| *** thorst has joined #openstack-infra | 00:01 | |
| *** thorst has quit IRC | 00:02 | |
| *** salv-orlando has quit IRC | 00:03 | |
| *** thorst has joined #openstack-infra | 00:03 | |
| *** markvoelker has quit IRC | 00:04 | |
| *** ArxCruz has quit IRC | 00:04 | |
| *** chlong has joined #openstack-infra | 00:05 | |
| *** gildub_ has joined #openstack-infra | 00:06 | |
| *** thorst has quit IRC | 00:11 | |
| *** tzn has quit IRC | 00:16 | |
| *** hogepodge has quit IRC | 00:16 | |
| *** piet has joined #openstack-infra | 00:17 | |
| *** hogepodge has joined #openstack-infra | 00:22 | |
| *** hogepodge has quit IRC | 00:27 | |
| *** piet has quit IRC | 00:35 | |
| *** salv-orl_ has quit IRC | 00:36 | |
| *** piet has joined #openstack-infra | 00:38 | |
| *** piet has quit IRC | 00:40 | |
| *** piet has joined #openstack-infra | 00:41 | |
| *** Qiming has joined #openstack-infra | 00:42 | |
| *** hogepodge has joined #openstack-infra | 00:43 | |
| *** dims has joined #openstack-infra | 00:43 | |
| *** otter768 has joined #openstack-infra | 00:43 | |
| *** otter768 has quit IRC | 00:48 | |
| *** piet has quit IRC | 00:57 | |
| *** triggerwarning has joined #openstack-infra | 01:00 | |
| *** triggerwarning has quit IRC | 01:01 | |
| *** triggerwarning has joined #openstack-infra | 01:01 | |
| *** dims has quit IRC | 01:02 | |
| *** sdake_ has joined #openstack-infra | 01:03 | |
| tonyb | is there anything strange going on in project-config and running local tests? | 01:03 |
|---|---|---|
| *** sdake has quit IRC | 01:04 | |
| tonyb | I can't get tox to pass on origin/master | 01:04 |
| *** markvoelker has joined #openstack-infra | 01:05 | |
| *** dcentral has quit IRC | 01:08 | |
| *** triggerwarning has quit IRC | 01:08 | |
| *** thorst has joined #openstack-infra | 01:09 | |
| *** markvoelker has quit IRC | 01:10 | |
| openstackgerrit | Tony Breeds proposed openstack-infra/project-config: Do not run heatclient functional tests on kilo https://review.openstack.org/259820 | 01:13 |
| *** armax has quit IRC | 01:15 | |
| *** thorst has quit IRC | 01:16 | |
| *** lykinsbd has joined #openstack-infra | 01:17 | |
| *** Qiming has quit IRC | 01:19 | |
| *** piet has joined #openstack-infra | 01:22 | |
| openstackgerrit | Diana Whitten proposed openstack-infra/project-config: Add openstack-ux repo https://review.openstack.org/259239 | 01:23 |
| *** piet has quit IRC | 01:27 | |
| *** lykinsbd has quit IRC | 01:28 | |
| *** lykinsbd has joined #openstack-infra | 01:29 | |
| *** salv-orlando has joined #openstack-infra | 01:37 | |
| *** jyuso1 has joined #openstack-infra | 01:41 | |
| *** piet has joined #openstack-infra | 01:41 | |
| fungi | clarkb: on the lack of swap, we can mkswap a sparse file on the rootfs and activate it if jobs want swap (we may in fact want to ensure we have roughly the same amount of swap in all providers) | 01:47 |
| * fungi goes back to not really being at a computer | 01:47 | |
| *** Qiming has joined #openstack-infra | 01:52 | |
| *** dims has joined #openstack-infra | 02:07 | |
| *** vilobhmm11 has joined #openstack-infra | 02:10 | |
| *** vilobhmm11 has quit IRC | 02:11 | |
| *** salv-orlando has quit IRC | 02:12 | |
| *** thorst has joined #openstack-infra | 02:14 | |
| *** thorst has quit IRC | 02:21 | |
| *** markvoelker has joined #openstack-infra | 02:36 | |
| *** woodster_ has quit IRC | 02:36 | |
| *** harlowja_at_home has joined #openstack-infra | 02:39 | |
| *** markvoelker has quit IRC | 02:40 | |
| *** BenJ has joined #openstack-infra | 02:44 | |
| *** BenJ is now known as Guest74291 | 02:44 | |
| *** otter768 has joined #openstack-infra | 02:44 | |
| *** Guest74291 has quit IRC | 02:45 | |
| *** rvasilets___ has quit IRC | 02:45 | |
| *** BenCW has joined #openstack-infra | 02:46 | |
| *** sdake_ has quit IRC | 02:47 | |
| *** otter768 has quit IRC | 02:49 | |
| openstackgerrit | Joshua Hesketh proposed openstack-infra/zuul: Fix regression in change tracking https://review.openstack.org/259281 | 02:51 |
| *** harlowja_at_home has quit IRC | 02:54 | |
| *** piet has quit IRC | 02:57 | |
| *** janonymous has joined #openstack-infra | 02:58 | |
| *** sdake has joined #openstack-infra | 03:04 | |
| *** sdake_ has joined #openstack-infra | 03:06 | |
| *** sdake has quit IRC | 03:09 | |
| *** zz_dimtruck is now known as dimtruck | 03:13 | |
| *** thorst has joined #openstack-infra | 03:19 | |
| dims | lifeless : have you seen this before? raise ValueError, "No closing quotation" | 03:20 |
| dims | lifeless : example in http://logs.openstack.org/07/259707/2/check/gate-releases-tox-validate/c18a4b5/console.html#_2015-12-21_03_09_48_019 | 03:20 |
| lifeless | dims: nope, but it sounds like it might be the thing ansible ran into - see the mail on -dev about renaming their functional test target | 03:22 |
| *** coolsvap|away is now known as coolsvap | 03:23 | |
| dims | lifeless : ack will check it out tomorrow. thanks | 03:23 |
| lifeless | dims: (or it may not be ;)) | 03:25 |
| *** links has joined #openstack-infra | 03:26 | |
| openstackgerrit | Tony Breeds proposed openstack-infra/project-config: Add oldstable jobs for python-barbicanclient https://review.openstack.org/259828 | 03:27 |
| *** thorst has quit IRC | 03:27 | |
| *** JbHuNt has joined #openstack-infra | 03:31 | |
| *** dims has quit IRC | 03:32 | |
| openstackgerrit | Tony Breeds proposed openstack-infra/project-config: Barbican does not support python34 in kilo https://review.openstack.org/259830 | 03:32 |
| *** coolsvap is now known as coolsvap|away | 03:33 | |
| *** coolsvap|away is now known as coolsvap | 03:40 | |
| openstackgerrit | Ian Wienand proposed openstack-infra/glean: Handle invalid symlinks in written files https://review.openstack.org/259831 | 03:40 |
| *** lykinsbd has quit IRC | 03:44 | |
| *** zhurong has joined #openstack-infra | 03:51 | |
| openstackgerrit | Ian Wienand proposed openstack-infra/glean: Minor testing cleanups https://review.openstack.org/259834 | 03:51 |
| *** otter768 has joined #openstack-infra | 04:07 | |
| openstackgerrit | Ian Wienand proposed openstack-infra/glean: Check for network service enablement https://review.openstack.org/259838 | 04:12 |
| *** zhurong has quit IRC | 04:20 | |
| *** claudiub has joined #openstack-infra | 04:22 | |
| *** thorst has joined #openstack-infra | 04:24 | |
| *** mahatic has joined #openstack-infra | 04:26 | |
| *** jyuso1 has quit IRC | 04:28 | |
| openstackgerrit | Ian Wienand proposed openstack/diskimage-builder: yum-minimal: leave behind dummy /etc/resolv.conf https://review.openstack.org/257173 | 04:29 |
| *** thorst has quit IRC | 04:31 | |
| *** jyuso1 has joined #openstack-infra | 04:34 | |
| *** markvoelker has joined #openstack-infra | 04:37 | |
| *** otter768 has quit IRC | 04:37 | |
| *** dimtruck is now known as zz_dimtruck | 04:40 | |
| *** markvoelker has quit IRC | 04:41 | |
| *** tlian2 has quit IRC | 04:44 | |
| *** zz_dimtruck is now known as dimtruck | 04:46 | |
| *** otter768 has joined #openstack-infra | 04:48 | |
| openstackgerrit | Joshua Hesketh proposed openstack-infra/zuul: Skip based off commit message https://review.openstack.org/251371 | 04:53 |
| openstackgerrit | Joshua Hesketh proposed openstack-infra/zuul: Filter events by commit message https://review.openstack.org/251373 | 04:53 |
| openstackgerrit | Joshua Hesketh proposed openstack-infra/zuul: Filter jobs off commit message https://review.openstack.org/251372 | 04:53 |
| openstackgerrit | Joshua Hesketh proposed openstack-infra/zuul: Fix regression in change tracking https://review.openstack.org/259281 | 04:53 |
| openstackgerrit | Joshua Hesketh proposed openstack-infra/zuul: Expose commitMessage as a Change attribute https://review.openstack.org/222791 | 04:53 |
| *** harlowja_at_home has joined #openstack-infra | 04:57 | |
| *** bharathm has joined #openstack-infra | 05:00 | |
| openstackgerrit | Ian Wienand proposed openstack/diskimage-builder: yum-minimal: do not configure eth0 & eth1 for DHCP automatically https://review.openstack.org/257172 | 05:04 |
| *** jyuso1 has quit IRC | 05:08 | |
| *** bharathm has quit IRC | 05:08 | |
| *** markvoelker has joined #openstack-infra | 05:10 | |
| *** zhurong has joined #openstack-infra | 05:12 | |
| *** jyuso1 has joined #openstack-infra | 05:15 | |
| *** amotoki has joined #openstack-infra | 05:21 | |
| *** harlowja_at_home has quit IRC | 05:23 | |
| *** mudassirlatif has joined #openstack-infra | 05:27 | |
| *** thorst has joined #openstack-infra | 05:29 | |
| *** otter768 has quit IRC | 05:30 | |
| *** maishsk has quit IRC | 05:32 | |
| *** thorst has quit IRC | 05:36 | |
| openstackgerrit | Joshua Hesketh proposed openstack-infra/project-config: Turn off -e in check-git-commit-message.sh https://review.openstack.org/259842 | 05:39 |
| openstackgerrit | Ken'ichi Ohmichi proposed openstack-infra/devstack-gate: Enable SameHostFilter and DifferentHostFilter https://review.openstack.org/259843 | 05:53 |
| *** julim has quit IRC | 05:56 | |
| *** yuanying has joined #openstack-infra | 05:56 | |
| *** nihilifer has joined #openstack-infra | 06:00 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack-infra/project-config: Normalize projects.yaml https://review.openstack.org/259847 | 06:02 |
| *** Thelo has quit IRC | 06:05 | |
| *** oomichi has joined #openstack-infra | 06:06 | |
| openstackgerrit | Gal Sagie proposed openstack-infra/project-config: Export enabled services from devstackgaterc in Dragonflow https://review.openstack.org/259848 | 06:15 |
| *** rossella_s has quit IRC | 06:21 | |
| *** rossella_s has joined #openstack-infra | 06:22 | |
| *** oomichi has quit IRC | 06:30 | |
| *** dimtruck is now known as zz_dimtruck | 06:32 | |
| *** abregman has joined #openstack-infra | 06:33 | |
| *** aopincar has joined #openstack-infra | 06:34 | |
| *** thorst has joined #openstack-infra | 06:34 | |
| *** thorst has quit IRC | 06:41 | |
| *** vilobhmm11 has joined #openstack-infra | 06:46 | |
| *** _nadya_ has joined #openstack-infra | 06:49 | |
| *** triggerwarning has joined #openstack-infra | 06:51 | |
| *** Ramanjaneya has joined #openstack-infra | 06:51 | |
| *** dcentral has joined #openstack-infra | 06:51 | |
| *** oomichi has joined #openstack-infra | 06:53 | |
| *** Thelo has joined #openstack-infra | 06:56 | |
| *** amotoki has quit IRC | 06:59 | |
| *** _nadya_ has quit IRC | 07:00 | |
| *** HeOS has joined #openstack-infra | 07:00 | |
| *** gildub has quit IRC | 07:03 | |
| *** gildub_ has quit IRC | 07:03 | |
| *** maishsk has joined #openstack-infra | 07:09 | |
| *** chlong has quit IRC | 07:13 | |
| Thelo | Hi all, in the third party documentation it is written that we need a "normal reviewer account in gerrit" to read the event stream. Is this different from a normal account? | 07:13 |
| *** sdake_ has quit IRC | 07:17 | |
| *** markvoelker has quit IRC | 07:21 | |
| *** triggerwarning has quit IRC | 07:24 | |
| openstackgerrit | Ian Wienand proposed openstack/diskimage-builder: [WIP] yum-minimal : install selinux policy packages https://review.openstack.org/259867 | 07:25 |
| *** markvoelker has joined #openstack-infra | 07:25 | |
| *** _nadya_ has joined #openstack-infra | 07:28 | |
| *** _nadya_ has quit IRC | 07:28 | |
| openstackgerrit | Anton Tcitlionok proposed openstack-infra/jenkins-job-builder: Clarify 'trigger-parametrized-builds' description https://review.openstack.org/259870 | 07:29 |
| openstackgerrit | Anton Tcitlionok proposed openstack-infra/jenkins-job-builder: Clarify 'trigger-parametrized-builds' description https://review.openstack.org/259870 | 07:31 |
| *** AJaeger has joined #openstack-infra | 07:31 | |
| *** irenab has quit IRC | 07:31 | |
| *** otter768 has joined #openstack-infra | 07:31 | |
| *** claudiub has quit IRC | 07:32 | |
| *** lezbar has joined #openstack-infra | 07:34 | |
| *** otter768 has quit IRC | 07:36 | |
| *** unicell has quit IRC | 07:38 | |
| *** unicell has joined #openstack-infra | 07:38 | |
| *** lezbar has quit IRC | 07:39 | |
| *** thorst has joined #openstack-infra | 07:39 | |
| *** mudassirlatif has quit IRC | 07:42 | |
| openstackgerrit | Andreas Jaeger proposed openstack-infra/project-config: Do not run heatclient functional tests on kilo https://review.openstack.org/259820 | 07:42 |
| *** lezbar__ has joined #openstack-infra | 07:43 | |
| openstackgerrit | Kirill Zaitsev proposed openstack-infra/project-config: Adds release notes jobs to murano deliverables https://review.openstack.org/259874 | 07:44 |
| *** flepied has quit IRC | 07:44 | |
| openstackgerrit | Merged openstack-infra/project-config: Normalize projects.yaml https://review.openstack.org/259847 | 07:45 |
| *** _nadya_ has joined #openstack-infra | 07:45 | |
| *** thorst has quit IRC | 07:46 | |
| *** BenCW has quit IRC | 07:52 | |
| openstackgerrit | Kirill Zaitsev proposed openstack-infra/project-config: Make python3 jobs voting for murano-dashboard https://review.openstack.org/259877 | 07:54 |
| *** migi has joined #openstack-infra | 07:57 | |
| openstackgerrit | Kirill Zaitsev proposed openstack-infra/project-config: Adds release notes jobs to murano deliverables https://review.openstack.org/259874 | 08:01 |
| *** vilobhmm11 has quit IRC | 08:03 | |
| openstackgerrit | Andreas Jaeger proposed openstack-infra/project-config: Revert "Check for correct DocImpact tag usage" https://review.openstack.org/259569 | 08:03 |
| *** salv-orlando has joined #openstack-infra | 08:11 | |
| openstackgerrit | yolanda.robla proposed openstack/diskimage-builder: Create new partitioning element. https://review.openstack.org/259881 | 08:11 |
| *** flepied has joined #openstack-infra | 08:12 | |
| *** irenab has joined #openstack-infra | 08:12 | |
| *** flepied1 has joined #openstack-infra | 08:13 | |
| *** salv-orlando has quit IRC | 08:14 | |
| *** ihrachys has joined #openstack-infra | 08:14 | |
| *** flepied has quit IRC | 08:17 | |
| *** Benjj has joined #openstack-infra | 08:19 | |
| Benjj | VolumeSizeExceedsAvailableQuota: Requested volume or snapshot exceeds allowed gigabytes quota. Requested 1G, quota is 1G and 1G has been consumed. | 08:20 |
| Benjj | I get this error msg. where should I change the settings? | 08:20 |
| Benjj | tried in cinder.conf but not works | 08:21 |
| openstackgerrit | Kirill Zaitsev proposed openstack-infra/project-config: Adds release notes jobs to murano deliverables https://review.openstack.org/259874 | 08:22 |
| *** unicell1 has joined #openstack-infra | 08:23 | |
| *** unicell has quit IRC | 08:24 | |
| openstackgerrit | Ian Wienand proposed openstack/diskimage-builder: yum-minimal : install selinux policy packages https://review.openstack.org/259867 | 08:24 |
| Benjj | I install it using devstack | 08:25 |
| openstackgerrit | Nicko Glayre proposed openstack-infra/jenkins-job-builder: Adding support for maven-metadata-plugin https://review.openstack.org/257430 | 08:25 |
| *** abregman has quit IRC | 08:28 | |
| *** chlong has joined #openstack-infra | 08:31 | |
| *** abregman has joined #openstack-infra | 08:33 | |
| *** jcoufal has joined #openstack-infra | 08:34 | |
| *** shardy has joined #openstack-infra | 08:37 | |
| *** dizquierdo has joined #openstack-infra | 08:40 | |
| *** markus_z has joined #openstack-infra | 08:40 | |
| *** thorst has joined #openstack-infra | 08:44 | |
| *** zhurong has quit IRC | 08:45 | |
| *** ifarkas has joined #openstack-infra | 08:45 | |
| *** Hal1 has joined #openstack-infra | 08:45 | |
| *** zhurong has joined #openstack-infra | 08:48 | |
| *** abregman has quit IRC | 08:49 | |
| *** thorst has quit IRC | 08:51 | |
| *** kzaitsev1mb has joined #openstack-infra | 08:53 | |
| *** asselin__ has joined #openstack-infra | 08:55 | |
| openstackgerrit | yolanda.robla proposed openstack/diskimage-builder: Create new partitioning element. https://review.openstack.org/259881 | 08:55 |
| *** asselin_ has quit IRC | 08:57 | |
| AJaeger | Benjj: for devstack better ask on #openstack-qa - or on #openstack-dev | 08:57 |
