sorta been assuming that the zuul doc build runs the `docs` tox job, but apparently that's not the case? | 17:35 |
| efried | (btw, another step backward is that this isn't using the new doc build process - that's another issue I'm sure we'll have to address soon) | 17:36 |
| fungi | build-openstack-sphinx-docs implements the docs pti described at https://governance.openstack.org/tc/reference/project-testing-interface.html#documentation | 17:36 |
| *** jamesmcarthur has joined #openstack-infra | 17:37 | |
| fungi | so installs doc/requirements.txt and bindep.txt contents and then runs `sphinx-build -b html doc/source doc/build` | 17:37 |
| fungi | it doesn't interact with tox.ini in any way | 17:37 |
| efried | I see. So in order to build specs, we would actually have to move the specs into a subdir of doc/source | 17:37 |
| efried | (right now they're in doc/specs) | 17:37 |
| fungi | and/or use a separate job | 17:37 |
| efried | What really ought to happen is for them to move their specs into cyborg-specs | 17:38 |
| efried | but apparently they're not gonna, for now. | 17:38 |
| efried | Okay, fungi, thanks for the help. I'll go off and do more things. | 17:38 |
| *** larsks has joined #openstack-infra | 17:40 | |
| larsks | Hey folks. Sometimes, when I submit a review with a closes-bug: tag, the review never gets associated with the corresponding bug in launchpad. What can cause that? And can I do something to make the review show up on the bug report (other than just leaving a comment)? | 17:41 |
| *** heyongli has quit IRC | 17:42 | |
| *** heyongli has joined #openstack-infra | 17:43 | |
| openstackgerrit | Andreas Jaeger proposed openstack-infra/infra-manual master: Remove system-required instructions for creators https://review.openstack.org/577310 | 17:43 |
| AJaeger | mordred, clarkb, fungi, what do you think of the above? ^ I tried wrapping my head around this for mordred ;) | 17:44 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul master: Don't add implied branch matchers to project-pipeline variants https://review.openstack.org/577881 | 17:44 |
| clarkb | larsks: if our LP user doesn't have bug admin (or whatever the lp term is) for that project it can't do things like close bugs irc | 17:44 |
| larsks | clarkb: but's not even adding the comment to the bug report. This is specifically for https://review.openstack.org/#/c/578498/, which fixes https://bugs.launchpad.net/kolla/+bug/1778963. | 17:45 |
| openstack | Launchpad bug 1778963 in kolla "The horizon container does not properly re-run maintenance tasks when configuration is modified" [Undecided,In progress] | 17:45 |
| *** auristor has quit IRC | 17:45 | |
| larsks | I've had this happen with a number of other reviews, too, for other projects. | 17:45 |
| fungi | larsks: also, if the commit message footer isn't present in the first patchset, it won't comment until the change merges | 17:45 |
| larsks | fungi: the original commit message included the closes-bug tag. | 17:46 |
| larsks | I'm just wondering if I'm doing something wrong to cause this... | 17:46 |
| fungi | larsks: is it consistent on that project? | 17:46 |
| openstackgerrit | Andreas Jaeger proposed openstack-infra/infra-manual master: Remove system-required instructions for creators https://review.openstack.org/577310 | 17:46 |
| *** electrofelix has quit IRC | 17:46 | |
| *** jamesmcarthur has quit IRC | 17:47 | |
| larsks | fungi: I can't say. I haven't submitted a bug against the kolla project recently. | 17:47 |
| fungi | larsks: what was the corresponding review for that bug? | 17:47 |
| fungi | i can take a look in logs | 17:47 |
| larsks | fungi: https://review.openstack.org/#/c/578498/ | 17:47 |
| fungi | another (less common) cause is if the user has multiple active gerrit accounts then the update_bug hook script can't reverse-resolve their openid to a particular lp account and breaks since it's unable to reassign the bugtask | 17:48 |
| fungi | and yet another common cause is sometimes the lp api just chokes and doesn't do what it's asked | 17:49 |
| openstackgerrit | Andreas Jaeger proposed openstack-infra/project-config master: Use wildcard to apply system-required template https://review.openstack.org/577307 | 17:49 |
| openstackgerrit | Andreas Jaeger proposed openstack-infra/project-config master: Use wildcard to apply system-required template https://review.openstack.org/577307 | 17:50 |
| larsks | fungi: e.g., the same thing happened on https://bugs.launchpad.net/tripleo/+bug/1777023 | 17:50 |
| openstack | Launchpad bug 1777023 in tripleo "neutron::services::lbaas runs too late, causing deploy to fail" [Medium,In progress] - Assigned to Lars Kellogg-Stedman (larsks) | 17:50 |
| clarkb | larsks: [2018-06-27 17:39:34,169] [HookQueue-1] INFO com.googlesource.gerrit.plugins.hooks.HookTask : hook[patchset-created] output: Regular users can assign and unassign only themselves and their teams. Only project owners, bug supervisors, drivers and release managers can assign others. | 17:51 |
| fungi | larsks: in the case of kolla, the bug supervisor is the kolla-drivers lp team which does not include our account | 17:51 |
| clarkb | I'm not completely sure that is your change as we don't seem to log the change or bug number but the timestamp matches up | 17:51 |
| fungi | so our gerrit is unable to reassign bugs for kolla | 17:51 |
| clarkb | and that causes a traceback which crashes the update | 17:51 |
| larsks | I see. Is that something that one would typically report to the upstream project as something that should be fixed? | 17:52 |
| AJaeger | config-core, I updated mordred's changes to use wildcards - could you review https://review.openstack.org/577307 and https://review.openstack.org/577310 together, please? | 17:52 |
| clarkb | larsks: yes | 17:52 |
| clarkb | larsks: I believe our projects creators doc in the infra manual should cover what is required | 17:52 |
| larsks | Awesome, I'll take a look and poke people. Thanks! | 17:52 |
| *** heyongli has quit IRC | 17:53 | |
| *** heyongli has joined #openstack-infra | 17:53 | |
| fungi | basically one of the administrators in https://launchpad.net/~kolla-drivers/+members needs to add the https://launchpad.net/~hudson-openstack user to that team | 17:53 |
| clarkb | AJaeger: firefox is attempting to render the diff now :0 | 17:53 |
| clarkb | this may be a local fetch and review change | 17:53 |
| AJaeger | clarkb: oh yeah - sorry, should have mentioned... | 17:53 |
| larsks | fungi: got it. | 17:54 |
| *** auristor has joined #openstack-infra | 17:55 | |
| corvus | dhellmann, fungi, clarkb, AJaeger, mordred: regarding the plan to move-template-invocations-out-of-project-config / python3 goal -- it occurs to me that the release jobs are different. those pipelines don't deal with branches, so zuul is going to grab all of the relevant project stanzas. so if we go with the approach of having a "publish-to-pypi-python3" template on master and "publish-to-pypi" | 17:56 |
| corvus | template on stable, for a release, zuul is going to collect both of those templates together and run both jobs. | 17:56 |
| clarkb | oh right refs/tags doesn't contain branch info | 17:57 |
| AJaeger | corvus: oh... any other jobs to tread this way? | 17:57 |
| AJaeger | corvus: good catch! | 17:57 |
| *** larsks has left #openstack-infra | 17:57 | |
| clarkb | I want to say it is pipeline specific | 17:57 |
| corvus | AJaeger: anything that runs in a release or tag pipeline | 17:58 |
| AJaeger | yep | 17:58 |
| corvus | yeah, it's pipelines that don't have branches | 17:58 |
| corvus | i *think* that doesn't apply to post, because we know the target branch, but i want to double check that | 17:58 |
| *** lpetrut has quit IRC | 17:59 | |
| fungi | specific to the kind of ref, yeah | 17:59 |
| corvus | yeah, post is okay, it uses the target branch | 18:00 |
| fungi | tags certainly aren't guaranteed to map to one and only one branch | 18:00 |
| fungi | so we can't reliably guess one for them | 18:00 |
| *** edmondsw has joined #openstack-infra | 18:01 | |
| corvus | i think there's two semi-independent questions here: 1) do we want to try to do run different release jobs for different "branches"? and 2) do we want to configure that in-repo or not? | 18:02 |
| fungi | instead we probably need to have some means of signalling in the repository which python to use when building release artifacts | 18:02 |
| corvus | fungi: yeah, alter my question #1 to include that possibility | 18:02 |
| corvus | i guess what i was really trying to ask is: 1) do we want to try to run different python versions to build releases off of different branches? | 18:03 |
| fungi | so projects which are wanting to use python3 can have some way to tell us that in some relevant file in the master branch and the job can choose the right python based on that | 18:03 |
| *** heyongli has quit IRC | 18:03 | |
| *** heyongli has joined #openstack-infra | 18:03 | |
| fungi | i think the "why" we want that is so that we can default to exercising python3 on as many python-based jobs as possible, while accepting that some of our stable branches may indefinitely only be able to build artifacts under python2 | 18:04 |
| fungi | i.e., too-old build dependencies, python2isms in setup.py, et cetera | 18:05 |
| clarkb | AJaeger: the two changes lgtm | 18:05 |
| *** cfriesen has joined #openstack-infra | 18:05 | |
| AJaeger | thanks, clarkb - hope I get another review for them to avoid merge conflicts and allow smaller changes ;) | 18:06 |
| fungi | though an interesting feature request there is... can zuul look at configuration in the content of a tagged repository state for pipelines which act on tag refs? | 18:06 |
| *** janki has quit IRC | 18:06 | |
| fungi | that might be insane/unworkable, just brainstorming | 18:06 |
| *** edmondsw has quit IRC | 18:06 | |
| dhellmann | corvus : I suspect we're going to need to be able to run the python2 job on older branches, but I may be over cautious there | 18:07 |
| corvus | fungi: theoretically it could. it would be somewhat resource intensive. | 18:07 |
| cfriesen | hope this is the right place to ask...I'm trying to figure out why https://review.openstack.org/#/c/578414/ isn't merged yet. I don't see anything useful at zuul.openstack.org but maybe I'm not doing it right. | 18:07 |
| dhellmann | fungi : we could also just have 1 job that reads a config file in the repo itself, right? | 18:07 |
| fungi | dhellmann: right, that's basically what i was suggesting initially | 18:07 |
| dhellmann | ok, I have quickly skimmed the scrollback so I missed that | 18:08 |
| AJaeger | cfriesen: is the "depends-on" change merged? | 18:09 |
| fungi | cfriesen: https://review.openstack.org/578414 was approved at 15:58z but declares a depends-on for a change (in another project with which it seems to not share a change queue) which didn't merge until 16:03z | 18:09 |
| dhellmann | corvus , fungi : it sounds like the simplest solution is what AJaeger proposes in https://review.openstack.org/#/c/578478/2/zuul.d/projects.yaml and just keep the release jobs in project-config along with the other things that we only want to have configured one time | 18:10 |
| corvus | fungi: essentially the process would be: ask the mergers for the in-repo config and dynamically reconfigure with cached config from all branches removed, and the newly read config added in. | 18:10 |
| dhellmann | although that doesn't help with the per-branch versioning question | 18:10 |
| fungi | cfriesen: you might try reapproving 578414 by adding a new workflow +1 or getting the original approver to remove and readd theirs | 18:10 |
| AJaeger | dhellmann: I just gave that -1 quickly based on discussion here - can change that anytime ;) | 18:10 |
| dhellmann | AJaeger : ok, yeah. I thought that was the right approach but now I'm not so sure. | 18:11 |
| clarkb | cfriesen: 2018-06-27 15:58:27,054 DEBUG zuul.Pipeline.openstack.gate: Failed to enqueue changes ahead of <Change 0x7f2a34288ac8 578414,1> seems to be the underlying issue, looking t osee if I can find more data than that | 18:11 |
| *** yolanda_ has joined #openstack-infra | 18:11 | |
| dhellmann | what if I add branch specifiers to the old release jobs so they will run under stable branches up through rocky and then we disable it on master and do a hard-cut over to the python3 version during stein? | 18:12 |
| dhellmann | oh, we don't know the branch, that's the whole point, isn't it | 18:12 |
| clarkb | cfriesen: https://review.openstack.org/#/c/578413 is the depends on of 578414, they don't share a gate queue between those projects so 413 had to merge before you can enqueue 414 to the gate | 18:12 |
| * dhellmann spins in circles | 18:12 | |
| clarkb | cfriesen: if you approve 414 now it should enqueu | 18:12 |
| corvus | dhellmann: i think keeping all unbranched jobs in project-config is viable. it's also viable to put them in repo with the caveat that if you change it in one branch, you need to backport that change. but you can see with that you have the same potential problem either way in that regardless, there isn't a way to have a "branched release job". | 18:12 |
| fungi | clarkb: i _think_ that's the message it logs when the depends-on isn't merged yet and doesn't share a queue with the current change | 18:12 |
| clarkb | fungi: yup | 18:13 |
| cfriesen | clarkb: okay...what's the "proper" way to handle this sort of thing when there are dependencies between projects? | 18:13 |
| cfriesen | and thanks for the help | 18:13 |
| *** heyongli has quit IRC | 18:13 | |
| dhellmann | corvus : are there other classes of jobs that are "unbranched" like the release jobs? | 18:13 |
| AJaeger | cfriesen: if they share a queue: You're fine. If not, approve a change with "depends-on" only once the dependend one is in. | 18:13 |
| dhellmann | or maybe other "instances" not classes | 18:13 |
| *** heyongli has joined #openstack-infra | 18:14 | |
| clarkb | cfriesen: if you don't want them sharing a queue then you just have to approve them after they merge in series. If you are ok with them sharing a queue like nova cinder neutron etc then you can specify a common queue name in your config | 18:14 |
| fungi | cfriesen: if you think these projects are likely to have interdependencies, you may want to declare a shared change queue for them | 18:14 |
| clarkb | cfriesen: let me find an example of that | 18:14 |
| *** Shrews has quit IRC | 18:14 | |
| *** trown|lunch is now known as trown | 18:14 | |
| *** yolanda has quit IRC | 18:14 | |
| corvus | dhellmann: i think it's anything that runs in the tag or release pipeline. post doesn't have this issue (it knows the branch). that makes docs publishing interesting -- you can have branched docs publishing jobs in the post pipeline, but the problem appears when you run the same job/jobs in release (you lose the branch context) | 18:15 |
| fungi | cfriesen: sharing a queue means that zuul will guarantee they're tested as if they will merge serially (which brings some additional merge delays as it has to make sure the changes ahead do in fact merge first), but for fairly tightly-coupled projects that's often what you want | 18:15 |
| *** Shrews has joined #openstack-infra | 18:15 | |
| cfriesen | fungi: makes sense | 18:16 |
| corvus | dhellmann, clarkb, fungi, AJaeger: note that you *can* have branch specifiers for jobs that run on tags, but they would match the *tag* -- ie, you can say "run this on 2.x, run that on 3.x" | 18:16 |
| clarkb | cfriesen: https://git.openstack.org/cgit/openstack-infra/project-config/tree/zuul.d/projects.yaml#n108 the queue: integrated there is how we set up the integrated gate between nova and friends. You can set queue: startlingx or similar on your projects to have them share one | 18:16 |
| clarkb | corvus: ah | 18:16 |
| dhellmann | corvus : ok, so that means the release notes jobs, the doc build jobs, and the packaging jobs | 18:16 |
| *** e0ne has joined #openstack-infra | 18:16 | |
| corvus | dhellmann: sounds about right | 18:16 |
| dhellmann | I'm inclined to encourage teams to just backport whatever changes they need to make those jobs work, but I don't have a good idea of what is included. | 18:16 |
| dhellmann | and to also say those need to stay configured in project-config so that they are only in one place and we don't end up with duplication | 18:17 |
| dhellmann | running multiple versions | 18:17 |