| *** kzaitsev1mb has quit IRC | 09:00 | |
| *** daemontool has joined #openstack-infra | 09:00 | |
| *** abregman has joined #openstack-infra | 09:04 | |
| *** yamamoto has joined #openstack-infra | 09:05 | |
| *** hashar has joined #openstack-infra | 09:05 | |
| *** [HeOS] has joined #openstack-infra | 09:18 | |
| *** HeOS has quit IRC | 09:19 | |
| *** sfinucan has joined #openstack-infra | 09:20 | |
| *** HeOS has joined #openstack-infra | 09:20 | |
| *** Benjj has quit IRC | 09:21 | |
| *** [HeOS] has quit IRC | 09:23 | |
| *** markmcd has quit IRC | 09:25 | |
| *** markvoelker has quit IRC | 09:25 | |
| *** dizquierdo has quit IRC | 09:26 | |
| *** markmcd has joined #openstack-infra | 09:27 | |
| *** HeOS has quit IRC | 09:29 | |
| openstackgerrit | Merged openstack-infra/project-config: Adds release notes jobs to murano deliverables https://review.openstack.org/259874 | 09:29 |
| openstackgerrit | Merged openstack-infra/project-config: Do not run heatclient functional tests on kilo https://review.openstack.org/259820 | 09:30 |
| *** mpavone has joined #openstack-infra | 09:31 | |
| *** otter768 has joined #openstack-infra | 09:32 | |
| openstackgerrit | Merged openstack-infra/project-config: Check for dead links in App Catalog https://review.openstack.org/259232 | 09:32 |
| *** HeOS has joined #openstack-infra | 09:32 | |
| *** bexelbie has joined #openstack-infra | 09:32 | |
| openstackgerrit | Merged openstack-infra/project-config: Add oldstable jobs for python-barbicanclient https://review.openstack.org/259828 | 09:33 |
| *** salv-orlando has joined #openstack-infra | 09:34 | |
| openstackgerrit | Merged openstack-infra/project-config: Fix rally jobs for networking-ovn, kuryr and dragonflow https://review.openstack.org/258347 | 09:35 |
| *** yamamoto has quit IRC | 09:36 | |
| *** otter768 has quit IRC | 09:36 | |
| *** HeOS has quit IRC | 09:38 | |
| *** Qiming has quit IRC | 09:38 | |
| *** openstack has joined #openstack-infra | 15:36 | |
| yolanda | here it comes | 15:36 |
| *** irenab has quit IRC | 15:36 | |
| AJaeger | thanks, yolanda ! | 15:36 |
| *** sdake has joined #openstack-infra | 15:36 | |
| fungi | short story is that those bots are extensions to supybot which is built on the twisted framework. twisted has an ssl+ipv6 incompatibility where it can't establish ssl sockets to ipv6 addresses but tries to do so anyway if the dns response for the destination includes at least one ipv6 address | 15:37 |
| *** xarses has joined #openstack-infra | 15:37 | |
| AJaeger | yolanda: Yeah, #success works again. | 15:37 |
| yolanda | nice | 15:37 |
| fungi | it's been that way for years and nobody working on twisted has had the time to solve it (apparently a fairly subtle/complex challenge) | 15:37 |
| yolanda | i'll know for next time | 15:37 |
| openstackgerrit | yolanda.robla proposed openstack/diskimage-builder: Add dib element to generate logical volumes https://review.openstack.org/252041 | 15:38 |
| *** bexelbie has joined #openstack-infra | 15:38 | |
| AJaeger | Time to do some shopping. Bye for now... | 15:38 |
| *** AJaeger has quit IRC | 15:38 | |
| *** otter768 has quit IRC | 15:39 | |
| *** jsavak has quit IRC | 15:39 | |
| *** jsavak has joined #openstack-infra | 15:40 | |
| *** maishsk has quit IRC | 15:40 | |
| asselin__ | fungi, if you have a chance, could you review this: https://review.openstack.org/#/c/253619/ | 15:42 |
| *** apuimedo_ has quit IRC | 15:44 | |
| *** apuimedo_ has joined #openstack-infra | 15:44 | |
| *** smarcet1981 has joined #openstack-infra | 15:45 | |
| *** jsavak has quit IRC | 15:45 | |
| *** aopincar has quit IRC | 15:45 | |
| *** sdake has quit IRC | 15:45 | |
| *** smarcet has quit IRC | 15:45 | |
| fungi | asselin__: it looks like we haven't had much diversity of reviewers on the related patches. i'd like to see nibalizer weigh in on that if he's around | 15:46 |
| fungi | i am not confident in my ability to spot the risks on that change | 15:46 |
| *** GheRiver1 has quit IRC | 15:47 | |
| *** sdake has joined #openstack-infra | 15:48 | |
| asselin__ | fungi, understood. | 15:49 |
| fungi | and the system-config change looks like sort of a fulcrum for the others | 15:50 |
| *** yamamoto has joined #openstack-infra | 15:50 | |
| *** smarcet has joined #openstack-infra | 15:51 | |
| *** smarcet1981 has quit IRC | 15:51 | |
| *** apuimedo_ has quit IRC | 15:51 | |
| *** apuimedo has quit IRC | 15:51 | |
| *** harlowja_at_home has joined #openstack-infra | 15:52 | |
| *** sdake_ has joined #openstack-infra | 15:52 | |
| *** sdake has quit IRC | 15:53 | |
| *** apuimedo has joined #openstack-infra | 15:53 | |
| *** yamamoto has quit IRC | 15:54 | |
| *** apuimedo_ has joined #openstack-infra | 15:54 | |
| *** mtanino has joined #openstack-infra | 15:54 | |
| *** yamamoto has joined #openstack-infra | 15:54 | |
| *** changbl has quit IRC | 15:54 | |
| asselin__ | fungi, we need to pick one or the other, having both fails miserable. | 15:54 |
| *** bexelbie has quit IRC | 15:54 | |
| *** tonytan4ever has joined #openstack-infra | 15:55 | |
| *** smarcet has quit IRC | 15:55 | |
| *** smarcet1981 has joined #openstack-infra | 15:55 | |
| fungi | asselin__: makes sense, just want to make sure we're not about to head down a road to somewhere we don't want to be with that module | 15:55 |
| *** vgridnev has quit IRC | 15:56 | |
| openstackgerrit | yolanda.robla proposed openstack-infra/system-config: Add instructions on how to troubleshoot bots https://review.openstack.org/260068 | 15:59 |
| *** sdake has joined #openstack-infra | 16:00 | |
| *** apuimedo has quit IRC | 16:00 | |
| *** vgridnev has joined #openstack-infra | 16:00 | |
| *** apuimedo_ has quit IRC | 16:00 | |
| *** dizquierdo has quit IRC | 16:01 | |
| jeblair | fungi: review-dev.openstack.org COMMAND=/usr/bin/service exit -- is that a known incident? | 16:01 |
| *** yamamoto has quit IRC | 16:02 | |
| *** apuimedo has joined #openstack-infra | 16:02 | |
| *** apuimedo_ has joined #openstack-infra | 16:02 | |
| *** mrmartin has joined #openstack-infra | 16:02 | |
| *** yamamoto has joined #openstack-infra | 16:02 | |
| *** sdake_ has quit IRC | 16:02 | |
| *** yamamoto has quit IRC | 16:03 | |
| fungi | jeblair: it is not. i assumed that was something zaro was testing but meant to look into it, then got sidetracked | 16:03 |
| jeblair | fungi: review.openstack.org COMMAND=/usr/bin/touch GerritSiteHeader.html -- also that | 16:03 |
| fungi | that one is known. i saw jhesketh/yolanda discussing it in scrollback i think after approving a patch to the javascript. i assume it was an errant sudo | 16:03 |
| fungi | the timimg matched up anyway so i didn't look further | 16:03 |
| *** vgridnev has quit IRC | 16:05 | |
| jeblair | zaro was logged in during that time | 16:05 |
| jeblair | (and was the only one) | 16:05 |
| *** esker has quit IRC | 16:06 | |
| yolanda | fungi, i don't follow that.. i didn't log into review-dev. You mean, touching html and css files on gerrit? i did that on review.opensatck | 16:06 |
| jeblair | fungi: and then sudo'd 'service gerrit stop' from zaro shortly thereafter, so, yeah looks harmless | 16:06 |
| yolanda | also, is something i wanted to raise the attention of some infra-root. Why these files are owned by root, and not by gerrit2? any reason? | 16:06 |
| jeblair | yolanda: because gerrit2 does not need to alter them | 16:07 |
| krotscheck | Do we have historical data on gerrit usage? i.e. how user behavior changed pre-upgrade to post-upgrade? | 16:07 |
| jeblair | yolanda: it's generally good for web applications not to be permitted to modify more than necessary | 16:07 |
| * krotscheck is curious about whether it's possible to extract the habituation time for a new user interface from that data. | 16:07 | |
| yolanda | ok | 16:07 |
| jeblair | yolanda: particularly so that if they are compromised in some small way, they can't have their own code changed by the attacker so that they could then be compromised in some laregr way. | 16:08 |
| fungi | yolanda: yep, i was talking about the second one | 16:08 |
| openstackgerrit | Valeriy Ponomaryov proposed openstack-infra/project-config: Add 'dsvm' to manila-image-elements dsvm-job name https://review.openstack.org/260051 | 16:08 |
| yolanda | makes sense | 16:08 |
| jeblair | krotscheck: i believe our only such metric is, erm, unsolicited user feedback on the mailing list and irc. | 16:08 |
| jeblair | krotscheck: it doesn't record dwell time metrics or anything like that | 16:09 |
| jeblair | krotscheck: we have 1 month of apache logs, from which one could, potentially, reconstruct some information based on the backend js api requests. however, i think that would be a big project. | 16:09 |
| *** [1]Thelo has joined #openstack-infra | 16:09 | |
| krotscheck | jeblair: Yeah, it would. | 16:10 |
| anteaya | krotscheck: also it has public knowledge that new screen is coming so some folks have been migrating to new screen using 2.8 voluntarily | 16:10 |
| jeblair | krotscheck: one could potentially get some really macro-scale stats from gerrit and/or stackalytics; like overall reviewer efficiency. | 16:11 |
| krotscheck | jeblair: I may be able to get things from the stackalytics data. Determine if pre-upgrade shows a notable change to net reviews performed. | 16:11 |
| krotscheck | right. that. | 16:11 |
| krotscheck | There's a statistical method by which one can determine variable causality too if the dataset is large enough. | 16:11 |
| krotscheck | But it's harder with aggregate things. | 16:11 |
| * krotscheck ponders... | 16:11 | |
| *** Thelo has quit IRC | 16:11 | |
| *** [1]Thelo is now known as Thelo | 16:11 | |
| * krotscheck looks at his todo list.... | 16:12 | |
| * krotscheck shelves this idea until later. | 16:12 | |
| jeblair | krotscheck: may want to consider whether gertty has an impact on that data; it's not a huge user set, but last i looked it was perhaps on the order of 100 | 16:12 |
| jeblair | and of course, this time of year has effects on reviewer throughput as well | 16:13 |
| asselin__ | jeblair, this patch needs some reviews please: https://review.openstack.org/#/c/252768/ | 16:13 |
| *** HeOS has quit IRC | 16:14 | |
| fungi | i was about to say, yeah i expect any impact on review volume will be impacted more by it being what is traditionally our slowest time of year, project-wise, and so any shift from interface learning curve is likely to end up as noise by comparison | 16:15 |
| *** admin0 has quit IRC | 16:15 | |
| *** lykinsbd has joined #openstack-infra | 16:15 | |
| krotscheck | jeblair: It really does. | 16:15 |
| openstackgerrit | Merged openstack-infra/shade: Add support for querying role assignments. https://review.openstack.org/259147 | 16:15 |
| krotscheck | My old prof showed me how to compensate for regular seasonal variation. That was over a decade ago though :/ | 16:16 |
| mordred | I will restate the unuseful "I find the new screen excruciatingly difficult to navigate" | 16:16 |
| jeblair | mordred: on that note, did you see i added project topics to gertty over the weekend? | 16:17 |
| mordred | jeblair: I did not - neat! | 16:17 |
| *** markus_z has quit IRC | 16:17 | |
| blogan | can anyone tell me how I can retrigger the gate job on a review that failed? | 16:17 |
| fungi | yeah, some people preemptively started getting themselves used to the new review screen in 2.8 in preparation for the upgrade, i preemptively shifted more of my workflow to gertty instead | 16:18 |
| *** bastelfreak has quit IRC | 16:18 | |
| jeblair | oh, i have some other fixes; i think i need to brown-paper-bag 1.3.1 | 16:18 |
| anteaya | blogan: well posting a comment with content 'recheck' on the patch sends it back to the check queue | 16:18 |
| Shrews | jeblair: making gertty easier while making gerrit more difficult, eh?... my theory that this is all a well calculated plot seems to have some validity! | 16:18 |
| Shrews | :) | 16:19 |
| blogan | anteaya: yeah but that won't run the gate merge jobs though no? | 16:19 |
| jesusaurus | mordred: hear hear, it is definitely more difficult | 16:19 |
| anteaya | blogan: if it exits the check queue with +1 from Jenkins and has a +1 workflow on it (with a +2 from at least one core reviewer) it will head into the gate queue | 16:19 |
| blogan | anteaya: could have sworn reverify worked in the past but doesn't now | 16:19 |
| anteaya | blogan: it has to pass check to pass gate | 16:19 |
| blogan | anteaya: ahh okay, thanks | 16:19 |
| anteaya | blogan: correct, reverify was removed | 16:19 |
| anteaya | blogan: if you are a core reviewer you can reapply +1 workflow | 16:20 |
| jeblair | Shrews: you're on to something. the two are definitely related; but the causality runs in the non-sinister direction. | 16:20 |
| jeblair | (which i guess means it runs to the right) | 16:20 |
| blogan | anteaya: i did that, but still didn't kick them off, ill do a recheck though | 16:20 |
| anteaya | blogan: which patch? | 16:20 |
| blogan | https://review.openstack.org/#/c/253884/ | 16:20 |
| Shrews | jeblair: so, quick novice gertty question... is there a visual cue to see when a review is dependent on another review that needs to be rebased (i.e., outdated)? | 16:20 |
| jesusaurus | jeblair: the dexter direction? | 16:20 |
| jeblair | Shrews: yes, it will say [OUTDATED] in the 'depends-on' area | 16:21 |
| jeblair | (just like oldscreen gerrit) | 16:21 |
| jeblair | jesusaurus: that's the one! | 16:21 |
| openstackgerrit | gordon chung proposed openstack-infra/project-config: fix stable/liberty telemetry integration job https://review.openstack.org/258642 | 16:21 |
| blogan | anteaya: so on that review, a recheck was posted on the 19th, but the last time the gate jobs ran was the 17th | 16:22 |
| anteaya | blogan: I see that | 16:22 |
| openstackgerrit | James E. Blair proposed openstack/gertty: Fix config validation to accept new keymap format https://review.openstack.org/259253 | 16:23 |
| openstackgerrit | James E. Blair proposed openstack/gertty: Fix multi-key handling at top level https://review.openstack.org/260084 | 16:23 |
| openstackgerrit | James E. Blair proposed openstack/gertty: Cache counts of project changes https://review.openstack.org/260085 | 16:23 |
| anteaya | blogan: yes kevinbenton posted a recheck on the 19th that returned successfuly and should have put that patch in the gate queue, but that didn't seem to happen | 16:23 |
| anteaya | blogan: I'm perplexed | 16:23 |
| blogan | anteaya: me too :) | 16:24 |
| mordred | jeblair: can I get gertty to prioritize download changes for the project I'm looking at? | 16:24 |
| jeblair | mordred: if you hit ctrl-r it will do that | 16:24 |
| jeblair | mordred: i have 'make that happen automatically when it makes sense' on my todo list | 16:24 |
| *** harlowja_at_home has quit IRC | 16:24 | |
| jeblair | oh | 16:24 |
| jeblair | sorry | 16:24 |
| blogan | anteaya: last resort will be to just push up another patchset and lobby for those +2s again | 16:25 |
| anteaya | blogan: stand by | 16:25 |
| jeblair | mordred: i misread that as 'the change i'm looking at', not the 'project i'm looking at' | 16:25 |
| blogan | anteaya: unless yall want to troubleshoot it | 16:25 |
| blogan | anteaya: sure thing | 16:25 |
| jeblair | mordred: 1 sec | 16:25 |
| anteaya | can any infra-root not currently addressing gerrit/gertty interfaces take a peek at the history of https://review.openstack.org/#/c/253884/ | 16:25 |
| anteaya | neither blogan nor myself can figure out why it refuses to enter the gate | 16:26 |
| fungi | anteaya: blogan: recheck on a change which already has a jenkins +1 doesn't enqueue into the gate. gerrit doesn't emit a vote change event when replacing a +1 with a +1 | 16:26 |
| fungi | needs a new workflow +1 added instead | 16:26 |
| jeblair | mordred: okay, yes, ctrl-r in a list of changes for a project will already do a high-priority sync for that project. | 16:26 |