| cfriesen | clarkb: thanks | 18:17 |
| fungi | dhellmann: aside from build-deps which lack python3 support (hopefully there aren't any?) that will probably work out okay | 18:17 |
| dhellmann | and at *that* point I wonder how much configuration we would actually be moving out of project-config | 18:17 |
| AJaeger | corvus, dhellmann: we don't run docs in release - only in post - so, we should be fine. Or do we have it in release anywhere? | 18:18 |
| corvus | dhellmann: yeah, if the process is "fix up your project to work with py3 in all branches, then flip the switch for the project as a whole" i think keeping those in p-c for now is the best fit. | 18:18 |
| dhellmann | we may end up doing a lot less work if we just move the jobs that aren't in templates | 18:18 |
| dhellmann | AJaeger : it runs on tag events so we get docs that include the tagged version, right? or did we change that? | 18:18 |
| dhellmann | I know we do release notes that way still | 18:18 |
| fungi | dhellmann: if there are projects which need to install their dependencies to be able to generate files they're embedding in an sdist or a wheel that might be a problem, but that seems like an edge case | 18:18 |
| dhellmann | fungi : hmm. do we still support that anywhere? | 18:19 |
| fungi | i hope not? | 18:19 |
| dhellmann | the new python3 packaging job would not support it | 18:19 |
| AJaeger | dhellmann: publish-openstack-sphinx-docs does not run on tags anymore | 18:20 |
| dhellmann | AJaeger : ah, good | 18:20 |
| dhellmann | still | 18:20 |
| AJaeger | dhellmann: release-notes-jobs runs on tag - perhaps we should change that for symmetry? | 18:20 |
| *** Parvej has quit IRC | 18:20 | |
| *** yamahata has joined #openstack-infra | 18:20 | |
| dhellmann | part of the reason for doing all of this moving around was to make it easier to change the jobs that we're now saying we don't want to move | 18:20 |
| dhellmann | which makes me think we should just drop that whole section of the plan | 18:21 |
| *** myoung|lunch is now known as myoung | 18:21 | |
| *** rh-jelabarre has joined #openstack-infra | 18:22 | |
| dhellmann | AJaeger : we need the release notes job to run on tag so that when we do a final release we combine the release notes for all of the pre-release versions into the final version | 18:22 |
| *** bobh has quit IRC | 18:22 | |
| corvus | dhellmann: still move the check/gate jobs in-repo? | 18:23 |
| *** bobh has joined #openstack-infra | 18:23 | |
| dhellmann | corvus : we could, but doing that doesn't simplify anything having to do with the underlying goal so I'm less inclined to roll all of the work together into 1 goal | 18:23 |
| *** heyongli has quit IRC | 18:23 | |
| dhellmann | those aren't the jobs we need them to change -- we need the packaging, doc, and release notes jobs changed | 18:24 |
| clarkb | your goal specifically being python3 first | 18:24 |
| *** heyongli has joined #openstack-infra | 18:24 | |
| *** dpawlik has quit IRC | 18:24 | |
| dhellmann | yes | 18:24 |
| corvus | dhellmann: ok, wasn't sure what the expected work balance was between adding py3 test jobs to branches vs the others. | 18:24 |
| corvus | sounds like it's mostly the others | 18:24 |
| dhellmann | most of the projects already have at least some python3 test jobs | 18:24 |
| *** dpawlik has joined #openstack-infra | 18:24 | |
| clarkb | I still think infra should push back on changes that can/should go in repo particularly if projects are pinging us due to blocking them (when we aren't) | 18:24 |
| dhellmann | I'm OK with that | 18:24 |
| corvus | yeah, just doesn't have to be part of the tc goal :) | 18:25 |
| dhellmann | I just have trouble considering it a blocker for the python3 stuff | 18:25 |
| dhellmann | we could that as a separate goal, even | 18:25 |
| clarkb | corvus: dhellmann yup and doesn't have to happen all at once | 18:25 |
| clarkb | I think it will happen relatively quickly if we advertise it as an option | 18:26 |
| dhellmann | I have a nifty tool to figure out which bits can move, too | 18:26 |
| dhellmann | https://review.openstack.org/#/c/578194/ | 18:26 |
| *** Shrews has quit IRC | 18:28 | |
| dhellmann | so we're left with a question about how to test the changes to the jobs | 18:28 |
| smcginnis | If someone other than Andreas has a moment, I would appreciate it if I could get eyes on https://review.openstack.org/#/c/578449/ to get some follow on work moving. | 18:28 |
| dhellmann | if I switch the release notes build with a change in project-config, I can't test that until it is approved, right? | 18:28 |
| dhellmann | I guess someone could add it to a patch in repo | 18:28 |
| *** dpawlik has quit IRC | 18:28 | |
| corvus | dhellmann: yep, i think that's about it | 18:29 |
| *** Shrews has joined #openstack-infra | 18:29 | |
| dhellmann | hmm | 18:30 |
| dhellmann | ok, so I guess I will work up the directions for doing that instead of the current plan | 18:30 |
| dhellmann | corvus : does zuul not look at the commit being tagged to determine the jobs it should run? | 18:31 |
| dhellmann | like if I propose a patch against master, I would expect it to read the zuul settings out of the master branch | 18:31 |
| dhellmann | and if I tag something, I would expect it to read the settings out of the commit being tagged | 18:31 |
| *** e0ne has quit IRC | 18:32 | |
| *** Shrews has quit IRC | 18:33 | |
| fungi | dhellmann: nope, that's basically the fanciful feature request i posted earlier | 18:33 |
| corvus | dhellmann: zuul does not look at the content of the commit unless it changes zuul's configuration, and even then it merges that configuration with everything else. | 18:33 |
| dhellmann | fungi : well no wonder I'm always so confused about how it decides what to run :-) | 18:33 |
| corvus | basically, zuul's configuration is global and exists outside of the context of any given commit | 18:33 |
| *** Shrews has joined #openstack-infra | 18:33 | |
| *** heyongli has quit IRC | 18:34 | |
| *** heyongli has joined #openstack-infra | 18:34 | |
| dhellmann | ok. that's a different model than I understood | 18:35 |
| corvus | that's at least somewhat necessary in order to implement things like "allow project-config to specify what jobs to run on nova changes". but as discussed with fungi earlier, that doesn't have to be the end of the story. :) | 18:36 |
| corvus | so fungi's suggestion is something we can do in v3 but couldn't in v2 | 18:37 |
| corvus | but a potential change to v3 that also could have worked in v2 would be to try to figure out what "branch" the tag is on. the question is, what if the answer comes back both "master and stable/foo"? | 18:38 |
| corvus | or "none" | 18:38 |
| fungi | and also comes with the potential for new security risks we'd want to think through _very_ carefully | 18:38 |
| dhellmann | For a lot of cases, I would actually be happy enough just having individual jobs look at their own config files in the repo. | 18:38 |
| fungi | though i guess assuming we expect that security controls around pushing tags are at least as tight as around approving changes and pull requests, then that's probably not representing any new risks | 18:39 |
| dhellmann | that makes the jobs more complicated, and I would need to modify the python3 job, again, but I could do that and then switch all projects using the python2 job to the python3 job in project-config if the default is python2 (which would be weird for a job called python3 so I guess I would rename it) | 18:39 |
| dhellmann | we could probably define 1 role that all of the relevant jobs could use | 18:40 |
| dhellmann | but even doing that doesn't make moving those job settings into the project tree make sense to me if we're going to have problems in the future switching from one job to another again | 18:41 |
| dhellmann | I guess we would have to just keep changing each job definition to look for new parameters | 18:41 |
| *** rlandy is now known as rlandy|afk | 18:41 | |
| dhellmann | (in the configuration file) | 18:41 |
| openstackgerrit | Merged openstack-infra/project-config master: Add openstack/operations-guide to zuul https://review.openstack.org/578449 | 18:42 |