| clarkb | in this case it is the depends on | 16:26 |
| mordred | cool | 16:26 |
| blogan | fungi: i did that last night htough | 16:26 |
| anteaya | fungi: blogan did so and I still don't see it in the status page | 16:26 |
| blogan | fungi: and it didn't enter the gate | 16:26 |
| clarkb | fungi anteaya blogan ^ | 16:26 |
| anteaya | clarkb: thanks | 16:26 |
| mordred | https://review.openstack.org/#/q/I940b541f4ef9c489126cd2d215b1d857f0624de0,n,z | 16:26 |
| fungi | anteaya: blogan: which can also be accomplished by the person who posted the previous workflow +1 removing and readding it | 16:26 |
| jeblair | mordred: (but if you're looking at a query that just happens to be for one project, it's not smart enough to figure that one out so it will sync all subscribed projects in that case) | 16:26 |
| blogan | oh that depends-on bc of the backports! | 16:27 |
| fungi | anteaya: blogan: yes, what clarkb said | 16:27 |
| blogan | good catch clarkb i remember that happening | 16:27 |
| jeblair | mordred: (but the 'hit enter on a project in the project list case' -- that's handled) | 16:27 |
| fungi | that will also keep it from getting enqueued | 16:27 |
| mordred | jeblair: nah, single project was fine | 16:27 |
| anteaya | blogan: that depends on points to 3 different patches: https://review.openstack.org/#/q/I940b541f4ef9c489126cd2d215b1d857f0624de0,n,z | 16:27 |
| *** mpavone has quit IRC | 16:27 | |
| anteaya | blogan: that is a poor choice for a depends-on workflow | 16:27 |
| blogan | anteaya: yeah, since the one got merged into master i can remove this depends-on | 16:27 |
| anteaya | clarkb: sorry, I missed that | 16:27 |
| blogan | anteaya: i didnt add the backport ones, just the one to master | 16:28 |
| anteaya | blogan: in future don't put a depends-on that links to 3 different patches | 16:28 |
| anteaya | blogan: well socialize a solution then, but zuul will take you literally if you state you depend on 3 different patches | 16:28 |
| *** stevemar_znc is now known as stevemar | 16:28 | |
| blogan | anteaya: i didn't, i think some person(s) forgot to remove the depends-on when doing a cherry-pick to | 16:28 |
| anteaya | no the cherry picks use the same change id | 16:29 |
| blogan | anteaya: will do, thanks again | 16:29 |
| anteaya | clarkb fungi thanks | 16:29 |
| blogan | anteaya: thats what i mean sorry, so in that case what would the proper workflow be? generate a new change-idf or the backports? | 16:29 |
| anteaya | to be honest I don't know | 16:30 |
| fungi | the zuul behavior with regard to change ids on multiple branches is at odds with the backport workflow duplicating change ids on other branches | 16:31 |
| fungi | especially if you end up with one of them getting abandoned. then you can never merge the depending-on change without removing the depends-on from its commit message first | 16:31 |
| fungi | so easiest solution for now is to just remove the depends-on from the commit message and then reapprove the change once the update passes jobs | 16:32 |
| jesusaurus | fungi: i wrote up 254957 to address exactly that | 16:32 |
| blogan | fungi: yeah thats what i figured would be the solution for now | 16:32 |
| openstackgerrit | Brant Knudson proposed openstack-infra/project-config: Change postgres job to use standard deployment https://review.openstack.org/258006 | 16:33 |
| fungi | jesusaurus: we've discussed and dismissed that behavior change in the past | 16:33 |
| jesusaurus | fungi: oh, i missed that discussion | 16:33 |
| fungi | jesusaurus: it averts a disaster where my change depends on some functionality in your change, and then your change gets abandoned and someone blindly approves mine | 16:33 |
| *** ifarkas has quit IRC | 16:34 | |
| *** yamamoto has joined #openstack-infra | 16:34 | |
| fungi | if zuul were free to ignore abandoned changes, then once it merges my change *boom* | 16:34 |
| jesusaurus | i see | 16:34 |
| blogan | anteaya, fungi, clarkb: thanks for the help and information | 16:34 |
| jeblair | Shrews, mordred: if you were considering firing up gertty, gimme a sec to fix some things i noticed over the weekend | 16:34 |
| fungi | admittedly this is partly because depends-on encodes two behaviors: 1. don't merge change B until A merges, 2. if possible test change B with change A applied | 16:35 |
| fungi | jesusaurus: so your change is an attempt to solve for behavior 2, while missing the safeguard it provided for behavior 1 | 16:35 |
| anteaya | blogan: thanks for your patience and understanding | 16:36 |
| blogan | fungi, jesusaurus: what if it checked only for the same branch? which would solve the problem i've run into? | 16:36 |
| *** reed has quit IRC | 16:36 | |
| *** reed has joined #openstack-infra | 16:36 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/shade: Fall back to nova on 404s from glance https://review.openstack.org/260087 | 16:36 |
| fungi | jeblair: i've been using 1.3.0 since friday, albeit lightly. i'll upgrade once you have tagged | 16:36 |
| *** zeih has joined #openstack-infra | 16:36 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/shade: Fall back to nova on 404s from glance https://review.openstack.org/260087 | 16:37 |
| mordred | jeblair, Shrews: ^^ this patch makes me sad | 16:37 |
| *** gyee has joined #openstack-infra | 16:37 | |
| fungi | blogan: depends-on is intended to support dependencies for upgrade tests as well, so for example we can post a kilo change that grenade needs in place to be able to merge some change on liberty | 16:37 |
| mordred | jeblair, Shrews: I cannot explain WHY I'm getting 404s on hpcloud | 16:37 |
| mordred | it's on ly HP | 16:37 |
| mordred | and it's not broken in the command line client | 16:37 |
| fungi | blogan: so cross-branch dependencies were an intentional part of the design | 16:37 |
| blogan | fungi: ah yeah, i understand the issue there now | 16:37 |
| *** flepied1 has quit IRC | 16:38 | |
| *** yamamoto has quit IRC | 16:38 | |
| *** flepied has joined #openstack-infra | 16:38 | |
| blogan | tough one to solve, this probably happens so infrequently that its not too big of a deal | 16:38 |
| *** EricGonczer_ has joined #openstack-infra | 16:38 | |
| anteaya | now that bots are back up, here is the paste of the third-party meeting earlier: http://paste.openstack.org/show/482421/ | 16:39 |
| *** f1ller is now known as filler | 16:41 | |
| *** jpr has quit IRC | 16:41 | |
| *** markvoelker has joined #openstack-infra | 16:43 | |
| *** maishsk has joined #openstack-infra | 16:44 | |
| notmorgan | mordred: weird only in the CLI? | 16:44 |
| openstackgerrit | Monty Taylor proposed openstack/requirements: Bump os-client-config constraint https://review.openstack.org/260092 | 16:45 |
| mordred | notmorgan: no, CLI works. shade list_images() does not | 16:45 |
| *** dimtruck is now known as zz_dimtruck | 16:45 | |
| mordred | notmorgan: also, I cannot for the life of me figure out how to get the glanceclient debug output to print when I'm consuming through shade | 16:45 |
| openstackgerrit | mathieu bultel proposed openstack-infra/shade: Add normalize stack function for heat stack_list https://review.openstack.org/254337 | 16:46 |
| *** apuimedo has quit IRC | 16:46 | |
| *** apuimedo_ has quit IRC | 16:47 | |
| openstackgerrit | James E. Blair proposed openstack/gertty: Fix multi-key handling at top level https://review.openstack.org/260084 | 16:47 |
| openstackgerrit | James E. Blair proposed openstack/gertty: Fix config validation to accept new keymap format https://review.openstack.org/259253 | 16:47 |
| openstackgerrit | James E. Blair proposed openstack/gertty: Support >= 2.9 query batching https://review.openstack.org/259705 | 16:47 |
| openstackgerrit | Jaume Devesa proposed openstack-infra/project-config: Add puppet-midonet to OpenStack infrastructure https://review.openstack.org/260093 | 16:47 |
| *** apuimedo has joined #openstack-infra | 16:47 | |
| *** flepied has quit IRC | 16:47 | |
| *** apuimedo_ has joined #openstack-infra | 16:48 | |
| *** flepied has joined #openstack-infra | 16:48 | |
| anteaya | If anyone cares, I'm out on Wednesday until after Christmas, will be back next week | 16:48 |
| *** markvoelker has quit IRC | 16:48 | |
| *** hurgleburgler has joined #openstack-infra | 16:48 | |
| jeblair | anteaya: thanks, enjoy! | 16:48 |
| openstackgerrit | Jaume Devesa proposed openstack-infra/project-config: Add fuel-plugin-midonet notifications at #midonet https://review.openstack.org/260094 | 16:49 |
| anteaya | jeblair: thanks | 16:49 |
| *** flepied has quit IRC | 16:49 | |
| anteaya | still here until Wednesday just sharing | 16:49 |
| *** vgridnev has joined #openstack-infra | 16:50 | |
| Shrews | mordred: weird | 16:51 |
| mordred | woot. I figured out http tracing via shade | 16:51 |
| jeblair | oh! mrunge found and fixed a long standing bug in 259992 i think | 16:51 |
| krotscheck | jeblair: I have a new update from the npm mirror team, they're basically saying "You need a TB". | 16:52 |
| mordred | Shrews: REQ: curl -g -i -X GET https://region-b.geo-1.images.hpcloudsvc.com:443/v1.0/v2/images?limit=1000 | 16:52 |
| mordred | that would be it | 16:52 |
| *** ihrachys has quit IRC | 16:52 | |
| mordred | Shrews, notmorgan: bug in that os-client-config pass version patch I think | 16:52 |
| anteaya | krotscheck: that presents a challenge | 16:52 |
| openstackgerrit | Diana Whitten proposed openstack-infra/project-config: Add openstack-ux repo https://review.openstack.org/259239 | 16:53 |
| Shrews | mordred: is that due to the occ change that just merged? | 16:54 |
| jeblair | krotscheck: neat. we have a couple hundred mb. so i think we need someone to a) figure out if we can get 1.5tb+ on each of our providers; b) see if we can use afs; c) look into some other gfs; or d) pivot to a caching proxy strategy rather than mirrors. | 16:54 |
| mordred | Shrews: yah. working on a fix now | 16:54 |
| Shrews | mordred: and a test, i'm sure ;) | 16:54 |
| *** apuimedo has quit IRC | 16:54 | |
| krotscheck | jeblair: Or e) Write an opportunistic mirror. | 16:54 |
| jeblair | mordred, fungi, clarkb, greghaynes: see my conv w/ krotscheck | 16:54 |
| jeblair | krotscheck: ya | 16:54 |
| *** apuimedo_ has quit IRC | 16:54 | |
| *** armax has joined #openstack-infra | 16:55 | |
| *** apuimedo has joined #openstack-infra | 16:55 | |
| *** apuimedo_ has joined #openstack-infra | 16:56 | |
| krotscheck | jeblair: Caching Proxy is basically an opportunistic mirror. | 16:56 |
| krotscheck | And it sanely handles the weirdness of "Some of our mirrors are normal mirrors and some we build ourselves (wheel)" | 16:56 |
| jeblair | krotscheck: with the advantage of being content agnostic | 16:56 |
| jeblair | i guess wheel building would fit in there by us running a single "real" wheel mirror and then just accesing it through the proxy like anything else. | 16:57 |
| krotscheck | Yep | 16:57 |
| mordred | jeblair: wow | 16:57 |
| krotscheck | So, would we prefer a swift thing or building our own thing? | 16:57 |
| *** zeih has quit IRC | 16:58 | |
| hashar | are you talking about caching wheels ? | 16:58 |
| jeblair | krotscheck: not following that one? | 16:58 |
| krotscheck | jeblair: We can configure swift to basically be a CDN no? Or we can build a server to be a CDN. | 16:58 |
| * krotscheck may have that wrong | 16:58 | |
| krotscheck | hashar: Yes | 16:58 |
| mordred | krotscheck: unfortunately, we need per-cloud-region things- so I think swift would be extra complex | 16:59 |
| jeblair | krotscheck: oh, i think that's "rax public access to swift is via cdn". i think generic cdn is a separate non-openstack service. | 16:59 |
| mordred | krotscheck: putting things in rax swift will not help our nodes in ovh in terms of not hitting internets | 16:59 |
| mordred | for instance | 16:59 |
| *** abregman has quit IRC | 17:00 | |
| mordred | krotscheck: does the npm format allow for filesystem storae? or does it need 1TB of contiguous block? | 17:00 |
| krotscheck | mordred: Ok, so apache's mod_disk_cache? | 17:00 |
| jeblair | mordred: ah i think i see where krotscheck is going -- if we build a mirror in swift, and publish it via cdn, we will have brought the mirror very close to ovh by virtue of the cdn | 17:01 |
| jeblair | however, it's still ovh-external traffic | 17:01 |
| mordred | yah. and it has the downside of not being able to do atomic updates | 17:01 |
| hashar | I myself looked at several possibilities to cache packages from various package managers. It is rather messy. | 17:01 |
| jeblair | so network traffic wise, not as optimal as us having something inside each provider | 17:01 |
| jeblair | mordred, krotscheck: or squid | 17:01 |
| krotscheck | mordred: The NPM Mirror tries to simulate a couchdb in the filesystem. | 17:02 |
| hashar | I looked at Squid with SSL interception (ssl-bump feature). Wrote my findings at https://phabricator.wikimedia.org/T116015 | 17:02 |
| mordred | jeblair: isn't there a problem with caching ... yeah, that ^^ | 17:02 |
| hashar | for python there is devpi that is rather straightforward to setup ( evaluated at https://phabricator.wikimedia.org/T114871 ) | 17:03 |
| mordred | oh golly no | 17:03 |
| hashar | hehe | 17:03 |
| mordred | devpi is way not a production service | 17:03 |
| krotscheck | mordred: And, well, those documents in couchdb point at CDN locations that npm itself manages. Those docs have to be rewritten on disk before they can find the package. | 17:03 |
| * krotscheck ponders squid. | 17:04 | |
| mordred | krotscheck: so part of our mirroring code would need to rewrite the documents as we store them, yeah? | 17:04 |
| krotscheck | mordred: Yep. There's a program out there called "registry-static" taht does this for us. | 17:04 |
| hashar | Artifactory does have support for npm/pip/gem caching but it is not included in the open source solution (though the code is free apparently) | 17:04 |
| krotscheck | And it's maintained by the yahoo people. | 17:04 |
| mordred | jeblair, krotscheck: I'd kind of love to make an earnest stab at AFS mirror | 17:04 |
| mordred | krotscheck: neat! | 17:05 |
| jeblair | mordred: i'm game | 17:05 |
| krotscheck | mordred: But that requires 1TB of disk. | 17:05 |
| mordred | krotscheck: not 1TB on a single server | 17:05 |
| jeblair | mordred: the infrastructure is basically ready to go | 17:05 |
| krotscheck | mordred: True. | 17:05 |
| openstackgerrit | Merged openstack-infra/jenkins-job-builder: Implement bzr SCM module. https://review.openstack.org/234365 | 17:05 |
| mordred | and we can get atomic updates for our things that need it | 17:05 |
| krotscheck | mordred, jeblair: Well, since we're creating brand new mirror boxes anyway... | 17:05 |
| *** _nadya_ has quit IRC | 17:06 | |
| hashar | krotscheck: a problem with squid / http caches, is that they will not cache compiled modules which might needed if you want to speed up install | 17:06 |
| fungi | yeah, content-agnostic proxies actually create new problems with package managers | 17:06 |
| zaro | morning | 17:06 |
| jeblair | krotscheck: afs lets us have a single instance of the mirror on one arbitrarily large filesystem (which can be spread across many servers), then we put mirror nodes in each provider which are simply afs clients with a large local cache | 17:06 |
| fungi | because they're not aware of the holistic state of what's getting cached | 17:06 |
| *** hdd has joined #openstack-infra | 17:06 | |
| krotscheck | jeblair: mirror master and afs client, right. | 17:07 |
| jeblair | krotscheck: as the local cache warms up, the local afs client mirror-serving host converges to "serve the subset of files we are interested in from local disk" automatically | 17:07 |
| *** ZZelle has joined #openstack-infra | 17:07 | |
| fungi | for example, apt has several smart proxy implementations available that deal with those consistency issues for cached access to apt package repositories | 17:07 |
| jeblair | anything it doesn't have in its local cache, it fetches over the internet from the afs server | 17:07 |
| mordred | but - it does so in a resilent way that assumes the internet will break :) | 17:07 |
| krotscheck | jeblair: The main issue that would cause with npm is that the URL rewrite to http:/mirror-master/foo.tgz would need to be different for each afs client. | 17:07 |
| openstackgerrit | Christopher Aedo proposed openstack-infra/project-config: Fixing app-catalog dead-link checker https://review.openstack.org/260115 | 17:08 |
| mordred | krotscheck: can it do file urls? | 17:08 |
| *** notnownikki has quit IRC | 17:08 | |
| mordred | krotscheck: or do they have to be http? | 17:08 |
| *** devananda has quit IRC | 17:08 | |
| krotscheck | mordred: Lemme check. | 17:08 |
| jeblair | krotscheck: when i was looking into afs for pypi mirroring, i did some mod-rewrite magic | 17:08 |
| jeblair | mordred: you're thinking of having the test nodes fetch from afs? | 17:09 |
| krotscheck | jeblair: whoa, mod-rewrite can parse json? | 17:09 |
| jeblair | krotscheck: no | 17:09 |
| krotscheck | oh. | 17:09 |
| openstackgerrit | Jaume Devesa proposed openstack-infra/project-config: Add fuel-plugin-midonet notifications at #midonet https://review.openstack.org/260094 | 17:09 |
| krotscheck | you had my hopes up! | 17:09 |
| jeblair | well, i mean, maybe | 17:09 |
| mordred | krotscheck: if they can do file, we can rewrite the docuemnts to contain file:///afs/openstack.org/mirror/npm/foo | 17:09 |
| jeblair | krotscheck: i don't understand the rewriting problem yet, just throwing that out there in case it makes sense | 17:09 |
| krotscheck | mordred: Would that actually provide a performance improvement? | 17:10 |
| mordred | which doesn't work well for pip, becuase pip deal poorly witih file urls | 17:10 |
| mordred | krotscheck: not really - but it would fix the url-in-json-document problem | 17:10 |
| jeblair | mordred: again -- that's for having our actual test nodes use afs -- is that what you want to do? | 17:10 |
| krotscheck | jeblair: Right. a package might be index.json -> { things, more things, package: 'https://location/foo.version.tgz'} | 17:10 |
| jeblair | krotscheck: and in "package: <location>", <location> must be an absolute rather than relative url? | 17:11 |
| mordred | jeblair: well, aiui, the above ^^ says that the files we store on disk in the mirror may contain explcit urls, rather than relative paths | 17:11 |
| krotscheck | jeblair: I _think_ yes. | 17:11 |
| krotscheck | jeblair: I can test that | 17:11 |
| krotscheck | jeblair: But I believe that it does, because the index.json is downloaded to "Wherever the npm command is invoked", which may not know the registry context. | 17:11 |
| jeblair | krotscheck: oh, then there's a separate invocation to actuall get the thing and it's lost the context of the original index download? | 17:12 |
| anteaya | morning zaro | 17:13 |
| krotscheck | jeblair: I'm not certain. Will have to rebuild a mirror with the new puppet modules to test. | 17:13 |
| hashar | krotscheck: for npm, I am pretty sure you also need to rewrite urls in the package.json files | 17:13 |
| krotscheck | hashar: Probably. | 17:13 |
| *** maishsk has quit IRC | 17:13 | |
| hashar | krotscheck: cause the ones coming from npmjs.org would have hardcoded npmjs.org url in their json payload :( | 17:13 |
| *** maishsk has joined #openstack-infra | 17:14 | |
| jeblair | krotscheck, mordred: so assuming that's the case, we can solve it either with mordred's idea of using file urls (which means afs clients on all our test nodes), or we could solve it by having a different _index_ for each mirror so it can have different urls, but still having a single set of packages. depending on the exact directory structure, there are some afs tricks we can do to make that easier. failing those, we can have mod_rewrite | 17:14 |
| hashar | krotscheck: one of Jfrog Artifactory / Sonatype Nexus have plugins that handles the rewrite for you. Aka an "intelligent" proxy/cache | 17:14 |
| *** devananda has joined #openstack-infra | 17:15 | |
| notmorgan | mordred: fwiw, we probably need to bump the min-req on OCC as well for novalicent. | 17:15 |
| krotscheck | hashar: Yeah, but those are $$ and not happily licensed. | 17:15 |
| notmorgan | mordred: not just the upper constraint | 17:15 |
| openstackgerrit | Randall Burt proposed openstack-infra/elastic-recheck: Add query for bug 1527483 https://review.openstack.org/260119 | 17:15 |
| openstack | bug 1527483 in neutron "VPNaaS - No providers specified for 'VPN' service" [Critical,Confirmed] https://launchpad.net/bugs/1527483 - Assigned to Martin Hickey (martin-hickey) | 17:15 |
| hashar | krotscheck: maybe the plugins are actually open source but just disabled via a feature flag :-D (not sure whether it will breach their license though) | 17:16 |
| krotscheck | jeblair, mordred: Is there a compelling other reason to add an AFS client to our nodes? It seems a bit... heavy handed. | 17:16 |
| *** woodster_ has quit IRC | 17:16 | |
| mordred | krotscheck, jeblair: new idea | 17:17 |
| *** rlandy has quit IRC | 17:17 | |
| mordred | we use the ready script to write an entry into /etc/hosts that points to a per-region IP of a server that's running Apache on AFS | 17:17 |
| mordred | and we rewrite the node documents to refer to that hostname | 17:17 |
| jeblair | krotscheck: if we did that, we could concievably drop mirrors altogether, however, warming the afs cache on each test node would be difficult. i agree, i think it's not our best option. | 17:17 |
| mordred | so that the docuemnts can be served | 17:17 |
| hashar | another idea we had is that when a change merge, we run a job that install the dependencies, save the package managers caches and send them to an object store like swift. Namespaced by repo/branch. Then jobs can grab that central cache to warm up the instance. | 17:17 |
| mordred | hashar: we still don't have a swift per region though | 17:18 |
| jeblair | mordred: i'm usually not excited about ideas that involve writing to /etc/hosts, but that actually sounds like a not terrible use of that | 17:18 |
| mordred | jeblair: in this case, I _think_ it would solve the problem | 17:18 |
| mordred | which is that we have some physical needs to put an identical thing in each region | 17:18 |
| krotscheck | mordred, jeblair: Would that cause issues with known_hosts? | 17:18 |
| jeblair | yeah, basically, npm wants to know what "the mirror" is, and we'll tell it "the mirror".. just differently for each region. | 17:19 |
| mordred | krotscheck: it shouldn't - becaues we'd pick a fake hostname that we'd only use for http | 17:19 |
| jeblair | i mean, if we had a more sophisticated dns system, we could do geolocal dns | 17:19 |
| krotscheck | Ok, that should be doable. | 17:19 |
| *** piet has quit IRC | 17:19 | |
| mordred | "mirror.local" or something | 17:19 |
| jeblair | which is a perfectly respectable thing to do and functionally equivalent to this :) | 17:19 |
| jeblair | i like it if we can get it past fungi :) | 17:19 |
| mordred | fungi: ? | 17:20 |
| krotscheck | jeblair: This NPM mirror effort of mine is becoming a VERY hairy yak ;) | 17:20 |
| *** _nadya_ has joined #openstack-infra | 17:20 | |
| notmorgan | jeblair: i would advocate dns but... | 17:20 |
| mordred | krotscheck: but it's kind of become a fun on | 17:20 |
| mordred | one | 17:20 |
| jeblair | krotscheck: i like that "pivot" has positive connotations. | 17:20 |
| *** maishsk has quit IRC | 17:20 | |
| mordred | Shrews: ok. found the occ problem | 17:21 |
| mordred | Shrews: it actualy was not in the version patch | 17:21 |
| Shrews | \o/ | 17:21 |
| fungi | jeblair: mordred: you know what a luddite i am. this reminds me of the global hosts file from the pre-dns days ;) | 17:22 |
| mordred | Shrews: you're not going to like the patch that solves it :) /me glares at glance for being different | 17:22 |
| fungi | i think it's not a terrible idea now for the same reasons it wasn't a terrible idea then | 17:22 |
| fungi | dns came about once the global hosts file had hundreds of entries being kept in sync via ftp from many hundreds of sites | 17:22 |
| fungi | it scaled pretty well below that | 17:22 |
| anteaya | fungi: yes luddite is the word that comes to mind when I think of you | 17:23 |
| *** mudassirlatif has joined #openstack-infra | 17:23 | |
| *** sfinucan has quit IRC | 17:24 | |
| * krotscheck has a bind9 ansible playbook around here somewhere.... | 17:24 | |
| fungi | krotscheck: i mostly just don't want to see the npm yak shave stop us from finishing the wheel building mirror work, which will have a huge positive impact on most of the jobs we run | 17:24 |
| * anteaya picks up her stone hammer | 17:24 | |
| *** hashar has quit IRC | 17:24 | |
| jeblair | fungi: whew, thanks for clarifying; i wasn't sure how many layers of sarcasm and irony i was seing in your initial response :) | 17:24 |
| krotscheck | fungi: It's easy enough to peel those apart, they're on separate branches. | 17:24 |
| jeblair | turns out it was just my own reflection | 17:24 |
| fungi | hah | 17:25 |
| krotscheck | fungi: If we can start with this, and only this patch, we can focus on the wheel things for a bit. | 17:26 |
| krotscheck | https://review.openstack.org/#/c/253236/ | 17:26 |
| docaedo | Any change I could get a quick look at https://review.openstack.org/260115 ? It's a one line change that fixes an experimental test that merged this morning (SHOULD fix it anyway) | 17:26 |
| krotscheck | (The followup patch is the one where we actually have to make servers) | 17:26 |
| fungi | krotscheck: awesome, i'll stick that on my priority pile now | 17:27 |
| *** piet has joined #openstack-infra | 17:27 | |
| krotscheck | fungi: Thanks :) | 17:28 |
| *** apuimedo has quit IRC | 17:28 | |
| openstackgerrit | James E. Blair proposed openstack/gertty: Add process mark to project list https://review.openstack.org/259701 | 17:28 |
| openstackgerrit | James E. Blair proposed openstack/gertty: Add project topics https://review.openstack.org/259671 | 17:28 |
| openstackgerrit | James E. Blair proposed openstack/gertty: Cache counts of project changes https://review.openstack.org/260085 | 17:28 |
| openstackgerrit | James E. Blair proposed openstack/gertty: Fix multi-key handling in diff view https://review.openstack.org/260123 | 17:28 |
| *** apuimedo_ has quit IRC | 17:28 | |
| openstackgerrit | James E. Blair proposed openstack/gertty: Fix list index out of range https://review.openstack.org/259992 | 17:29 |
| openstackgerrit | James E. Blair proposed openstack/gertty: Fix multi-key handling in diff view https://review.openstack.org/260123 | 17:29 |
| jeblair | (yeah, i'm using gertty to stack up commits into different branches) | 17:30 |
| *** fawadkhaliq has quit IRC | 17:30 | |
| EmilienM | how can we merge two patches that Depends-On each others (if possible)? (note that they only pass unit/functional tests with that Depends-On) | 17:30 |
| *** apuimedo has joined #openstack-infra | 17:30 | |
| fungi | EmilienM: presumably only one needs to depends-on the other? | 17:30 |
| openstackgerrit | Thanh Ha proposed openstack-infra/jenkins-job-builder: Add support for credentials properties in authorization matrix https://review.openstack.org/199547 | 17:30 |
| *** apuimedo_ has joined #openstack-infra | 17:30 | |
| *** AJaeger has joined #openstack-infra | 17:30 | |
| EmilienM | fungi: I have a chicken&egg issue | 17:30 |
| jeblair | mordred, Shrews: this is how i use gertty to understand patch series: http://paste.openstack.org/show/482434/ | 17:31 |
| EmilienM | I'll need to land one first | 17:31 |
| *** hashar has joined #openstack-infra | 17:31 | |
| fungi | EmilienM: two patches should not interdepend. that breaks continuous deployment assumptions. is it possible that if it's staged into three or four changes it can be implemented incrementally and asynchronously instead? | 17:31 |
| krotscheck | jeblair: So... we're now doing AFS mirrors and geolocal dns? | 17:31 |
| jeblair | mordred, Shrews: would it help if that view also indicated whether something is outdated? | 17:31 |
| EmilienM | fungi: yes it can | 17:32 |
| EmilienM | fungi: if I touch to unit testing | 17:32 |
| jeblair | krotscheck: it seems among our better options. since we have an afs cell, we should probably do some quick informal testing with npm | 17:32 |
| fungi | EmilienM: zuul assumes all changes are serialized when gating, so if your changes have to be introduced in lock-step it can't actually test at least one of them | 17:32 |
| jeblair | krotscheck: just to make sure there isn't a showstopper | 17:32 |
| EmilienM | fungi: ok, I see | 17:32 |
| EmilienM | fungi: I'll split commit in commits | 17:32 |
| openstackgerrit | Sean M. Collins proposed openstack-infra/devstack-gate: Stop hard coding SERVICE_HOST https://review.openstack.org/229111 | 17:33 |
| krotscheck | jeblair: Works for me. Is this a thing you'd like me to do, or someone else? | 17:33 |
| jeblair | krotscheck: maybe we could write an index file and drop a couple packages in afs, and then serve that through apache and try installing some things that way | 17:33 |
| jeblair | krotscheck: you up for installing a local afs client? | 17:33 |
| krotscheck | jeblair: I can try! | 17:33 |
| fungi | EmilienM: what i've seen in similar past situations is that it also makes reviewing easier, because you are clearly calling out the order in which things change | 17:33 |
| krotscheck | jeblair: Not today though | 17:33 |
| *** unicell1 has quit IRC | 17:33 | |
| jeblair | krotscheck: i think we even have instructions | 17:33 |
| jeblair | krotscheck: http://docs.openstack.org/infra/system-config/afs.html#client-configuration | 17:34 |
| jeblair | krotscheck: so we could make a principal for you and give you full access to a volume, and then you can probably do most of the work from there | 17:34 |
| mordred | ++ | 17:34 |
| krotscheck | jeblair: That sounds like authority. | 17:35 |
| *** apoorvad has joined #openstack-infra | 17:35 | |
| * krotscheck is allergic to authority | 17:35 | |
| mordred | distributed kerberized filesystems fwi | 17:35 |
| mordred | ftw | 17:35 |
| jeblair | mordred: fyi its ftw | 17:35 |
| mordred | krotscheck: nah. just write access to a volume in our AFS cell | 17:35 |
| krotscheck | Oh good. | 17:35 |
| mordred | krotscheck: so that you can write files to it | 17:35 |
| mordred | krotscheck: without needing one of us to do something | 17:35 |
| notmorgan | mordred: amazing when technology works | 17:35 |
| mordred | notmorgan: inorite? | 17:35 |
| jeblair | i'll make the principal and volume now while i'm thinking about it | 17:35 |
| *** otter768 has joined #openstack-infra | 17:35 | |
| krotscheck | So I could run a registry-mirroring script for a few cycles, write into that cell with the appropriate settings, and then see if I can install from there. | 17:36 |
| anteaya | jeblair: pretty | 17:36 |
| mordred | krotscheck: yes. I think that's a great idea - can you do that on a small subset of packages? | 17:36 |
| krotscheck | mordred: I can. There's a "Just run this for one package" setting. | 17:36 |
| mordred | neat! | 17:36 |
| mordred | that sounds like a stellar test | 17:36 |
| jeblair | krotscheck: username krotscheck? | 17:37 |
| krotscheck | jeblair: Works for me. | 17:37 |
| *** maishsk has joined #openstack-infra | 17:37 | |
| * krotscheck feels like he's in college again :) | 17:38 | |
| mordred | krotscheck: it's out main goal most days | 17:38 |
| mordred | krotscheck: "make these computers work almost as well as the ones from college" | 17:38 |
| krotscheck | mmmmmspark | 17:38 |
| mordred | s/out/our/ | 17:38 |
| *** BobBall is now known as BobBall_AWOL | 17:39 | |
| mordred | notmorgan, Shrews: https://review.openstack.org/260126 | 17:39 |