| corvus | yeah, that's a reasonable-sounding solution too. gets at the question: "is the job 'build the package' or 'build the package a certain way'?" | 18:43 |
| dhellmann | corvus : I appreciate you bringing this up before we got too far down the wrong path. At the same time, I feel like that's the 2nd (or more?) time it has happened sort of accidentally. How can I get some of your time to sit down with me (virtually or otherwise) to make sure we have a good plan before we all travel to denver? | 18:43 |
| dhellmann | ideally the conversion for oslo would be done before we start anyone else, because the oslo team has agreed to go early and test all of this | 18:43 |
| corvus | dhellmann: heh, i had the same feeling and was thinking we should totally sit down in denver. i should revise my mental timetable. :) | 18:44 |
| dhellmann | yeah, I need to be ready to explain all of this to other people by then | 18:44 |
| openstackgerrit | Sean McGinnis proposed openstack-infra/project-config master: Readd openstack/operations-guide repo https://review.openstack.org/578441 | 18:44 |
| *** heyongli has quit IRC | 18:44 | |
| *** tobasco_afk is now known as tobasco | 18:44 | |
| *** heyongli has joined #openstack-infra | 18:44 | |
| *** e0ne has joined #openstack-infra | 18:44 | |
| *** felipemonteiro_ has quit IRC | 18:46 | |
| corvus | dhellmann: i can certainly take a few hours and really dig into the proposal as written, and maybe have a call with you so we can say lots of words at each other which will hopefully increase understanding and give ample opportunity to catch pitfalls. but also, i think proceeding with oslo.config early and exercising it will be important. 2 weeks ago, i was thinking about this a lot, and still | 18:48 |
| corvus | didn't quite connect these two dots, until today when i was working on a change zuul branch matching code. going through the process completely (ie oslo) is probably the best way to shake things out. | 18:48 |
| dhellmann | so the release job runs the build-python-release role from zuul-jobs which uses the release_python variable to determine which python to use. is there some way to set that variable from another role? | 18:48 |
| dhellmann | ok, if we think things are really likely to break I will probably pick something other than oslo.config to test with; maybe oslo.i18n since we don't see many changes there | 18:49 |
| corvus | dhellmann: yes you can use the set_fact task in ansible to override the variable, but note that it will only have effect within the same playbook. | 18:49 |
| *** felipemonteiro_ has joined #openstack-infra | 18:49 | |
| openstackgerrit | Sean McGinnis proposed openstack-infra/project-config master: Readd openstack/operations-guide repo https://review.openstack.org/578441 | 18:49 |
| dhellmann | ok, so we can define a role that reads a file from the repo and uses it to set the variable and add that role to all of the playbooks that need it | 18:50 |
| dhellmann | and then adding the patch to set the configuration will be somewhat self-testing | 18:50 |
| openstackgerrit | Paul Belanger proposed openstack-infra/zuul master: Add min_avail_hdd governor for zuul-executor https://review.openstack.org/577948 | 18:50 |
| fungi | that should work, yes | 18:51 |
| corvus | dhellmann: it might be good to have a test case which isn't... perfect :) ie, if we can make releases from an old branch which breaks under py3.... | 18:51 |
| dhellmann | I don't know of any projects that actually break like that. I'm just assuming that there is one. | 18:51 |
| fungi | would be easy enough to adjust the setup.py in the release-test repo to break if sys.version>=3 or something | 18:52 |
| corvus | we will know after denver | 18:52 |
| dhellmann | does a task know what playbook is calling it? | 18:52 |
| dhellmann | fungi : oh, sure, good idea | 18:52 |
| dhellmann | I'm thinking of how this configuration file will work. We probably want a separate setting for each job, right? so if the job or playbook name is somehow knowable then the role can use that to decide which value to read out of the file | 18:53 |
| corvus | dhellmann: i'm sure there's a way, but that sounds sort of like a "layer violation" | 18:53 |
| dhellmann | alternatively we need a *separate* variable to tell the role what to read in every place we add the role to a playbook? | 18:54 |
| dhellmann | or it's all or nothing | 18:54 |
| fungi | i feel like a single switch is probably fine? | 18:54 |
| *** heyongli has quit IRC | 18:54 | |
| corvus | all or nothing doesn't sound bad? we're basically asserting that the project can be released under py3? | 18:54 |
| fungi | this is mostly as a workaround for not being able to exclude some older, frozen code | 18:54 |
| dhellmann | and the release notes build and the docs build and ... | 18:54 |
| *** heyongli has joined #openstack-infra | 18:55 | |
| dhellmann | so they may have to create a series of patches in order to turn the switch on | 18:55 |
| dhellmann | but that's not the end of the world, I suppose | 18:55 |
| fungi | so you know, if you can't build tarballs under python3 on stable/pike and that means we also just don't build your stable/pike docs under python3, is that a problem? | 18:55 |
| dhellmann | I'm thinking about how to turn the flag on in the master branch, when several otherwise unrelated code changes may be needed to make it work | 18:56 |
| fungi | we should be able to assume people will have the freedom to make master work, and then flip the "it's done" switch | 18:56 |
| dhellmann | yeah, that works for me | 18:56 |
| dhellmann | one hacky thing I've done before is look at the trove classifiers in setup.cfg | 18:56 |
| fungi | the tricky bit is that exercising release jobs on random repos isn't easy outside of actually tagging new releases | 18:56 |
| corvus | basepython= in tox.ini could be a signal too | 18:56 |
| dhellmann | although that feels potentially dangerous in this case | 18:56 |
| dhellmann | corvus : there are many instances of basepython in the file, so I'm not sure which one to use | 18:57 |
| dhellmann | fungi : yeah, we can encourage folks to turn the flag on and ask for an alpha tag | 18:57 |
| dhellmann | oh, actually, there's a check packaging job now | 18:57 |
| fungi | right, until tox 3.1 fixes the need to have basepython in every testenv | 18:57 |
| dhellmann | that's part of the new job template | 18:57 |
| corvus | dhellmann: i was thinking if it showed up in plain old [testenv] | 18:57 |
| dhellmann | corvus : yeah. it almost never will, though, because until tox 3.1 that will break all of the other envs | 18:58 |
| dhellmann | py27 would use that unless there was an override, etc. | 18:58 |
| dhellmann | and only some of the projects are using explicit environments for the ones tox generates by default | 18:58 |
| fungi | the baseputhon feature is basically buggy/misimplemented | 18:58 |
| fungi | er, basepython | 18:58 |
| dhellmann | yeah | 18:58 |
| dhellmann | so I think a new file is going to be the simplest thing to understand | 18:59 |
| corvus | oh neat, didn't know that. | 18:59 |
| fungi | in retrospect we shouldn't have started to rely on it | 18:59 |
| dhellmann | I will call it .openstack.ini and it will live in the root of the repo and we can put other things in it later | 18:59 |
| fungi | if you're a python3-only project and only test against the default available "python3" in a given context then basepython=python3 in your [testenv] works fine | 18:59 |
| corvus | dhellmann: could you add it as a new thing to setup.cfg? | 19:00 |
| dhellmann | I guess? it's not really setup.py or pbr related. | 19:00 |
| dhellmann | is there an ansible module for reading values out of yaml files? | 19:01 |
| corvus | yeah, you can read yaml, json, ini easily | 19:01 |
| fungi | i'd also be wary of extending setup.cfg since that's also now something setuptools can use and we're trying to maintain reasonable parity between setuptools and pbr config section and option names in there | 19:01 |
| dhellmann | I know there's one for ini | 19:01 |
| fungi | could we stick a shebang in setup.py? | 19:02 |
| fungi | and then parse it | 19:02 |
| dhellmann | what's wrong with a new file that is clearly for our job settings? | 19:02 |
| dhellmann | maybe .openstack-jobs.ini | 19:03 |
| fungi | nothing particularly, i suppose other than metafile fatigue | 19:03 |
| fungi | we already put in confgs for setuptools/pbr, zuul, coverage, tox... what's one more? | 19:04 |
| fungi | git-review | 19:04 |
| dhellmann | hmm, there's an ini "plugin" but I don't see a module for reading from a file like that | 19:04 |
| corvus | it makes how you build an openstack project more openstack specific | 19:04 |
| *** heyongli has quit IRC | 19:04 | |
| *** heyongli has joined #openstack-infra | 19:05 | |
| corvus | to be fair, the alternative is "you just know to use python3" which isn't great | 19:05 |
| dhellmann | *ideally* it doesn't matter | 19:05 |
| dhellmann | can I set variables in the zuul configuration files somehow? | 19:06 |
| corvus | i'm not too worried about this one. i think the thing i'm most apprehensive about is if .openstack-jobs.ini becomes a thing that has more than just this setting. | 19:06 |
| fungi | that's sort of been my argument against some of the pti changes to stop building docs and sdists/wheels in tox is we've traded "here's a wrapper which does the special setup you're going to need" for "you need to do special things and they're documented over here" | 19:06 |
| dhellmann | there's already a disconnect between what's in the repo and how to build it. Knowing the incantation of sphinx, for example. | 19:07 |
| dhellmann | right, exactly, fungi | 19:07 |
| corvus | dhellmann: yes, you can set "vars" on jobs, but that likely has the same branch context problems as changing the jobs themselves | 19:07 |
| dhellmann | ah, yeah | 19:07 |
| dhellmann | of course | 19:07 |
| fungi | not to bikeshed too hard on this, but maybe if the filename made it clear this was a single-purpose, transitional flag file? | 19:08 |
| dhellmann | like at the end of the cycle we switch the jobs to use python3 by default and then ignore the file? | 19:08 |
| fungi | you wouldn't even need a parser then, just test for the existence of .run-my-jobs-with-python3 (name to be chosen by someone better at this) | 19:09 |
| dhellmann | so maybe .python3-first.ini | 19:09 |
| dhellmann | ah, yeah | 19:09 |
| *** hasharAway is now known as hashar | 19:09 | |
| openstackgerrit | Dean Troyer proposed openstack-infra/project-config master: Put StarlingX projects into a common gate queue https://review.openstack.org/578532 | 19:10 |
| corvus | i have to go forage for food now | 19:12 |
| fungi | and that way, once the job is obsolete, let projects know they're free to delete it at their convenience | 19:12 |
| dhellmann | yeah | 19:12 |
| fungi | er, once the file is obsolete | 19:12 |
| fungi | which of course will result in somebody with a qq.com or 163.com e-mail address proposing deletion patches to 800 repositories | 19:13 |
| fungi | but whatever. cleanup like that is always going to involve mass changes | 19:13 |
| dhellmann | ok, I'll spend a little time figuring out what changes are needed in the relevant jobs | 19:14 |
| *** heyongli has quit IRC | 19:15 | |
| dhellmann | I'll want to rename the job and the project-template, so that means duplicating some stuff until I can replace the use of the old name with the new one, I suppose | 19:15 |
| *** heyongli has joined #openstack-infra | 19:15 | |
| dtroyer | Is https://review.openstack.org/#/c/572793/ (creating git.starlingx.io) lacking something to move forward? Possibly policy and not technical? | 19:15 |
| fungi | or reviewer bandwidth | 19:17 |
| dhellmann | corvus , fungi , clarkb : I have moved the goal description to https://etherpad.openstack.org/p/goal-python3-first so it will be easier to collaborate on changes to the process | 19:17 |
| fungi | dtroyer: oh, i think it's also delayed waiting for someone to get an ssl cert issued | 19:18 |
| dtroyer | ah, ok. thanks | 19:19 |
| *** Shrews has quit IRC | 19:21 | |
| fungi | clarkb: not to heap yet one more cert purchase on you, but 572793 needs one | 19:22 |
| clarkb | ya | 19:22 |
| clarkb | will need to figure out the verification for that one too | 19:23 |
| * clarkb checks whois | 19:23 | |
| clarkb | simiarly lacking whois contact info, will have to do dns based verification too | 19:23 |
| *** heyongli has quit IRC | 19:25 | |
| *** heyongli has joined #openstack-infra | 19:25 | |
| *** Shrews has joined #openstack-infra | 19:26 | |
| clarkb | dtroyer: fungi: I'll do that after lunch today, get hiera updated and approve the change | 19:27 |
| fungi | thanks clarkb! | 19:28 |
| clarkb | dtroyer: fungi is there an update to project-config queued too to set the cgit-alias data? | 19:30 |
| AJaeger | clarkb: https://review.openstack.org/572801 | 19:30 |
| AJaeger | clarkb: so, yes | 19:30 |
| fungi | excellent | 19:30 |
| fungi | thanks AJaeger! | 19:30 |
| AJaeger | dtroyer: a "needed-by" would have been nice to find that directly | 19:30 |
| auristor | as a result of the EU's GDPR whois is now useless | 19:31 |
| fungi | yep, we've noticed | 19:31 |
| fungi | or at least increasingly useless | 19:31 |
| fungi | whois for some of my .com/.net domains still have all my contact info, as does a domain i have in .is | 19:32 |
| fungi | but my .org domains are all now devoid of useful contact info | 19:32 |
| auristor | it depends on the registrar but if the registrar does business with EU persons it has little choice but to hide the data | 19:33 |
| openstackgerrit | Andreas Jaeger proposed openstack-infra/project-config master: Use wildcard to apply system-required template https://review.openstack.org/577307 | 19:33 |
| AJaeger | clarkb: had to move the regex to the top for alphabetical order, could you review again, please? ^ | 19:34 |
| fungi | my registrar is dyndns, now owned by oracle. i'm sure they do business with the eu, but some registrars are only filtering contact info for registrants with eu addresses | 19:34 |
| clarkb | auristor: my understadning is that they only have to remove it if requested | 19:34 |
| clarkb | auristor: and explain what the data is used for otherwise | 19:34 |
| *** rh-jelabarre has quit IRC | 19:34 | |
| fungi | also, yes, a lot of the registries within the eu are still publishing direct e-mail contact info for domains under their tlds and just removing them when requested by the registrant | 19:34 |
| dhellmann | fungi : regarding deleting that file, we basically can't until stable/rocky is closed. So, effectively never. | 19:35 |
| dhellmann | or at least far enough in the future that I don't expect to be around to remember to do it | 19:35 |
| *** heyongli has quit IRC | 19:35 | |
| fungi | dhellmann: well, we can... keep in mind that the extended maintenance guidelines require those branches to keep up with changes in testing | 19:35 |
| *** heyongli has joined #openstack-infra | 19:36 | |
| dhellmann | so you're saying that we have to make all the old branches work under python 3 for packaging, docs, and release notes? | 19:36 |
| dhellmann | that was not my plan. | 19:36 |
| fungi | we could still set a deadline and if they don't fix the branches before then those jobs will stop being useful for them | 19:36 |
| dhellmann | I don't want to prevent releases from those older branches as a result of this transition. | 19:36 |
| fungi | the deadline could be when we retire xenial images, for example | 19:36 |
| dhellmann | when is that likely to be? | 19:37 |
| fungi | we agreed we weren't going to keep tagging releases on em branches | 19:37 |
| fungi | (or so i thought?) | 19:37 |
| dhellmann | they aren't "em" yet though | 19:37 |
| fungi | sure, they reach em at the old "eol" date | 19:37 |
| fungi | i assumed the transitional period to keep those files around was going to be on the order of 18-24 months | 19:38 |
| dhellmann | ok, I was thinking end of cycle | 19:38 |
| dhellmann | so different expectations | 19:38 |
| dhellmann | 18-24 months seems ok | 19:38 |
| dhellmann | I anticipate still being present to remember to do something then :-) | 19:38 |
| *** jamesmcarthur has joined #openstack-infra | 19:38 | |
| fungi | but you don't need to be. just put a #TODO comment in with the obsolescence date | 19:39 |