| mordred | notmorgan, Shrews: need to add tests - just wanted you to see the thing | 17:39 |
| *** otter768 has quit IRC | 17:40 | |
| krotscheck | mordred, jeblair: While I'm at it, I'll go ahead and create a different server with a public IP and an apache vhost. | 17:40 |
| openstackgerrit | James E. Blair proposed openstack-infra/system-config: Add krotscheck to users https://review.openstack.org/260129 | 17:41 |
| jeblair | mordred, fungi: ^ instapprove to get the uid reservation? | 17:41 |
| mordred | jeblair, fungi: +2 | 17:42 |
| *** zeih has joined #openstack-infra | 17:42 | |
| jeblair | krotscheck: different server for what? | 17:42 |
| openstackgerrit | Michael Krotscheck proposed openstack-infra/system-config: Add krotscheck to users https://review.openstack.org/260129 | 17:43 |
| *** admin0 has joined #openstack-infra | 17:43 | |
| * krotscheck fixed a typo ^^ | 17:43 | |
| krotscheck | jeblair: One to run the mirror script on. One to run apache. Full test of transfer to a 'remote' mirror server. | 17:43 |
| *** maishsk has quit IRC | 17:44 | |
| fungi | i need to go run some lunch errands, back in a little while | 17:44 |
| krotscheck | AFK as well, back in 20 | 17:44 |
| *** markvoelker has joined #openstack-infra | 17:44 | |
| jeblair | krotscheck: (sorry for the typo, i blame this chiclet keyboard) | 17:44 |
| *** yaume has quit IRC | 17:44 | |
| *** admin0 has quit IRC | 17:46 | |
| *** zz_dimtruck is now known as dimtruck | 17:46 | |
| *** maishsk has joined #openstack-infra | 17:46 | |
| bkero | Always blame chiclet keyboard | 17:47 |
| *** e0ne has quit IRC | 17:49 | |
| nibalizer | asselin__: fungi likely we can land that patch but pabelanger_ is doing work to have us simply use the apache module | 17:49 |
| *** markvoelker has quit IRC | 17:49 | |
| *** jpr has joined #openstack-infra | 17:49 | |
| *** apuimedo has quit IRC | 17:50 | |
| *** ayoung has quit IRC | 17:51 | |
| *** amuller is now known as amuller_afk | 17:51 | |
| *** gyee has quit IRC | 17:51 | |
| *** apuimedo has joined #openstack-infra | 17:53 | |
| *** maishsk has quit IRC | 17:53 | |
| michaelxin | Thanks for your help all the time. I have two CR: https://review.openstack.org/#/c/260120/ and https://review.openstack.org/#/c/260099/ Both are in the stage of "Needs Verified". But Jenkins already +1 both. Any idea why? Thanks. | 17:53 |
| *** apuimedo_ has quit IRC | 17:54 | |
| openstackgerrit | Merged openstack/gertty: Fix config validation to accept new keymap format https://review.openstack.org/259253 | 17:54 |
| *** hashar has quit IRC | 17:54 | |
| *** maishsk has joined #openstack-infra | 17:54 | |
| *** apuimedo_ has joined #openstack-infra | 17:54 | |
| *** ZZelle has quit IRC | 17:54 | |
| asselin | nibalizer, if pabelanger_ change won't happen soon, we just merge this one and switch to apache module from there. | 17:55 |
| openstackgerrit | James E. Blair proposed openstack/gertty: Fix multi-key handling at top level https://review.openstack.org/260084 | 17:56 |
| openstackgerrit | Merged openstack/gertty: Fix multi-key handling at top level https://review.openstack.org/260084 | 17:56 |
| jroll | michaelxin: those patches depend on each other | 17:57 |
| openstackgerrit | Merged openstack-infra/system-config: Add krotscheck to users https://review.openstack.org/260129 | 17:57 |
| *** smarcet1981 has quit IRC | 17:57 | |
| michaelxin | jroll: How to fix it? Thanks. | 17:57 |
| Shrews | jeblair: I would LOVE for that view to show outdated somehow | 17:57 |
| jroll | michaelxin: you have the first patch dependent on the second via depends-on, and the second dependent on the first because they were submitted together in that order | 17:57 |
| jroll | michaelxin: you need to remove the depends-on there | 17:57 |
| jeblair | Shrews: cool, i bet it won't be too hard | 17:58 |
| michaelxin | jroll: Got it. Thanks. | 17:59 |
| jroll | np | 17:59 |
| *** apuimedo_ has quit IRC | 17:59 | |
| *** Swami has joined #openstack-infra | 18:00 | |
| *** salv-orl_ has joined #openstack-infra | 18:00 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/shade: Add option to enable HTTP tracing https://review.openstack.org/260134 | 18:00 |
| jeblair | Shrews, mordred: were you having a problem with zuul enqueing changes? | 18:00 |
| *** apuimedo_ has joined #openstack-infra | 18:00 | |
| mordred | jeblair: not to my knowledge | 18:00 |
| mordred | Shrews, notmorgan: ^^ that's the easy way to get http trace logging when hacking with shade | 18:01 |
| jeblair | ok, wasn't sure if your troubles with dependent changes extended to that | 18:01 |
| mordred | nah. it was just UI not being able to figure out what the hell was going on | 18:01 |
| *** eharney has quit IRC | 18:01 | |
| jeblair | i am, and i suspect it's another manifestation of the change caching thing that hopefully that patch from jhesketh fixes | 18:01 |
| mordred | notmorgan, Shrews: https://review.openstack.org/260126 fixes the glance issue with HP and adds more tests | 18:01 |
| openstackgerrit | James E. Blair proposed openstack/gertty: Support >= 2.9 query batching https://review.openstack.org/259705 | 18:01 |
| openstackgerrit | Travis McPeak proposed openstack-infra/project-config: Adding voting Bandit on Bandit gate https://review.openstack.org/260135 | 18:02 |
| openstackgerrit | Merged openstack/gertty: Support >= 2.9 query batching https://review.openstack.org/259705 | 18:02 |
| openstackgerrit | James E. Blair proposed openstack/gertty: Fix multi-key handling in diff view https://review.openstack.org/260123 | 18:02 |
| *** apuimedo has quit IRC | 18:02 | |
| *** zeih has quit IRC | 18:02 | |
| openstackgerrit | James E. Blair proposed openstack/gertty: Fix list index out of range https://review.openstack.org/259992 | 18:02 |
| openstackgerrit | Merged openstack/gertty: Fix multi-key handling in diff view https://review.openstack.org/260123 | 18:02 |
| *** hurgleburgler has quit IRC | 18:03 | |
| openstackgerrit | Merged openstack/gertty: Fix list index out of range https://review.openstack.org/259992 | 18:03 |
| *** apuimedo has joined #openstack-infra | 18:03 | |
| *** salv-orlando has quit IRC | 18:03 | |
| notmorgan | mordred: going to need to bug someone about shade functional tests against a live cloud (if there is a good doc on it). Seems wierd. | 18:03 |
| *** jsavak has joined #openstack-infra | 18:03 | |
| openstackgerrit | James E. Blair proposed openstack/gertty: Add process mark to project list https://review.openstack.org/259701 | 18:03 |
| openstackgerrit | James E. Blair proposed openstack/gertty: Add project topics https://review.openstack.org/259671 | 18:03 |
| openstackgerrit | James E. Blair proposed openstack/gertty: Cache counts of project changes https://review.openstack.org/260085 | 18:03 |
| jeblair | mordred, fungi, Shrews: okay, gertty master should be much happier with gerrit 2.11. | 18:04 |
| zaro | was wondering if anybody here can help me setup a smtp (outgoing mail only) server? | 18:05 |
| jeblair | mordred, fungi, Shrews: or check out https://review.openstack.org/260085 for project topics if you are adventurous (it has a schema migration, so you may want to keep a db backup) | 18:05 |
| EmilienM | it would be useful if zuul UI could filter by Gerrit topic, what do you think? | 18:05 |
| jeblair | zaro: "apt-get install exim-daemon-light" | 18:05 |
| mordred | zaro: ++ ^^ | 18:05 |
| zaro | thanks, was trying to do with postfix but couldn't get it to work. | 18:06 |
| *** dtantsur is now known as dtantsur|afk | 18:06 | |
| zaro | will give exim-daemon-light a try | 18:07 |
| *** apuimedo has quit IRC | 18:07 | |
| *** apuimedo_ has quit IRC | 18:07 | |
| openstackgerrit | Merged openstack/requirements: Updated from generate-constraints https://review.openstack.org/259741 | 18:07 |
| *** jsavak has quit IRC | 18:07 | |
| *** otter768 has joined #openstack-infra | 18:08 | |
| notmorgan | mordred: so now occ only uses endpoint override, just if the user requests we populate the override instead of endpoint to avoid double version construction? | 18:08 |
| *** yamahata has joined #openstack-infra | 18:08 | |
| *** jsavak has joined #openstack-infra | 18:08 | |
| *** apuimedo has joined #openstack-infra | 18:08 | |
| notmorgan | Bases on your patch to fix glance. | 18:08 |
| mordred | notmorgan: yah | 18:08 |
| mordred | notmorgan: for glance that is | 18:08 |
| notmorgan | Not a bad dodge | 18:08 |
| *** pblaho has quit IRC | 18:08 | |
| *** apuimedo_ has joined #openstack-infra | 18:09 | |
| notmorgan | Will check it again when Jenkins finishes and I am done with coffee. | 18:09 |
| notmorgan | But it looks good at a first glance | 18:09 |
| mordred | coolio. thanks | 18:09 |
| mordred | flaper87: if you are bored, you might want to look at https://review.openstack.org/260126 | 18:10 |
| anteaya | my vlc media player is sporting a christmas hat | 18:11 |
| anteaya | how festive | 18:12 |
| *** jsavak has quit IRC | 18:13 | |
| *** bastelfreak has joined #openstack-infra | 18:13 | |
| *** jerryz has joined #openstack-infra | 18:14 | |
| *** Guest95009 is now known as jgriffith | 18:15 | |
| anteaya | EmilienM: by zuul UI do you mean http://status.openstack.org/zuul/ | 18:15 |
| EmilienM | anteaya: yes | 18:15 |
| anteaya | that's not a bad idea | 18:15 |
| *** apuimedo has quit IRC | 18:15 | |
| anteaya | I always forget where the code for that view lives | 18:15 |
| * anteaya tries system-config | 18:16 | |
| *** apuimedo has joined #openstack-infra | 18:16 | |
| *** yamahata has quit IRC | 18:16 | |
| *** yamahata has joined #openstack-infra | 18:16 | |
| EmilienM | anteaya: somewhere around https://github.com/openstack-infra/zuul/blob/master/zuul/webapp.py | 18:17 |
| AJaeger | EmilienM: could you check https://review.openstack.org/#/c/260093/ , please? puppet-midonet repo... | 18:17 |
| *** ayoung has joined #openstack-infra | 18:18 | |
| EmilienM | AJaeger: ok | 18:18 |
| EmilienM | AJaeger: thx for highlight... | 18:18 |
| openstackgerrit | Merged openstack-infra/system-config: Use httpd::mod instead of httpd_mod https://review.openstack.org/253619 | 18:18 |
| anteaya | EmilienM: I think this should tell you where all the files are that come together to make that page: http://git.openstack.org/cgit/openstack-infra/system-config/tree/modules/openstack_project/manifests/status.pp#n161 | 18:20 |
| *** apuimedo has quit IRC | 18:20 | |
| *** apuimedo_ has quit IRC | 18:20 | |
| openstackgerrit | Travis McPeak proposed openstack-infra/project-config: Adding voting Bandit on Bandit gate https://review.openstack.org/260135 | 18:21 |
| EmilienM | AJaeger: done | 18:21 |
| *** rossella_s has quit IRC | 18:21 | |
| AJaeger | thanks, EmilienM ! | 18:21 |
| *** electrichead is now known as redrobot | 18:21 | |
| *** rossella_s has joined #openstack-infra | 18:22 | |
| *** apuimedo has joined #openstack-infra | 18:22 | |
| EmilienM | anteaya: http://git.openstack.org/cgit/openstack-infra/system-config/tree/modules/openstack_project/templates/zuul/status.js.erb | 18:22 |
| *** gyee has joined #openstack-infra | 18:22 | |
| openstackgerrit | Merged openstack-infra/puppet-openstackci: Allow logserver to run on CI server (part 2) https://review.openstack.org/254128 | 18:22 |
| *** apuimedo_ has joined #openstack-infra | 18:22 | |
| *** ZZelle has joined #openstack-infra | 18:22 | |
| anteaya | EmilienM: wonderful | 18:22 |
| EmilienM | anteaya: I wanted to make sure it's useful | 18:23 |
| *** unicell has joined #openstack-infra | 18:23 | |
| *** Hal2 has joined #openstack-infra | 18:23 | |
| *** notnownikki has joined #openstack-infra | 18:23 | |
| *** smarcet has joined #openstack-infra | 18:24 | |
| *** tonytan4ever has quit IRC | 18:24 | |
| *** gyee has quit IRC | 18:25 | |
| *** Hal1 has quit IRC | 18:26 | |
| anteaya | I would use it | 18:26 |
| *** tzn has joined #openstack-infra | 18:26 | |
| *** apuimedo has quit IRC | 18:26 | |
| * anteaya is afk for a bit | 18:27 | |
| *** apuimedo_ has quit IRC | 18:27 | |
| *** apuimedo has joined #openstack-infra | 18:27 | |
| *** unicell has quit IRC | 18:27 | |
| *** apuimedo_ has joined #openstack-infra | 18:28 | |
| *** ZZelle has quit IRC | 18:29 | |
| *** maishsk has quit IRC | 18:30 | |
| *** jpr has quit IRC | 18:30 | |
| *** woodster_ has joined #openstack-infra | 18:30 | |
| *** apuimedo has quit IRC | 18:32 | |
| *** apuimedo_ has quit IRC | 18:32 | |
| krotscheck | jeblair, mordred: There's already a mirror directory in this cell, shall I just add /mirror/npm? | 18:33 |
| *** apuimedo has joined #openstack-infra | 18:33 | |
| *** unicell has joined #openstack-infra | 18:34 | |
| *** apuimedo_ has joined #openstack-infra | 18:34 | |
| krotscheck | Oh, I can't. | 18:34 |
| krotscheck | Nevermind :) | 18:34 |
| *** unicell has quit IRC | 18:34 | |
| *** unicell1 has joined #openstack-infra | 18:34 | |
| *** e0ne has joined #openstack-infra | 18:34 | |
| *** otter768 has quit IRC | 18:35 | |
| *** yamamoto has joined #openstack-infra | 18:37 | |
| *** hurgleburgler has joined #openstack-infra | 18:38 | |
| openstackgerrit | Merged openstack-infra/project-config: Fixing app-catalog dead-link checker https://review.openstack.org/260115 | 18:40 |
| *** apuimedo has quit IRC | 18:40 | |
| *** apuimedo_ has quit IRC | 18:41 | |
| *** tmcpeak has joined #openstack-infra | 18:41 | |
| AJaeger | hurgleburgler: Regarding openstack-ux: You stated that openstack-ux is an official project, correct? | 18:41 |
| *** yamamoto has quit IRC | 18:42 | |
| *** apuimedo has joined #openstack-infra | 18:42 | |
| tmcpeak | AJaeger: regarding your comment on #260135 - this is a different kind of Bandit gate and projects may want to use both. This one is a baseline based on project changes, it runs a different script | 18:42 |
| jeblair | krotscheck: i'll set up mirror/npm and give you perms there | 18:42 |
| tmcpeak | so it's a different tox env name on purpose | 18:42 |
| *** apuimedo_ has joined #openstack-infra | 18:42 | |
| krotscheck | jeblair: Cool, I'm almost ready to run this thing. | 18:42 |
| mordred | krotscheck: woot | 18:42 |
| krotscheck | I'm going to use 'mirror.local' for now. | 18:43 |
| AJaeger | tmcpeak: so, why not use bandit-baseline? Those two names are unrelated... | 18:43 |
| tmcpeak | Ajaeger: we considered bandit baseline, but thought codesec sounded more intuitive in terms of a project that hasn't used Bandit before and will start with this | 18:43 |
| hurgleburgler | AJaeger It is an official project | 18:43 |
| AJaeger | hurgleburgler: I double checked governance repo and UX is indeed an official project, sorry for missing that before. So, you need to create a change for goverance repo which adds the nrew repo to the UX lists | 18:44 |
| *** jyuso1 has joined #openstack-infra | 18:44 | |
| AJaeger | And then your PTL has to +1 the changes. | 18:44 |
| *** ZZelle has joined #openstack-infra | 18:44 | |
| AJaeger | hurgleburgler: http://docs.openstack.org/infra/manual/creators.html#add-new-repository-to-the-governance-repository | 18:44 |
| hurgleburgler | AJaeger Gotcha, thanks for the direction! I'll do that now | 18:45 |
| AJaeger | Mkae the governance change Depends-On the project-config change and add as comment in project-config a link to the governance one | 18:45 |
| AJaeger | But that still leaves specs.openstack.org, do you really consider what you publish as specifications? | 18:45 |
| AJaeger | Otherwise you can publish on docs.openstack.org/developer/openstack-ux | 18:46 |
| AJaeger | tmcpeak: would that other project need two targets? codesec and bandit? Or only either of these? | 18:47 |
| tmcpeak | AJaeger: most likely both | 18:47 |
| tmcpeak | the original Bandit reports all issues, so it isn't as good for a gate but is still useful to have, the new one, codesec is what we're planning to use for a gate | 18:48 |
| *** maishsk has joined #openstack-infra | 18:48 | |