| openstackgerrit | boden proposed openstack-infra/project-config master: increase timeout on neutron-lib periodic for vmware-nsx https://review.openstack.org/578539 | 19:40 |
| fungi | can be the content of those flag files even... #TODO this file is scheduled to be obsolete after 2020-07-01, but check that <foo> is set to <bar> in <baz xyzzy quux> before removing | 19:40 |
| dtroyer | clarkb: thanks | 19:41 |
| dtroyer | AJaeger: roger that, thanks | 19:41 |
| clarkb | fungi: dtroyer the best part is I already filed an expense report for last weeks certs :P | 19:41 |
| dhellmann | fungi : oh, sure, good point | 19:41 |
| clarkb | that will learn me | 19:41 |
| AJaeger | boden: could you rebase https://review.openstack.org/575399 , please? | 19:42 |
| fungi | clarkb: you filed that fast! | 19:42 |
| fungi | i'll add the cname in dns | 19:43 |
| fungi | or did you want to do that while you're in there adding the ca temporary record? | 19:43 |
| fungi | it'll be the same zone either way | 19:43 |
| clarkb | fungi: ya I can do both updates | 19:43 |
| clarkb | don't need dns until that change has certs and is approved anyway | 19:43 |
| clarkb | fungi: I've found filing them right away is the best way to remember to file them :) | 19:44 |
| *** Shrews has quit IRC | 19:44 | |
| fungi | i set a reminder on my calendar to file them during a normally free hour on the friday before payday | 19:45 |
| *** heyongli has quit IRC | 19:45 | |
| *** heyongli has joined #openstack-infra | 19:46 | |
| fungi | since it's less work for me to batch up the last two weeks' expenses | 19:46 |
| *** jamesmcarthur has quit IRC | 19:46 | |
| clarkb | I think I need to update certcheck to check airship and starlingx git certs too | 19:47 |
| openstackgerrit | boden proposed openstack-infra/project-config master: remove lower constraints to vmware-nsx gate pipeline https://review.openstack.org/575399 | 19:47 |
| fungi | oh, yep | 19:47 |
| *** edmondsw has joined #openstack-infra | 19:50 | |
| *** Shrews has joined #openstack-infra | 19:53 | |
| *** jamesmcarthur has joined #openstack-infra | 19:54 | |
| *** edmondsw has quit IRC | 19:54 | |
| *** heyongli has quit IRC | 19:56 | |
| *** heyongli has joined #openstack-infra | 19:56 | |
| *** jamesmcarthur has quit IRC | 20:01 | |
| dhellmann | corvus , fungi , clarkb : I believe I have a complete new draft in https://etherpad.openstack.org/p/goal-python3-first when you have time to look it over | 20:02 |
| dhellmann | I left some of the old content there struck out to make the changes clearer | 20:02 |
| *** jamesmcarthur has joined #openstack-infra | 20:03 | |
| *** heyongli has quit IRC | 20:06 | |
| *** heyongli has joined #openstack-infra | 20:06 | |
| *** eharney has quit IRC | 20:13 | |
| *** Hal has joined #openstack-infra | 20:14 | |
| *** tpsilva has quit IRC | 20:14 | |
| *** Hal is now known as Guest80661 | 20:14 | |
| *** ihar has quit IRC | 20:16 | |
| *** heyongli has quit IRC | 20:16 | |
| *** heyongli has joined #openstack-infra | 20:17 | |
| *** boden has quit IRC | 20:17 | |
| *** dpawlik has joined #openstack-infra | 20:22 | |
| *** heyongli has quit IRC | 20:26 | |
| *** heyongli has joined #openstack-infra | 20:27 | |
| *** zul has quit IRC | 20:27 | |
| *** boden has joined #openstack-infra | 20:29 | |
| *** Guest80661 has quit IRC | 20:29 | |
| *** heyongli has quit IRC | 20:37 | |
| *** heyongli has joined #openstack-infra | 20:37 | |
| clarkb | dtroyer: fungi just waiting on cert signing verification now | 20:38 |
| clarkb | I'm taking the hieradata lock | 20:40 |
| openstackgerrit | Sean McGinnis proposed openstack-infra/project-config master: Readd openstack/operations-guide repo https://review.openstack.org/578441 | 20:40 |
| *** jesslampe has quit IRC | 20:40 | |
| *** kgiusti has left #openstack-infra | 20:40 | |
| *** e0ne has quit IRC | 20:42 | |
| clarkb | the biggest downside to dns record verification so far is that you have no way to trigger the poll | 20:42 |
| clarkb | so you sit around waiting and hoping you copy pastad the record details properly | 20:43 |
| *** jamesmcarthur has quit IRC | 20:43 | |
| *** yolanda_ has quit IRC | 20:43 | |
| *** yolanda_ has joined #openstack-infra | 20:44 | |
| *** heyongli has quit IRC | 20:47 | |
| *** heyongli has joined #openstack-infra | 20:47 | |
| *** jamesmcarthur has joined #openstack-infra | 20:49 | |
| openstackgerrit | Colleen Murphy proposed openstack-infra/puppet-nodepool master: Update libxslt package name for Xenial https://review.openstack.org/577519 | 20:49 |
| fungi | yet one more thing where working with computers involves sitting around and waiting | 20:52 |
| *** yolanda has joined #openstack-infra | 20:53 | |
| *** jamesmcarthur has quit IRC | 20:53 | |
| clarkb | alright hiera lock released | 20:54 |
| *** yolanda_ has quit IRC | 20:55 | |
| clarkb | I've approved the puppet change | 20:55 |
| *** trown is now known as trown|outtypewww | 20:55 | |
| *** camunoz has quit IRC | 20:56 | |
| clarkb | DNS updated too | 20:57 |
| *** boden has quit IRC | 20:57 | |
| *** heyongli has quit IRC | 20:57 | |
| lbragstad | o/ i went to push a trivial fix to openstack/python-keystoneclient-kerberos (which is apparently read-only), i assume we won't be able to do something like this? http://paste.openstack.org/show/724453/ | 20:58 |
| *** heyongli has joined #openstack-infra | 20:58 | |
| lbragstad | it looks read-only because it's technically retired since keystoneauth1 took responsibility of the functionality | 20:58 |
| *** yolanda_ has joined #openstack-infra | 20:59 | |
| johnsom | Is there a link I can grab the latest nodepool xenial image? I want to use the exact image the gates are using in tracking down this nested virt kvm bug. | 20:59 |
| johnsom | I tried hunting for a link last night, but didn't find one. | 21:00 |
| openstackgerrit | Clark Boylan proposed openstack-infra/system-config master: Update certcheck to include new domains https://review.openstack.org/578556 | 21:00 |
| clarkb | johnsom: https://nl01.openstack.org/images and https://nl02.openstack.org/images | 21:01 |
| johnsom | Awesome, thank you. | 21:01 |
| clarkb | johnsom: we have distributed builders now so any given image will be on one of them | 21:01 |
| clarkb | lbragstad: if it has been retired you'll have to unretire it | 21:02 |
| fungi | in the future the dashboard will hopefully be able to link to them too | 21:02 |
| *** dpawlik has quit IRC | 21:02 | |
| fungi | (nodepool images) | 21:02 |
| lbragstad | clarkb: ah - not a huge deal, just something i noticed when i stumbled across it | 21:02 |
| *** dpawlik has joined #openstack-infra | 21:02 | |
| *** yolanda has quit IRC | 21:02 | |
| johnsom | clarkb https times out and http gives a /images not found | 21:03 |
| *** jtomasek has quit IRC | 21:03 | |
| clarkb | johnsom: oh sorry nb01 and nb02 | 21:03 |
| fungi | lbragstad: worth noting. proper retirement would remove all content except a readme | 21:03 |
| clarkb | NodepoolBuilder not NodepoolLauncher | 21:03 |
| *** boden has joined #openstack-infra | 21:03 | |
| johnsom | Bingo, thanks | 21:04 |
| lbragstad | fungi: yeah - it looks like it was retired properly, but i noticed a bad link/reference in the readme :) http://paste.openstack.org/show/724453/ | 21:04 |
| *** zul has joined #openstack-infra | 21:05 | |
| clarkb | johnsom: the images use glean which will look for config drive data and configure root's ssh key. Or you can just modify the image slgihtly before you boot it | 21:05 |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul master: WIP: Match tag items against containing branches https://review.openstack.org/578557 | 21:06 |
| openstackgerrit | Colleen Murphy proposed openstack-infra/puppet-openstackci master: Update Gemfile for Zuulv3 https://review.openstack.org/577521 | 21:07 |
| openstackgerrit | Colleen Murphy proposed openstack-infra/puppet-openstackci master: Add beaker tests for nodepool https://review.openstack.org/577522 | 21:07 |
| openstackgerrit | Colleen Murphy proposed openstack-infra/puppet-openstackci master: Add beaker-rspec bindep dependencies https://review.openstack.org/578559 | 21:07 |
| johnsom | clarkb roger | 21:07 |
| *** Hal has joined #openstack-infra | 21:07 | |
| fungi | lbragstad: ahh. yeah probably not worth the trouble to fix. we can assume the reader will have some common sense ;) | 21:07 |
| lbragstad | ++ works for me | 21:07 |
| *** heyongli has quit IRC | 21:07 | |
| *** Hal is now known as Guest60457 | 21:07 | |
| dpawlik | johnsom: im here | 21:08 |
| *** heyongli has joined #openstack-infra | 21:08 | |
| *** priteau has quit IRC | 21:10 | |
| *** zul has quit IRC | 21:12 | |
| openstackgerrit | Clark Boylan proposed openstack-infra/project-config master: Add kata runtime and tests repos to zuul https://review.openstack.org/578562 | 21:14 |
| openstackgerrit | Clark Boylan proposed openstack-infra/project-config master: Add ubuntu job to new kata repos in zuul https://review.openstack.org/578563 | 21:14 |
| *** ldnunes has quit IRC | 21:15 | |
| *** heyongli has quit IRC | 21:18 | |
| *** heyongli has joined #openstack-infra | 21:18 | |
| *** boden has quit IRC | 21:20 | |
| mriedem | what does "ready to submit" mean? https://review.openstack.org/#/c/569149/ | 21:25 |
| mriedem | i'm not sure what's holding that up | 21:26 |
| clarkb | mriedem: it means the votes are all in a state that allows submitting | 21:26 |
| *** Guest60457 has quit IRC | 21:26 | |
| clarkb | +2 verified +2 code review +1 workflow | 21:26 |
| clarkb | not immediately sure why zuul didn't submit and merge it | 21:27 |
| clarkb | mriedem: the parent commit is for an older patchset so needs a rebase | 21:28 |
| mriedem | done, thanks | 21:28 |
| clarkb | mriedem: since old patchsets can never merge (and in this case current patchset did merge) you can't merge it with that parent | 21:28 |
| *** heyongli has quit IRC | 21:28 | |
| ianw | can anyone get to https://cn1.linaro.cloud:35357 ? | 21:28 |
| *** heyongli has joined #openstack-infra | 21:28 | |
| clarkb | ianw: initial testing from here is no | 21:28 |
| ianw | yeah i think it's more than flaky network and actually gone away :/ | 21:29 |
| cmurphy | o.0 are you expecting access to the admin interface? | 21:29 |
| ianw | cmurphy: just any sort of connection | 21:29 |
| clarkb | port 5000 also not talking | 21:30 |
| ianw | but no, it should just be the keystone endpoint | 21:30 |
| openstackgerrit | Merged openstack-infra/system-config master: Add git.starlingx.io https://review.openstack.org/572793 | 21:30 |
| clarkb | and 443 | 21:30 |
| fungi | china offline | 21:30 |
| fungi | back later | 21:30 |
| ianw | gema: ^ i will file something. we still have a couple of active hosts though | 21:31 |
| clarkb | bah I think the starlingx change merged 20 seconds after the git update for systemconf was completed | 21:31 |
| fungi | :( | 21:32 |
| fungi | next pulse | 21:32 |
| *** jtomasek has joined #openstack-infra | 21:36 | |
| openstackgerrit | Merged openstack-infra/zuul master: executor: add inventory_file path to the zuul.executor job variable https://review.openstack.org/578235 | 21:37 |
| *** felipemonteiro_ has quit IRC | 21:38 | |
| *** heyongli has quit IRC | 21:38 | |
| *** edmondsw has joined #openstack-infra | 21:38 | |
| *** heyongli has joined #openstack-infra | 21:39 | |
| openstackgerrit | Merged openstack-infra/zuul master: Add pause function to executor https://review.openstack.org/577827 | 21:40 |
| *** myoung is now known as myoung|pto | 21:41 | |
| *** hashar has quit IRC | 21:42 | |
| *** agopi has quit IRC | 21:43 | |
| *** edmondsw has quit IRC | 21:43 | |
| openstackgerrit | James E. Blair proposed openstack-infra/zuul master: WIP: Match tag items against containing branches https://review.openstack.org/578557 | 21:45 |
| *** tonytan4ever_brb has joined #openstack-infra | 21:45 | |
| *** tonytan4ever has quit IRC | 21:45 | |
| *** heyongli has quit IRC | 21:48 | |
| *** heyongli has joined #openstack-infra | 21:49 | |
| *** rcernin has joined #openstack-infra | 21:50 | |
| *** mriedem has quit IRC | 21:51 | |
| smcginnis | My next step patch for getting the operations guide docs going - https://review.openstack.org/#/c/578441/ | 21:52 |
| smcginnis | Not time critical, but I do like to keep the momentum going. ;) | 21:52 |
| clarkb | I think we should try to get AJaeger/mordreds change in that conflicts with that first | 21:53 |
| clarkb | then we can all rebase our changes | 21:54 |
| clarkb | (otherwise ajaeger/mordreds change will never merge) | 21:54 |
| clarkb | oh maybe it i s already in /me checks | 21:54 |
| clarkb | hrm no, why isn't it in the conflicts with list anymore | 21:55 |
| clarkb | config-core https://review.openstack.org/#/c/577307/ is the change I am talking about above. It greatly simplifies our project-config zuul projects configuration btu is going to conflict with many things, smcginnis and I both have changes up that conflict with it I think | 21:56 |
| clarkb | would probably be good if get another set of eyes on that and merge it | 21:56 |
| smcginnis | Oh right, I looked at that one earlier. | 21:57 |
| smcginnis | And like last time it is locking up my browser. :/ | 21:57 |
| clarkb | I had to fetch it locally and git diff in a terminal | 21:58 |
| smcginnis | So looks like it doesn't conflict in code, but would conflict in intent since at least mine adds system-required which will either need to be removed with this patch if it comes second, or be cleaned up later. | 21:59 |
| *** heyongli has quit IRC | 21:59 | |
| smcginnis | I agree, good to get this through first. | 21:59 |
| *** heyongli has joined #openstack-infra | 21:59 | |
| clarkb | ya I've got one up that will need to be rebased too, probably best to just rip the bandaid off :) | 21:59 |
| dtroyer | clarkb: no worries on timing, it took me this long to follow-up. is systemconf on a periodic cycle? | 21:59 |
| *** bobh has quit IRC | 22:00 | |
| smcginnis | Change in mine prepped pending that one merging. | 22:00 |
| clarkb | dtroyer: yup it runs every 15 minutes (so is starting right now), but takes about 25 minutes to complete an update so really its every 30 minute sor so | 22:00 |
| *** eharney has joined #openstack-infra | 22:02 | |
| ianw | clarkb: looks ok to me ... but also don't feel like i know every possible thing that might go wrong with it either. as you say, probably easiest to just try it and see | 22:03 |
| clarkb | ianw: zuul should mostly check the syntax is correct | 22:04 |
| clarkb | I think its mostly safe to do | 22:04 |
| clarkb | dtroyer: https://git.starlingx.io/cgit is a thing now. Going to approve the project-config update to add projects to that alias now | 22:05 |
| ianw | yeah, i'm just a bit shy on "changes that touch basically every project" :) | 22:06 |
| clarkb | mnaser: as an FYI I am going to approve https://review.openstack.org/#/c/572801/1 which has a -1 from you bceause git.starlingx.io is a thing now | 22:07 |
| smcginnis | It does get rid of quite a few lines from the file, which is kind of nice. | 22:08 |
| clarkb | smcginnis: ya and simplifies setup for new projects too | 22:08 |
| smcginnis | The department of redundancy reduction department would approve. | 22:08 |
| openstackgerrit | Merged openstack-infra/zuul master: Add min_avail_hdd governor for zuul-executor https://review.openstack.org/577948 | 22:09 |
| efried | fungi: https://governance.openstack.org/tc/reference/project-testing-interface.html#documentation <== this says the command for doc build is | 22:09 |
| efried | sphinx-build -b html doc/source doc/build | 22:09 |
| efried | but is it in fact | 22:09 |
| efried | sphinx-build -b html doc/source doc/build/html | 22:09 |
| efried | ? | 22:09 |
| *** heyongli has quit IRC | 22:09 | |
| *** heyongli has joined #openstack-infra | 22:09 | |
| smcginnis | It is build, but the chosen default build type is html. | 22:09 |
| efried | smcginnis: afaict this is causing built docs to go to doc/build/foo instead of doc/build/html/foo as I would have expected. | 22:10 |
| smcginnis | efried: So you could potentially have other build types under doc/build | 22:10 |
| smcginnis | efried: Hmm, that may be an issue with something in your conf.py. | 22:10 |
| smcginnis | efried: Which repo? | 22:10 |
| efried | cyborg, but... | 22:11 |
| efried | I'm trying to fix them up so their specs actually build. | 22:11 |