| AJaeger | tmcpeak: that's confusing for me and might be for others. But let's look at something else first: bandit is a linting target, isn't it? | 18:49 |
| *** jyuso1 has quit IRC | 18:50 | |
| tmcpeak | AJaeger: how do you mean that? | 18:50 |
| tmcpeak | -linting-target | 18:51 |
| AJaeger | bandit does code-analysis, doesn't it? | 18:51 |
| tmcpeak | yep | 18:51 |
| *** amuller_afk is now known as amuller | 18:51 | |
| AJaeger | found it - README has "security linter" ;) | 18:51 |
| AJaeger | why not add bandit target to the pep8 tox environment which is our usual "lint" environment? | 18:52 |
| tmcpeak | I'd like to have separate results for pep8 and Bandit | 18:52 |
| AJaeger | You can add any number of commands to it... | 18:52 |
| *** eharney has joined #openstack-infra | 18:52 | |
| *** rvasilets___ has joined #openstack-infra | 18:52 | |
| tmcpeak | separate reports, separate command line, everything | 18:52 |
| *** maishsk has quit IRC | 18:52 | |
| *** gyee has joined #openstack-infra | 18:53 | |
| AJaeger | tmcpeak: if both are voting, it's expensive. The runtimes are short but for each we start a new virtual machine... | 18:53 |
| tmcpeak | ohhh | 18:53 |
| tmcpeak | yeah, that is expensive | 18:53 |
| tmcpeak | is it possible to set both to run at the same time but keep output separate? | 18:53 |
| *** apuimedo has quit IRC | 18:53 | |
| AJaeger | not possible | 18:54 |
| *** apuimedo_ has quit IRC | 18:54 | |
| tmcpeak | AJaeger: hmm.. can I at least have them fail or pass independently? | 18:54 |
| *** apuimedo has joined #openstack-infra | 18:54 | |
| AJaeger | that gets tricky, let me check something... | 18:55 |
| tmcpeak | AJaeger: thank you | 18:55 |
| *** apuimedo_ has joined #openstack-infra | 18:55 | |
| notmorgan | mordred: +2 on your glance fix, but i think it means we need to get a larger integration test (betamax?) spun up for OCC | 18:56 |
| mordred | notmorgan: agree | 18:56 |
| mordred | notmorgan: because without testing against actual in-the-wild clouds, all of this is synthetic | 18:56 |
| notmorgan | exactly | 18:56 |
| notmorgan | the KSA betamax thing landed, not sure if we've released since then. | 18:57 |
| AJaeger | tmcpeak: I'm not seeing an obvious solution. Let's ask others. | 18:57 |
| tmcpeak | AJaeger: ok, thank you for checking | 18:57 |
| AJaeger | mordred, fungi, clarkb : What do you propose here? Add bandit targets everywhere? Or have bandit run as part of pep8? And can we somehow show easily what fails? | 18:58 |
| *** shardy is now known as shardy_afk | 18:58 | |
| mordred | AJaeger: it's a good question | 18:58 |
| *** piet has quit IRC | 18:58 | |
| *** tonytan4ever has joined #openstack-infra | 18:59 | |
| fungi | AJaeger: giving tox multiple commands works (exits nonzero if at least one command exits nonzero) but can be a pain to track down in the console log since it doesn't short-circuit | 18:59 |
| mordred | AJaeger: part of me wants to say "add logic to run-pep8.sh to detect if bandit is there and if so run it- but it makes it hard for a project to add a non-voting bandit, then make it voting" | 18:59 |
| notmorgan | fungi: ++ | 18:59 |
| *** piet has joined #openstack-infra | 18:59 | |
| notmorgan | from someone trying to debug out things such as bandit - i really don't like lumping more and more into the linting | 18:59 |
| fungi | i wonder if bandit could easily be run as a flake8 plugin/extension? | 18:59 |
| notmorgan | this isn't strictly linting | 19:00 |
| mordred | so I think honestly just add bandit targets everywhere | 19:00 |
| tmcpeak | notmorgan: that was kind of my feeling as well | 19:00 |
| fungi | yeah, the fact that it's not a linter makes the flake8 plugin path suboptimal | 19:00 |
| mordred | and we can deal with making it more efficient as part of zuul v3 / container execution possibilities | 19:00 |
| pleia2 | sick family member care this morning, aroundish now, but kind of in and out | 19:00 |
| AJaeger | fungi, do you mean tox -e command1,command2 - that fails currently AFAIR | 19:00 |
| notmorgan | AJaeger: it can work. | 19:00 |
| fungi | pleia2: go focus on important family things, there's nothing urgent here | 19:00 |
| mordred | but knowing that we _should_ run command1,command2 would be hard | 19:01 |
| pleia2 | fungi: very useful summary, thank you | 19:01 |
| AJaeger | fungi: tox 2 short-circuits now | 19:01 |
| fungi | oh? that's new, then | 19:01 |
| AJaeger | fungi: http://tox.readthedocs.org/en/latest/config.html#confval-ignore_errors=True|False%28default%29 | 19:01 |
| notmorgan | mordred: you can do it from within the tox.ini but ick. | 19:01 |
| *** pvaneck has joined #openstack-infra | 19:01 | |
| fungi | ahh, it's configurable. good to know | 19:01 |
| AJaeger | fungi: and the default changed from tox 1 to tox 2 ;( | 19:01 |
| fungi | oh, that's... interesting | 19:02 |
| *** angdraug has joined #openstack-infra | 19:02 | |
| fungi | but i guess it's a major version bump | 19:02 |
| notmorgan | i would still say a "security" target might be better. - we can move all security things into it including bandit and force tooling around making it easy to detect the failures once you move beyond bandit only? | 19:02 |
| fungi | so anything's game | 19:02 |
| notmorgan | i mean... it's more VMs. but it isn't a linter, and debugging things in multi commands is ick.. | 19:02 |
| notmorgan | unless you move bandit to being more of a linter - then it can be flake8 included and that'd be, in my mind the right time to put it into pep8 test | 19:03 |
| * notmorgan gets off soapbox | 19:03 | |
| mordred | yah. I think we should do what's 'right' | 19:03 |
| AJaeger | notmorgan: read http://git.openstack.org/cgit/openstack/bandit/tree/README.rst#n4 | 19:03 |
| fungi | so anyway, i've either not seen or quickly dismissed the arguments against having a separate bandit job | 19:03 |
| tmcpeak | what's the distinction of linter/non-linter here? | 19:03 |
| mordred | and deal with optimization more systemically | 19:03 |
| *** tonytan4ever has quit IRC | 19:03 | |
| AJaeger | "A security linter from OpenStack Security" | 19:03 |
| notmorgan | AJaeger: i am familiar with it | 19:04 |
| tmcpeak | that might not be correct, we just call it that | 19:04 |
| notmorgan | but it's not "really" a linter right now. | 19:04 |
| clarkb | I have long said using the existing linting target is simplest | 19:04 |
| notmorgan | it's somewhere between linter and other anaylsis | 19:04 |
| clarkb | because it exists wverywhere and people can opt in wirhout our review | 19:04 |
| notmorgan | it could easily be swing back towards something that easily works as a flake8 plugin | 19:05 |
| clarkb | and it is a linter as far as I know | 19:05 |
| notmorgan | which would be a nice bonus since each of the catch types could be included/excluded in a familiar way | 19:05 |
| *** weshay_pto has quit IRC | 19:05 | |
| notmorgan | if the bandit team is open to that | 19:05 |
| *** changbl has joined #openstack-infra | 19:05 | |
| openstackgerrit | Diana Whitten proposed openstack-infra/project-config: Add openstack-ux repo https://review.openstack.org/259239 | 19:05 |
| clarkb | as a dev keeping track of a bunch of special targets is not fun | 19:06 |
| clarkb | especially whe you push to jenkins after running tox and something new fails | 19:06 |
| notmorgan | which case linking to pep8 with that as the next step would be also acceptable | 19:06 |
| clarkb | because tox is misconfigured | 19:06 |
| notmorgan | with my dev hat on | 19:06 |
| fungi | [JARGON FILE] lint: A Unix C language processor which carries out more thorough checks on the code than is usual with C compilers. Lint is named after the bits of fluff it supposedly picks from programs. Judging by references on Usenet this term has become a shorthand for desk check at some non-Unix shops, even in languages other than C. Also used as delint. | 19:06 |
| fungi | vague enough i'd be willing to argue that there's no clear reason to say bandit is not a "linter" | 19:07 |
| AJaeger | So, should we take the christmas break and rename pep8 to lint and include bashate, bandit, etc in it? | 19:07 |
| *** maishsk has joined #openstack-infra | 19:07 | |
| AJaeger | (as tox environment I mean) | 19:07 |
| tmcpeak | so then will we get separate pass/fail or no? | 19:08 |
| tmcpeak | AJaeger: ^ | 19:08 |
| AJaeger | tmcpeak: no separate pass/fail, just a general: linting failed | 19:08 |
| notmorgan | tmcpeak: it wouldn't be a separate fail - it would be a "pep8" job fail, but you can know which linter failed | 19:09 |
| AJaeger | like today "pep8 failed" | 19:09 |
| notmorgan | if you looked at the log | 19:09 |
| tmcpeak | AJaeger: ok so the idea is that means something in the linters isn't happy, which could be code style, security, etc as opposed to "something in the unit tests failed" | 19:09 |
| AJaeger | tmcpeak: yes | 19:09 |
| tmcpeak | AJaeger: and then if somebody wanted to use the command line in their local dev environment, they also run them all at once? | 19:10 |
| * notmorgan wonders how hard it would be to make bandit checks also flake8 targets. | 19:10 | |
| fungi | i'm certainly in favor of a new and more generic tox target for a collection of linters, given how misnamed the "pep8" env is for many projects at this point | 19:10 |
| nibalizer | fungi: ++ | 19:10 |
| notmorgan | fungi: ++ | 19:10 |
| fungi | this also might make a good addition to the cross-projects spec for the standard testing interface | 19:11 |
| AJaeger | tmcpeak: you can have as many tox environments as you want - so one that includes all linters but also separate ones if you want... | 19:11 |
| *** apuimedo has quit IRC | 19:11 | |
| fungi | if we come up with a good new recommendation for something like that | 19:11 |
| nibalizer | https://wiki.openstack.org/wiki/Testing mentions 'style checks' | 19:11 |
| notmorgan | AJaeger: we can also do some limited templating in tox.ini to address multiples/overrides | 19:11 |
| *** apuimedo_ has quit IRC | 19:11 | |
| AJaeger | inframanual has "Running the style checks" and invokes pep8 | 19:11 |
| notmorgan | so if you want to limit/eliminate some linters it can be done via tox if we wanted to go down that road | 19:12 |
| *** apuimedo has joined #openstack-infra | 19:12 | |
| fungi | i'd almost argue for tox -e lint except, as evidenced by my jargon file quote above, we'd just be committing the same sin all over again (if perhaps at least giving credit to the first well-known tool name for this sort of thing?) | 19:12 |
| *** apuimedo_ has joined #openstack-infra | 19:13 | |
| AJaeger | fungi, "style"? | 19:13 |
| AJaeger | fungi, do you have a reference to the cross-spec? | 19:13 |
| notmorgan | fungi: -elinters would be fine even in light of the jargon | 19:13 |
| jeblair | krotscheck: you have full access on /afs/openstack.org/mirror/npm; that volume has a 100G quota; is that enough for now? | 19:14 |
| notmorgan | fungi: as long as we keep it to style/lint-style checks imo | 19:14 |
| AJaeger | notmorgan: the plural is nice to say it's more than one... | 19:15 |
| notmorgan | AJaeger: that was my goal :) | 19:15 |
| fungi | AJaeger: http://governance.openstack.org/reference/project-testing-interface.html i guess it predates the cross-project specs repo | 19:15 |
| *** _nadya_ has quit IRC | 19:16 | |
| jeblair | krotscheck: i also added read perms for system:anyuser to that volume | 19:16 |
| AJaeger | thanks, fungi | 19:16 |
| jeblair | krotscheck: we have a decidedly unofficial apache server that just serves afs, so you can probably use it for testing: http://afstest.openstack.org/mirror/ | 19:17 |
| *** jcoufal has quit IRC | 19:17 | |
| *** apuimedo has quit IRC | 19:17 | |
| *** rlandy has joined #openstack-infra | 19:17 | |
| jeblair | krotscheck: (or, of course, you can just run your own; there's nothing special about it) | 19:17 |
| *** tzn has quit IRC | 19:17 | |
| *** apuimedo_ has quit IRC | 19:17 | |
| AJaeger | so, next step would be a spec for openstack-specs for this? | 19:18 |
| jeblair | fungi, mordred: i am happy to report that all of the krb/afs commands i needed to do worked, so the cell does not seem to have bitrotted. | 19:18 |
| fungi | AJaeger: not necessarily. having at least one example implementation in the wild is probably first | 19:18 |
| fungi | jeblair: wow! i did actually not expect that | 19:18 |
| fungi | especially since we had some crashes in the past | 19:19 |
| *** apuimedo has joined #openstack-infra | 19:19 | |
| AJaeger | any guinea pigs around where we can do this? tmcpeak can we use bandit? ;) | 19:19 |
| tmcpeak | AJaeger: yep, volunteered | 19:19 |
| *** apuimedo_ has joined #openstack-infra | 19:19 | |
| AJaeger | tmcpeak: how do you like me to help with that? | 19:20 |
| tmcpeak | AJaeger: let me make sure I understand | 19:20 |
| fungi | cross-project specs are, as i understand (being not as involved in them as i should), an attempt to encode our community established practices where successful and promote wider adoption | 19:20 |
| tmcpeak | we're going to add bandit to the pep8 section with the idea that pep8 will be renamed in some way? | 19:20 |
| fungi | not so much a sounding board for "it would be nice if..." | 19:20 |
| AJaeger | tmcpeak: create a new environemnt called "linters" | 19:21 |
| *** reed has quit IRC | 19:21 | |
| fungi | we could in fact add bandit to a new [linters] section which runs the same things the current [pep8] env does plus bandit, and then add a job to run the linters env commands | 19:21 |
| fungi | yeah, what AJaeger said in fewer words | 19:21 |
| tmcpeak | fungi, AJaeger: got it | 19:21 |
| tmcpeak | I'll do the change and throw you guys on the review | 19:22 |
| fungi | at which point the pep8 env for whatever project that is can be deprecated and no longer run as a stand-alone job | 19:22 |
| tmcpeak | cool, sounds good | 19:22 |
| fungi | and then we can point to how well this is working on $project and wouldn't more projects like to take advantage of such splendor here's how | 19:22 |
| tmcpeak | fungi: perfect | 19:23 |
| AJaeger | thanks, tmcpeak, fungi, et al | 19:23 |
| tmcpeak | AJaeger, fungi, et al: thanks! | 19:23 |
| AJaeger | fungi, clarkb, do you have time for another discussion? what shall we do with the nova DocImpact change? It's currently completely broken. We can a) fix it - https://review.openstack.org/259842 - b) revert change https://review.openstack.org/#/c/259569/ c) leave scripts in, disable check for now (259842+new change to disable it). What's your opinon? | 19:24 |
| AJaeger | tmcpeak: add me to any change and I'll happily review - and help push this through in project-config | 19:25 |
| tmcpeak | AJaeger: sounds great, thank you | 19:25 |
| *** reed has joined #openstack-infra | 19:25 | |
| *** apuimedo_ has quit IRC | 19:26 | |
| *** apuimedo_ has joined #openstack-infra | 19:28 | |
| openstackgerrit | yolanda.robla proposed openstack/diskimage-builder: Add dib element to generate logical volumes https://review.openstack.org/252041 | 19:29 |
| *** doug-fish has quit IRC | 19:29 | |
| *** apuimedo has quit IRC | 19:30 | |
| *** moravec1 has quit IRC | 19:31 | |
| *** apuimedo has joined #openstack-infra | 19:31 | |
| *** apuimedo_ has quit IRC | 19:33 | |
| *** apuimedo_ has joined #openstack-infra | 19:33 | |
| *** angdraug has quit IRC | 19:33 | |
| krotscheck | jeblair: Should be enough, thanks | 19:34 |
| *** angdraug has joined #openstack-infra | 19:34 | |
| krotscheck | jeblair: That decidedly unoffical one is refusing my connections | 19:34 |
| *** moravec has joined #openstack-infra | 19:34 | |
| fungi | AJaeger: i'll defer to nova leadership on whether to revert it or patch it up while fleshing out the better reno-oriented solution | 19:35 |
| *** ZZelle has quit IRC | 19:36 | |
| fungi | AJaeger: i don't really have a preference as long as nova and docs teams agree there is some benefit to the current job | 19:36 |
| AJaeger | for that we first have to merge 259842 - let me approve that one and if johnthetubaguy speaks up, we can still disable/revert. Ok? | 19:36 |
| AJaeger | johnthetubaguy: are you around for docimpact discussion? Otherwise feel free to ping me tomorrow | 19:37 |
| *** apuimedo_ has quit IRC | 19:38 | |
| notmorgan | fungi: do we support OS X for python-*client libs/CLIs? [in any reasonable way] (cc clarkb, mordred, pleia2, anteaya) - as in "best effort" or just "nope not supporteD" | 19:39 |
| notmorgan | fungi, clarkb, mordred, and i need to ask how we handle patches proposed on LP and not through gerrit [it appears someone posted a patch to LP and someone else converted to gerrit] | 19:39 |
| *** apuimedo_ has joined #openstack-infra | 19:39 | |
| * notmorgan would love to be able to disable patch posting on lp. | 19:40 | |
| clarkb | I do not provide any os x support | 19:41 |
| openstackgerrit | Ramy Asselin proposed openstack-infra/elastic-recheck: Include query results in graph https://review.openstack.org/260188 | 19:42 |
| *** apuimedo has quit IRC | 19:43 | |
| *** notnownikki has quit IRC | 19:44 | |
| *** apuimedo_ has quit IRC | 19:44 | |
| *** markvoelker has joined #openstack-infra | 19:45 | |
| *** apuimedo has joined #openstack-infra | 19:45 | |
| fungi | notmorgan: considering our "if it's not tested it's broken" slogan and the corollary "we don't support what's broken" i have to assume the official answer is "no" | 19:45 |
| *** apuimedo_ has joined #openstack-infra | 19:45 | |
| openstackgerrit | Andreas Jaeger proposed openstack-infra/project-config: Replace bashate/pep8 in project-config by linters https://review.openstack.org/260192 | 19:46 |
| AJaeger | tmcpeak, fungi, clarkb: Let's lead by example in project-config ^ | 19:46 |
| fungi | notmorgan: as for handling patches posted to lp/github/notgerrit, someone else can forward the patch in but they are most likely violating the icla in so doing | 19:46 |
| notmorgan | fungi: ok i'll comment on the bug. | 19:47 |
| *** reed has quit IRC | 19:47 | |
| fungi | that's not to say that i object personally, but their employer's lawyers might be concerned about it | 19:47 |
| notmorgan | fungi: yeah i am also not super worried cause the fix is low | 19:48 |
| crinkle | mordred: is https://review.openstack.org/#/c/198417/ still on your radar? it would be cool for our bifrost server to be able to use that | 19:48 |
| notmorgan | innovation - it's just a "fix what opts are passed to socket" on OS X | 19:48 |
| fungi | the vmt sometimes reproposes into gerrit patches attached to embargoed vulnerability reports in lp, but that's generally patches proposed by existing contributors who were temporarily avoiding gerrit so they didn't prematurely end an embargo | 19:49 |
| tmcpeak | AJaeger: this is awesome | 19:49 |
| openstackgerrit | Philipp Marek proposed openstack-infra/ciwatch: Performance improvement for loading. https://review.openstack.org/260193 | 19:49 |
| tmcpeak | AJaeger: I was trying to grok the necessary magics to do the change but you're fast :) | 19:49 |
| *** apuimedo has quit IRC | 19:49 | |
| AJaeger | tmcpeak: once my patch is in, you can use gate-{name}-linters for it ;) | 19:50 |
| tmcpeak | sweet | 19:50 |
| AJaeger | tmcpeak: I wanted more than one guinea pig ;) | 19:50 |
| *** apuimedo_ has quit IRC | 19:50 | |
| tmcpeak | AJaeger: awesome, thanks man! | 19:50 |
| *** markvoelker has quit IRC | 19:50 | |
| *** apuimedo has joined #openstack-infra | 19:50 | |
| *** apuimedo_ has joined #openstack-infra | 19:51 | |
| *** angdraug has quit IRC | 19:52 | |
| *** lucasagomes is now known as lucas-dinner | 19:53 | |
| *** esker has joined #openstack-infra | 19:53 | |
| *** reed has joined #openstack-infra | 19:54 | |
| *** esker has quit IRC | 19:54 | |
| *** tzn has joined #openstack-infra | 19:54 | |
| *** esker has joined #openstack-infra | 19:54 | |
| openstackgerrit | Andreas Jaeger proposed openstack-infra/project-config: Replace bashate/pep8 in project-config by linters https://review.openstack.org/260192 | 19:54 |
| anteaya | notmorgan: I also do not provide any os x support | 19:55 |
| AJaeger | fungi, thanks for review - unfortunately the jenkins tests failed and I had to fix the test tool as well | 19:55 |
| AJaeger | see updated patch ^ | 19:55 |
| *** apuimedo has quit IRC | 19:55 | |
| *** lezbar has joined #openstack-infra | 19:56 | |
| *** apuimedo has joined #openstack-infra | 19:56 | |
| notmorgan | anteaya: commented on the bug saying "yeah you can land this fix, but it'll break again" | 19:56 |
| openstackgerrit | Merged openstack-infra/project-config: Turn off -e in check-git-commit-message.sh https://review.openstack.org/259842 | 19:56 |
| anteaya | pleia2: hope family members feel better soon | 19:56 |
| fungi | AJaeger: oh, yep! does that need to be replicated in any checks in the jjb repo too? | 19:56 |
| mordred | crinkle: yes. I think that would be lovely | 19:57 |
| AJaeger | yeah, I guess I should add description everywhere- let me split the change after checking jjb | 19:57 |
| anteaya | notmorgan: ha ha ha | 19:57 |
| jeblair | mordred, fungi, krotscheck: if the nodepool ready script is basically just 'host $NODEPOOL_PROVIDER_THINGIE.openstack.org >>/etc/hosts' there's basically no downside from using dns; we'll get timely address updates in general. | 19:57 |
| jeblair | krotscheck: oh, i think i need to disable puppet on it | 19:57 |
| fungi | jeblair: also, less flaky than additional dns lookups? (local nscd aside) | 19:58 |
| mordred | jeblair: ++ | 19:58 |
| notmorgan | ugh nscd :( boo ... | 19:58 |
| pleia2 | anteaya: thanks :) | 19:58 |
| *** ZZelle has joined #openstack-infra | 19:58 | |
| * notmorgan was hurt in the past by nscd :P | 19:58 | |
| *** lezbar__ has quit IRC | 19:58 | |
| fungi | and you still carry the scars, obviously | 19:58 |
| notmorgan | fungi: exactly | 19:58 |
| notmorgan | fungi: :P | 19:59 |
| jeblair | krotscheck: afstest fixed | 19:59 |
| TheJulia | Question, on using the file publishing, where does source setting expect the files to be located on the filesystem? | 20:00 |
| *** lezbar has quit IRC | 20:00 | |
| *** reed has quit IRC | 20:01 | |
| jroll | TheJulia: the root of the project the job is on, e.g. https://github.com/openstack-infra/project-config/blob/master/jenkins/jobs/ironic-python-agent-jobs.yaml#L31 | 20:02 |
| *** hashar has joined #openstack-infra | 20:03 | |
| *** reed has joined #openstack-infra | 20:03 | |
| TheJulia | jroll: as in $WORKSPACE? | 20:03 |
| fungi | TheJulia: yep | 20:04 |
| TheJulia | Ok, thanks! | 20:04 |
| *** ZZelle has quit IRC | 20:04 | |
| *** dcentral has joined #openstack-infra | 20:04 | |
| fungi | TheJulia: if you're using the ftp publisher there are options to strip/munge some of the prepended path within the workspace. if using the scp publisher which lacks those options, there are workarounds involving copying/renaming file paths around and i can dig up examples if you need | 20:05 |
| openstackgerrit | Ramy Asselin proposed openstack-infra/elastic-recheck: Include query results in graph https://review.openstack.org/260188 | 20:06 |
| TheJulia | fungi: I'll just make it like jroll's IPA job, I just need to change the folder things are in so the publisher finds the files. | 20:06 |
| *** irenab has joined #openstack-infra | 20:06 | |
| fungi | should be easy then | 20:06 |
| *** irenab has quit IRC | 20:06 | |
| TheJulia | yup | 20:06 |
| openstackgerrit | Andreas Jaeger proposed openstack-infra/project-config: Replace bashate/pep8 in project-config by linters https://review.openstack.org/260192 | 20:07 |
| openstackgerrit | Andreas Jaeger proposed openstack-infra/project-config: Add description label to jenkins checker https://review.openstack.org/260202 | 20:07 |
| AJaeger | fungi, updated and split into two changes ^ | 20:08 |
| openstackgerrit | greghaynes proposed openstack-infra/nodepool: Make secure.conf optional and backwards-compat https://review.openstack.org/232800 | 20:08 |
| openstackgerrit | Julia Kreger proposed openstack-infra/project-config: Build a Debian simple-init images https://review.openstack.org/254892 | 20:08 |
| *** irenab has joined #openstack-infra | 20:09 | |
| *** moravec has quit IRC | 20:10 | |
| *** apuimedo has quit IRC | 20:10 | |
| fungi | AJaeger: waiting to make sure that passes since not actually sure what jjb will do with the description on a template | 20:10 |
| *** ZZelle has joined #openstack-infra | 20:11 | |
| *** irenab has quit IRC | 20:11 | |
| *** apuimedo has joined #openstack-infra | 20:11 | |
| AJaeger | fungi, just found a bug... | 20:11 |
| tmcpeak | AJaeger: happy you're taking this on instead of me,.. 'description': v.All(str), would have been WAYYY over my head | 20:12 |
| AJaeger | fungi, see tests/yamlparser/fixtures/add_managed_to_desc.yaml in jenkins-job-builder | 20:12 |
| AJaeger | tmcpeak: pabelanger_ showed me that and I extendet it ;) | 20:12 |
| openstackgerrit | Andreas Jaeger proposed openstack-infra/project-config: Add description label to jenkins checker https://review.openstack.org/260202 | 20:12 |
| openstackgerrit | Andreas Jaeger proposed openstack-infra/project-config: Replace bashate/pep8 in project-config by linters https://review.openstack.org/260192 | 20:12 |
| tmcpeak | ahhh | 20:13 |
| *** apuimedo_ has quit IRC | 20:14 | |
| *** dkranz has quit IRC | 20:15 | |
| *** rkukura has quit IRC | 20:16 | |
| openstackgerrit | Andreas Jaeger proposed openstack-infra/project-config: Replace bashate/pep8 in project-config by linters https://review.openstack.org/260192 | 20:16 |
| *** Swami has quit IRC | 20:16 | |
| *** apuimedo_ has joined #openstack-infra | 20:16 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/shade: Add option to enable HTTP tracing https://review.openstack.org/260134 | 20:17 |
| openstackgerrit | Matthew Treinish proposed openstack/requirements: Bump minimum tempest-lib version https://review.openstack.org/259266 | 20:18 |
| *** apuimedo has quit IRC | 20:18 | |
| openstackgerrit | John Trowbridge proposed openstack-infra/tripleo-ci: DO NOT MERGE changes needed for delorean promotion https://review.openstack.org/258140 | 20:19 |
| *** apuimedo has joined #openstack-infra | 20:19 | |
| *** irenab has joined #openstack-infra | 20:20 | |
| *** _nadya_ has joined #openstack-infra | 20:21 | |
| krotscheck | jeblair: Can I trouble you to put DirectoryIndex index.html index.json into the aftest mirror? | 20:23 |
| *** dkranz has joined #openstack-infra | 20:23 | |
| *** _nadya_ has quit IRC | 20:25 | |
| *** e0ne has quit IRC | 20:25 | |
| *** janonymous_ has quit IRC | 20:26 | |
| *** apuimedo_ has quit IRC | 20:26 | |
| *** apuimedo has quit IRC | 20:26 | |
| *** apuimedo has joined #openstack-infra | 20:27 | |
| *** apuimedo_ has joined #openstack-infra | 20:27 | |
| *** rkukura has joined #openstack-infra | 20:28 | |
| *** moravec has joined #openstack-infra | 20:29 | |
| onovy | hi, someone for final +A for 256671 and 256668? thanks | 20:29 |
| *** e0ne has joined #openstack-infra | 20:29 | |
| onovy | and 256755 - three same patches | 20:30 |
| *** irenab has quit IRC | 20:31 | |
| jeblair | krotscheck: done | 20:31 |
| *** irenab has joined #openstack-infra | 20:31 | |
| *** apuimedo_ has quit IRC | 20:32 | |
| *** irenab has quit IRC | 20:32 | |
| *** apuimedo_ has joined #openstack-infra | 20:33 | |
| *** e0ne has quit IRC | 20:34 | |
| *** irenab has joined #openstack-infra | 20:35 | |
| *** otter768 has joined #openstack-infra | 20:36 | |
| openstackgerrit | Alex Schultz proposed openstack-infra/project-config: Add fuel-plugin-etckeeper repository for Fuel https://review.openstack.org/260208 | 20:36 |
| openstackgerrit | Andreas Jaeger proposed openstack-infra/project-config: Replace bashate/pep8 in project-config by linters https://review.openstack.org/260192 | 20:36 |
| *** irenab has quit IRC | 20:36 | |
| tmcpeak | AJaeger: making you work for it, huh? ;) | 20:36 |
| AJaeger | tmcpeak: yeah, interesting bugs I have to fix first... | 20:36 |
| *** apuimedo_ has quit IRC | 20:37 | |
| *** _nadya_ has joined #openstack-infra | 20:37 | |
| *** e0ne has joined #openstack-infra | 20:38 | |
| *** apuimedo has quit IRC | 20:38 | |
| *** apuimedo_ has joined #openstack-infra | 20:38 | |
| *** apuimedo has joined #openstack-infra | 20:39 | |
| fungi | hrm, almost ready with the changes to strip the branch-dependent worker parameter functions, but double-checking the jobs which are being explicitly overridden to precise is time-consuming | 20:40 |
| *** baoli has quit IRC | 20:40 | |
| mordred | fungi: woot | 20:40 |
| *** otter768 has quit IRC | 20:40 | |
| *** mwhahaha has joined #openstack-infra | 20:40 | |
| AJaeger | great, fungi! | 20:41 |
| *** esker has quit IRC | 20:42 | |
| *** apuimedo_ has quit IRC | 20:43 | |
| *** irenab has joined #openstack-infra | 20:43 | |
| *** esker has joined #openstack-infra | 20:43 | |
| *** apuimedo_ has joined #openstack-infra | 20:43 | |
| *** apuimedo has quit IRC | 20:44 | |
| fungi | just down to checking into the chef jobs which have icehouse/juno branches pinned to precise because of berkshelf/gecode dependencies | 20:45 |
| *** e0ne has quit IRC | 20:45 | |
| *** mrmartin has quit IRC | 20:45 | |
| *** apuimedo has joined #openstack-infra | 20:45 | |
| *** _nadya_ has quit IRC | 20:46 | |
| *** Hal2 has quit IRC | 20:46 | |
| *** apoorvad has quit IRC | 20:46 | |
| fungi | also this thing has probably already been merge-conflicted in the hour i've been writing it | 20:47 |
| *** ZZelle has quit IRC | 20:48 | |
| *** irenab has quit IRC | 20:48 | |
| *** apuimedo_ has quit IRC | 20:48 | |
| *** apoorvad has joined #openstack-infra | 20:49 | |
| AJaeger | 260202 and 260192 passed the tests now - the linters change for project-config. Could I get some reviews, please? Then tmcpeak can base the bandit changes on top of these... | 20:49 |
| tmcpeak | woohoo | 20:49 |
| AJaeger | fungi, I doubt we merged anything in the last hour... | 20:49 |
| *** irenab has joined #openstack-infra | 20:49 | |
| *** e0ne has joined #openstack-infra | 20:50 | |
| AJaeger | fungi: you're lucky - last merge was 54 mins ago to a script, and the one before 4h ago, so you shouldn't have any conflicts... | 20:50 |
| *** apuimedo_ has joined #openstack-infra | 20:50 | |
| fungi | the one i'm stuck on now is the gate-openstack-chef-repo job | 20:51 |
| krotscheck | jeblair, mordred: So, with a hosts entry pointing mirror.local to afstest, this works http://paste.openstack.org/show/482447/ | 20:51 |
| fungi | oh, i see | 20:51 |
| *** hichihara has joined #openstack-infra | 20:51 | |
| mordred | krotscheck: sweet! and it only downloaded things from mirror.local? | 20:51 |
| krotscheck | yep | 20:52 |
| jeblair | krotscheck: i notice that http://afstest.openstack.org/mirror/npm/ looks like it might be dir-per-package... is that the case? | 20:52 |
| *** ZZelle has joined #openstack-infra | 20:52 | |
| krotscheck | i had to mirror all those deps before it worked | 20:52 |
| krotscheck | jeblair: i believe so | 20:53 |
| *** irenab has quit IRC | 20:53 | |
| jeblair | krotscheck: how many packages we talkin about? | 20:53 |
| krotscheck | 218,275 | 20:53 |
| krotscheck | as of right now | 20:53 |
| tmcpeak | AJaeger: question - how is python-jobs going to change? | 20:53 |
| tmcpeak | AJaeger: we're currently just using the python-jobs template in Bandit | 20:54 |
| jeblair | krotscheck: in our local mirror scripty thingy, can we diretory-hash those? so they end up in eg "mirror/npm/j/ja/jason/" | 20:54 |
| AJaeger | tmcpeak: for now: Not at all. In the future: First step the jenkins template will get it... | 20:54 |
| *** irenab has joined #openstack-infra | 20:54 | |
| AJaeger | tmcpeak: just add 'gate-{name}-linters' to bandit in jenkins/jobs/projects.yaml | 20:54 |
| AJaeger | See the two places I changed it for project-config - in projects.yaml and in zuul/layout.yaml | 20:55 |