| efried | so there's not actually an example out in the world yet. | 22:11 |
| efried | unless you can tell something from http://logs.openstack.org/55/578455/2/check/build-openstack-sphinx-docs/aac4229/job-output.txt.gz -- I can't, looks like the commands themselves are suppressed. | 22:11 |
| corvus | clarkb: i reckon you can +W 577307; i would but i'm about to afk for a bit | 22:11 |
| smcginnis | efried: Is it cyborg, or cyborg-specs? | 22:11 |
| efried | smcginnis: cyborg (which is another story) | 22:12 |
| clarkb | corvus: thanks, will get it | 22:12 |
| *** Shrews has quit IRC | 22:12 | |
| clarkb | here goes | 22:12 |
| efried | smcginnis: We agree their specs should go in cyborg-specs, but they're keeping them in cyborg at least for the time being. (So yeah, all of this is going to have to be redone for cyborg-specs probably in Stein, plus redirects galore...) | 22:12 |
| smcginnis | efried: Hmm, I don't think it should make a difference, but I see they are using oslosphinx instead of openstackdocstheme. | 22:13 |
| efried | smcginnis: I checked on that in nova-specs and nova, and those are the same | 22:13 |
| efried | oh, no, nova is openstackdocstheme | 22:13 |
| smcginnis | efried: Oh, actually, I think you may be right on the output path needing to be doc/source/html. | 22:14 |
| smcginnis | efried: Just double checked the cinder docs job definition and it is specifying the html path. | 22:14 |
| efried | smcginnis: Okay, I'll make that change in the tox.ini (which is where I found it). I think it's correct in the actual job, just the doc that's wrong. | 22:15 |
| *** rlandy|afk is now known as rlandy | 22:15 | |
| efried | smcginnis: If you can tell me where that doc lives, I can propose the edit | 22:16 |
| fungi | propose an update to the pti and let's see if it flies | 22:16 |
| openstackgerrit | Merged openstack-infra/zuul master: Improve logging of GithubTriggerEvents https://review.openstack.org/575014 | 22:16 |
| fungi | efried: governance reference subdir | 22:16 |
| smcginnis | efried: Looks like there's a different issue there - http://logs.openstack.org/55/578455/2/check/build-openstack-sphinx-docs/aac4229/job-output.txt.gz#_2018-06-27_19_30_28_327503 | 22:16 |
| efried | smcginnis: Yes, I got that all fixed up. | 22:17 |
| efried | And the next one, and the next.... | 22:17 |
| smcginnis | efried: Ah. ;) | 22:17 |
| efried | new patch sets on the way in a sec... | 22:17 |
| *** agopi has joined #openstack-infra | 22:17 | |
| *** dpawlik has quit IRC | 22:17 | |
| *** dpawlik has joined #openstack-infra | 22:18 | |
| *** heyongli has quit IRC | 22:19 | |
| *** heyongli has joined #openstack-infra | 22:20 | |
| *** kienha has joined #openstack-infra | 22:22 | |
| openstackgerrit | Merged openstack-infra/project-config master: Set up cgit aliases for StarlingX https://review.openstack.org/572801 | 22:22 |
| efried | smcginnis, fungi: https://review.openstack.org/578573 Fix doc output path in PTI reference | 22:22 |
| *** Shrews has joined #openstack-infra | 22:22 | |
| *** dpawlik has quit IRC | 22:22 | |
| fungi | efried: thanks! documentation is rarely perfect | 22:23 |
| openstackgerrit | Merged openstack-infra/project-config master: Use wildcard to apply system-required template https://review.openstack.org/577307 | 22:23 |
| smcginnis | I suppose technically it was correct with what it stated before. | 22:23 |
| efried | smcginnis: Yes, it didn't actually *say* what the job itself does... | 22:23 |
| smcginnis | I think this is better for consistency though. | 22:23 |
| efried | smcginnis: Right, I used it as a cue to figure out how to fix up the tox.ini and get builds working locally, so it's useful for those to match. | 22:24 |
| fungi | the prose was right, the invocation it quoted was for sure wrong | 22:25 |
| fungi | and the prose while right, was still arguably misleading | 22:25 |
| openstackgerrit | Sean McGinnis proposed openstack-infra/project-config master: Readd openstack/operations-guide repo https://review.openstack.org/578441 | 22:26 |
| *** hongbin has quit IRC | 22:27 | |
| *** nicolasbock has quit IRC | 22:29 | |
| clarkb | zuul seems to still be working after the system-required update | 22:29 |
| openstackgerrit | Merged openstack-infra/zuul master: Fix missing dequeue on synchronized pull request https://review.openstack.org/577355 | 22:29 |
| *** heyongli has quit IRC | 22:29 | |
| *** heyongli has joined #openstack-infra | 22:30 | |
| openstackgerrit | Clark Boylan proposed openstack-infra/project-config master: Add kata runtime and tests repos to zuul https://review.openstack.org/578562 | 22:31 |
| openstackgerrit | Clark Boylan proposed openstack-infra/project-config master: Add ubuntu job to new kata repos in zuul https://review.openstack.org/578563 | 22:31 |
| openstackgerrit | Clark Boylan proposed openstack-infra/project-config master: Cleanup system-required stragglers https://review.openstack.org/578574 | 22:33 |
| clarkb | dtroyer: https://git.starlingx.io/cgit now with repos | 22:34 |
| *** eharney has quit IRC | 22:35 | |
| *** heyongli has quit IRC | 22:40 | |
| *** heyongli has joined #openstack-infra | 22:40 | |
| *** jamesmcarthur has joined #openstack-infra | 22:40 | |
| *** jamesmcarthur has quit IRC | 22:45 | |
| *** heyongli has quit IRC | 22:50 | |
| *** kzaitsev1pi has joined #openstack-infra | 22:50 | |
| *** heyongli has joined #openstack-infra | 22:50 | |
| *** kzaitsev_pi has quit IRC | 22:52 | |
| fungi | now with happening | 22:52 |
| *** Shrews has quit IRC | 22:55 | |
| *** Shrews has joined #openstack-infra | 22:58 | |
| *** heyongli has quit IRC | 23:00 | |
| *** heyongli has joined #openstack-infra | 23:00 | |
| mnaser | clarkb: no problem regarding the approve -1 | 23:05 |
| mnaser | I’ll remove it once I’m near a computer for completeness | 23:05 |
| mnaser | :) | 23:05 |
| fungi | you won't be able to remove it at this point since it's already merged | 23:06 |
| dtroyer | clarkb, fungi: thanks! it looks good | 23:09 |
| *** heyongli has quit IRC | 23:10 | |
| *** heyongli has joined #openstack-infra | 23:11 | |
| *** yolanda__ has joined #openstack-infra | 23:11 | |
| *** Shrews has quit IRC | 23:13 | |
| *** tonytan4ever_brb has quit IRC | 23:14 | |
| *** tonytan4ever has joined #openstack-infra | 23:14 | |
| *** Shrews has joined #openstack-infra | 23:15 | |
| *** yolanda_ has quit IRC | 23:15 | |
| *** ianychoi has quit IRC | 23:16 | |
| *** yolanda__ has quit IRC | 23:18 | |
| *** ianychoi has joined #openstack-infra | 23:19 | |
| *** boden has joined #openstack-infra | 23:20 | |
| *** yolanda has joined #openstack-infra | 23:20 | |
| *** heyongli has quit IRC | 23:21 | |
| *** heyongli has joined #openstack-infra | 23:21 | |
| *** yolanda_ has joined #openstack-infra | 23:21 | |
| *** yolanda has quit IRC | 23:24 | |
| *** boden has quit IRC | 23:24 | |
| *** yolanda__ has joined #openstack-infra | 23:25 | |
| *** edmondsw has joined #openstack-infra | 23:27 | |
| *** yolanda_ has quit IRC | 23:29 | |
| *** yolanda has joined #openstack-infra | 23:29 | |
| *** yolanda__ has quit IRC | 23:31 | |
| *** heyongli has quit IRC | 23:31 | |
| *** heyongli has joined #openstack-infra | 23:31 | |
| *** edmondsw has quit IRC | 23:32 | |
| *** mihalis68 has quit IRC | 23:34 | |
| *** tosky has quit IRC | 23:35 | |
| *** heyongli has quit IRC | 23:41 | |
| *** heyongli has joined #openstack-infra | 23:42 | |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul master: zk: retry initial zookeeper connection attempts https://review.openstack.org/576048 | 23:45 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul master: zk: use kazoo retry facilities https://review.openstack.org/536209 | 23:45 |
| *** heyongli has quit IRC | 23:51 | |
| *** heyongli has joined #openstack-infra | 23:52 | |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool master: zk: retry initial zookeeper connection attempts https://review.openstack.org/576047 | 23:54 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool master: zk: use kazoo retry facilities https://review.openstack.org/535537 | 23:54 |
| *** kienha has quit IRC | 23:57 | |
| *** alex_xu has quit IRC | 23:59 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!