| jeblair | krotscheck: afs has a directory limit of (well it's complicated but for discussion let's say) 64k entries. | 20:55 |
| krotscheck | jeblair: not if we want it to work, i think | 20:55 |
| *** bardia has joined #openstack-infra | 20:55 | |
| krotscheck | It's faking a CouchDB instance after all | 20:56 |
| jeblair | krotscheck: aren't we writing the url ourselves anyway? | 20:56 |
| tmcpeak | AJaeger: sure, but if I just add it, won't it run twice now since "python-jobs" is already calling pep8 too? | 20:56 |
| krotscheck | jeblair: We are rewriting the index.json file, yes, but the path of {mirror}/{packagename}/index.json still has to resolve. | 20:57 |
| tmcpeak | for project-config you guys weren't using that python-jobs template (which includes a bunch of other stuff) | 20:57 |
| *** irenab has quit IRC | 20:57 | |
| AJaeger | tmcpeak: indeed, you have to change that, don't use python-jobs template in zuul/layout.yaml but the single jobs. | 20:57 |
| tmcpeak | in Bandit we use 'python-jobs' to run pep8, py27 and py3 unit tests, etc | 20:57 |
| tmcpeak | AJaeger: ok | 20:57 |
| tmcpeak | AJaeger: won't those then run on separate VM's too? | 20:58 |
| AJaeger | tmcpeak: they already all run on separate VMs | 20:58 |
| tmcpeak | AJaeger: ahh ok | 20:58 |
| jeblair | krotscheck, mordred: this seems to be a disease common to language package mirror authors. | 20:58 |
| AJaeger | python-jobs is just syntactic sugar to not have to write these everytime | 20:58 |
| jeblair | krotscheck: we may be able to solve that with a rewrite rule | 20:58 |
| tmcpeak | it seems like we'd need to split python-jobs to "with linter target" and "without linter target" otherwise projects won't be able to use it anymore | 20:58 |
| AJaeger | tmcpeak: yeah, that's the next step. Let me put one change on top of my latest for this... | 20:59 |
| jhesketh | Morning | 20:59 |
| *** irenab has joined #openstack-infra | 20:59 | |
| krotscheck | jeblair: I think in this case it's the npm authors, not the mirror authors. | 20:59 |
| tmcpeak | AJaeger: ok cool, is that the way you see it splitting out? two different "python-jobs" - one running with legacy pep8 and one running with linter target? | 20:59 |
| *** moravec has quit IRC | 20:59 | |
| mordred | krotscheck, jeblair: we're already rewriting urls when we write out index.json in the tool that the yahoo folks maintain | 21:00 |
| krotscheck | jeblair: I think the npm mirror daemon has hooks we can add, let me check. | 21:00 |
| mordred | we could add a feature to add in {firstletter}/{packagename} | 21:00 |
| mordred | and they might land it quicker than the bandersnatch folks | 21:00 |
| jeblair | mordred: if i'm understanding krotscheck it's the initial lookup of "mirror/packagename" that needs to be solved, yeah? | 21:00 |
| openstackgerrit | Jeremy Stanley proposed openstack-infra/project-config: Get rid of precise||trusty job node declarations https://review.openstack.org/260213 | 21:01 |
| openstackgerrit | Jeremy Stanley proposed openstack-infra/project-config: Remove precise/trusty label mapping in Zuul config https://review.openstack.org/260214 | 21:01 |
| fungi | AJaeger: clarkb: mordred: ^ | 21:01 |
| mordred | oh - does it? | 21:01 |
| krotscheck | Yep, and that's an npm thing, not a registry-static thing. | 21:01 |
| mordred | jeblair: in that case, yeah, we'll need rewrite rule there too | 21:01 |
| jeblair | mordred: like when you say "npm install foo" it does "GET /mirror/foo" | 21:01 |
| mordred | yah | 21:01 |
| jeblair | krotscheck: i have a set of rewrite rules (they are actually pretty simple) that will let us serve "mirror/npm/foo' from 'mirror/f/o/foo' | 21:01 |
| krotscheck | As long as we can get the mirroring script to build that structure, yes? | 21:02 |
| jeblair | krotscheck: yep | 21:02 |
| krotscheck | Alright, lemme look for hooks | 21:02 |
| openstackgerrit | Andreas Jaeger proposed openstack-infra/project-config: Add new python-jobs-linters template https://review.openstack.org/260215 | 21:02 |
| jeblair | krotscheck: cool, if you can build that structure, i'll put the rewrite rules in place on afstest | 21:02 |
| *** vilobhmm11 has joined #openstack-infra | 21:02 | |
| *** irenab has quit IRC | 21:03 | |
| *** dprince has quit IRC | 21:03 | |
| jeblair | krotscheck: (one letter package names should just hash to one dir level deep; 2 and greater to 2 levels deep) | 21:03 |
| *** moravec has joined #openstack-infra | 21:03 | |
| *** dizquierdo has joined #openstack-infra | 21:04 | |
| tmcpeak | AJaeger: perfect! | 21:05 |
| *** irenab has joined #openstack-infra | 21:05 | |
| *** irenab has quit IRC | 21:06 | |
| *** migi has quit IRC | 21:06 | |
| *** apuimedo has quit IRC | 21:08 | |
| *** migi has joined #openstack-infra | 21:08 | |
| *** irenab has joined #openstack-infra | 21:08 | |
| *** ociuhandu has quit IRC | 21:09 | |
| AJaeger | fungi, what a change, woow | 21:09 |
| *** apuimedo has joined #openstack-infra | 21:09 | |
| fungi | AJaeger: it's ambitious... i don't expect it to pass on the first attempt | 21:09 |
| *** weshay_xchat has joined #openstack-infra | 21:09 | |
| openstackgerrit | greghaynes proposed openstack-infra/nodepool: Decouple nodepool db and config from builders https://review.openstack.org/222398 | 21:10 |
| fungi | and even once it does, i expect there will be at least some nuance we miss in review which causes us some post-facto correcting | 21:10 |
| *** vilobhmm111 has joined #openstack-infra | 21:10 | |
| fungi | but it needs to get done, and the sooner the better | 21:10 |
| fungi | we could have done this a cycle ago | 21:10 |
| openstackgerrit | greghaynes proposed openstack-infra/nodepool: Builders distinguish between failure and exception https://review.openstack.org/242131 | 21:10 |
| *** vilobhmm11 has quit IRC | 21:10 | |
| openstackgerrit | greghaynes proposed openstack-infra/nodepool: Decouple builders from nodepool instance https://review.openstack.org/232135 | 21:11 |
| ianw | greghaynes / SpamapS : so ... do you think i could get F23 minimal for christmas? there's 3 dib changes out there, pretty small, but with that i get a booting f23 in rax with glean | 21:11 |
| openstackgerrit | greghaynes proposed openstack-infra/nodepool: Builders can be run on their own https://review.openstack.org/232251 | 21:11 |
| SpamapS | ianw: I was literally just opening up gertty so I'll prioritize dib :) | 21:11 |
| fungi | AJaeger: also it will collect merge conficts so rapidly that we probably can't merge it unless we prioritize it over other changes | 21:11 |
| greghaynes | ianw: haha, I'll have a look | 21:12 |
| *** Swami has joined #openstack-infra | 21:12 | |
| * AJaeger votes for prioritizing it | 21:12 | |
| *** gchamoul has quit IRC | 21:12 | |
| ianw | SpamapS: thanks; https://review.openstack.org/#/c/257172/ (the eth0/eth1 thing) is maybe the mostly interesting one | 21:12 |
| *** irenab has quit IRC | 21:13 | |
| fungi | AJaeger: i'm around for the next few days, so happy to be the rooter to run jjb manually everywhere and also to make whatever adjustments we find are lacking afterward | 21:13 |
| *** changbl has quit IRC | 21:14 | |
| SpamapS | ianw: it's confusing to me that glean considers a DHCP-configured interface as something it shouldn't touch, since last I checked every Ubuntu cloud image ships with eth0 setup for DHCP | 21:14 |
| fungi | AJaeger: hrm, those changes succeeded their jobs on the first try. i'm suspicious--that almost never happens to me | 21:14 |
| greghaynes | ianw: re the eth0/1 thing - doesn't infra still run f21? | 21:14 |
| greghaynes | ianw: I am specifically worried that we will break them | 21:14 |
| *** irenab has joined #openstack-infra | 21:14 | |
| SpamapS | Guessing nobody has tried the combination of Ubuntu-not-minimal, configdrive-only cloud, and glean | 21:14 |
| *** irenab has quit IRC | 21:14 | |
| ianw | SpamapS: it's in the systemd .service file -- it checks if the eth0 script exists, and if it does, it does not run glean | 21:14 |
| greghaynes | SpamapS: I have by accident a few times | 21:14 |
| ianw | greghaynes: that's why i want f23 :) to kill off f21 in infra :) | 21:15 |
| SpamapS | Why would glean do that? | 21:15 |
| greghaynes | ianw: Yes, but we probably should not break them in doing so... | 21:15 |
| greghaynes | ianw: hrm, actually, I think if glean sees no config drive it will write out dhcp config | 21:15 |
| ianw | greghaynes: it's totally broken at the moment, i can't run devstack with missing packgaes | 21:15 |
| greghaynes | ianw: so maybe we wont | 21:15 |
| *** gchamoul has joined #openstack-infra | 21:16 | |
| ianw | SpamapS: https://github.com/openstack-infra/glean/blob/master/glean/init/glean%40.service#L5 | 21:16 |
| *** dizquierdo has quit IRC | 21:16 | |
| AJaeger | fungi, your change looks fine at first sight, I'm willing to take the risk. | 21:16 |
| greghaynes | SpamapS: I think we want to change that glean behavior FWIW | 21:16 |
| ianw | SpamapS: glean triggers this on udev net rules | 21:16 |
| fungi | AJaeger: now we just need a second brave soul to agree with you ;) | 21:16 |
| AJaeger | fungi, if we want to go through with it, we should not merge anything before your change goes in... | 21:16 |
| greghaynes | SpamapS: probably by making it do the cloud-init thing where it writes out the last applied config to disk and if it sees the same config on disk as in config drive it bails | 21:17 |
| fungi | i would like clarkb to give 260213 and 260214 a once-over since i recall he did a lot of the legwork on the original transition to run jobs on trusty after the icehouse release | 21:17 |
| fungi | and i'm basically attempting to unravel that | 21:18 |
| AJaeger | yes, clarkb reviewing these would be great. | 21:18 |
| *** irenab has joined #openstack-infra | 21:18 | |
| SpamapS | greghaynes: Ah yes I've been suggesting that for a while. OK. | 21:18 |
| *** claudiub has quit IRC | 21:18 | |
| jeblair | i'm going to go back into my zuulv3 hole for a while | 21:19 |
| *** irenab has quit IRC | 21:19 | |
| *** shardy_afk is now known as shardy | 21:19 | |
| anteaya | jeblair: hope it is cozy | 21:19 |
| AJaeger | fungi, if there's any fallout tomorrow morning European time, I can keep an eye on it - and if yolanda is around, so can she... | 21:19 |
| fungi | sounds cozy. especially if there are cookies there | 21:19 |
| fungi | AJaeger: thanks | 21:19 |
| AJaeger | cookies and coffee? Or something stronger? jeblair enjoy! | 21:19 |
| jeblair | fungi: there definitely are: http://www.hanescookies.com/hanes_cs/default.asp | 21:20 |
| AJaeger | fungi, if you have time to review my linters series starting at 26020, it would be nice - but let's get your changes in first! | 21:20 |
| jeblair | AJaeger: ^ :) | 21:20 |
| fungi | jeblair: from north carolina! | 21:21 |
| AJaeger | jeblair: my current favorites: http://www.lebkuchen-mirus.de/ ;) Sorry, German only | 21:21 |
| jeblair | fungi: my hometown even :) | 21:21 |
| * anteaya looks at 260213 and puts the kettle on | 21:21 | |
| fungi | jeblair: even better | 21:21 |
| AJaeger | jeblair: enjoy! | 21:22 |
| AJaeger | thanks, anteaya ! Make it a strong tea/coffee ;) | 21:22 |
| anteaya | heh | 21:22 |
| *** marekd has joined #openstack-infra | 21:23 | |
| anteaya | the cookies look delicious | 21:23 |
| *** apuimedo has quit IRC | 21:23 | |
| krotscheck | AJaeger: Oh, yum. | 21:23 |
| * krotscheck is baking his own stollen this year | 21:23 | |
| krotscheck | I had to candy my own lemon peel. | 21:23 |
| AJaeger | krotscheck: yummy! | 21:23 |
| fungi | the console log of the xml diff for the job change is >13mb | 21:24 |
| AJaeger | fungi: new record ;) | 21:24 |
| anteaya | krotscheck: that sounds like fun | 21:24 |
| anteaya | ha ha ha | 21:24 |
| *** apuimedo has joined #openstack-infra | 21:25 | |
| *** kgiusti has left #openstack-infra | 21:26 | |
| anteaya | fungi: so we don't _have_ to get rid of quotes they were just a casualty of your script? https://review.openstack.org/#/c/260213/1/jenkins/jobs/astara.yaml | 21:26 |
| *** bnemec has quit IRC | 21:26 | |
| *** irenab has joined #openstack-infra | 21:26 | |
| fungi | anteaya: yeah, it was accompanying cleanup. easier to normalize the quoting on them as i went | 21:26 |
| *** irenab has quit IRC | 21:27 | |
| AJaeger | anteaya: those quotes are optional, I'm happy to see them gone | 21:27 |
| anteaya | fungi: makes sense | 21:27 |
| fungi | the reason most of them were there is because someone copied them from somewhere without realizing they were unnecessary | 21:27 |
| anteaya | AJaeger: fair enough, just trying to understand the changes | 21:27 |
| AJaeger | sure, anteaya | 21:27 |
| anteaya | fungi: fair enough | 21:28 |
| fungi | anteaya: mostly it made checking them after the fact slightly easier | 21:28 |
| anteaya | fungi: yup | 21:28 |
| anteaya | so next question, this file: https://review.openstack.org/#/c/260213/1/jenkins/jobs/chef-jobs.yaml | 21:28 |
| anteaya | why did bare-precise win vs bare-trusty here? | 21:28 |
| *** bnemec has joined #openstack-infra | 21:28 | |
| fungi | anteaya: those were being explicitly set to bare-precise in the zuul configuration | 21:29 |
| fungi | you'll find some groups and openstackid jobs with a similar fate | 21:29 |
| AJaeger | anteaya: see https://review.openstack.org/#/c/260214/1/zuul/layout.yaml | 21:30 |
| bknudson_ | there used to be a button in gerrit to edit the commit message | 21:30 |
| AJaeger | anteaya: line 698 | 21:30 |
| AJaeger | bknudson_: it's still there - but now for every file | 21:30 |
| AJaeger | Review the commit message and you have the button | 21:30 |
| fungi | bknudson_: now you can edit anything, not just commit messages | 21:30 |
| anteaya | fungi: thank you, I will continue to look for that | 21:30 |
| bknudson_ | ok, I found the edit button on the main page | 21:30 |
| *** markvoelker has joined #openstack-infra | 21:30 | |
| anteaya | AJaeger: ah thanks | 21:31 |
| *** irenab has joined #openstack-infra | 21:31 | |
| *** irenab has quit IRC | 21:31 | |
| AJaeger | I have to leave now, looking forward to the outcome of this! Bye... | 21:32 |
| anteaya | doe and fawn in my garden | 21:32 |
| anteaya | AJaeger: night | 21:32 |
| *** AJaeger has quit IRC | 21:32 | |
| *** dcentral has quit IRC | 21:32 | |
| *** dims_ has quit IRC | 21:33 | |
| *** dims has joined #openstack-infra | 21:33 | |
| *** irenab has joined #openstack-infra | 21:35 | |
| *** irenab has quit IRC | 21:35 | |
| *** 18VAACSN1 has quit IRC | 21:35 | |
| *** michaelxin has quit IRC | 21:35 | |
| *** lindsayk has quit IRC | 21:35 | |
| *** markvoelker has quit IRC | 21:36 | |
| tonyb | Can anyone see where this https://review.openstack.org/#/c/259830/ change is? The last thing form jenkins was starting jobs check zuul but I can't see it in zuul (I may be blind) | 21:36 |
| tonyb | is there somewhere else I can check? | 21:36 |
| *** apuimedo_ has quit IRC | 21:37 | |
| *** ZZelle has quit IRC | 21:38 | |
| anteaya | you've lost your change? | 21:38 |
| *** apuimedo_ has joined #openstack-infra | 21:39 | |
| anteaya | I don't see it in status.openstack.org/zuul | 21:39 |
| openstackgerrit | Ian Wienand proposed openstack/diskimage-builder: yum-minimal: do not configure eth0 & eth1 for DHCP automatically https://review.openstack.org/257172 | 21:39 |
| tonyb | anteaya: yeah and I miss it ;P | 21:39 |
| *** angdraug has joined #openstack-infra | 21:39 | |
| ianw | SpamapS: ^ yeah, you're right, using "-n" there isn't a good idea because you can't unset easily | 21:39 |
| ianw | changed it to a 0/1 match | 21:39 |
| anteaya | tonyb: :( | 21:39 |
| *** irenab has joined #openstack-infra | 21:39 | |
| anteaya | jenkins said it was starting gate jobs about 12 hours ago and then it didn't finish | 21:40 |
| *** irenab has quit IRC | 21:40 | |
| anteaya | tonyb: I'll recheck it to get a +1 verified on it again | 21:40 |
| anteaya | tonyb: if it passes recheck find me and I'll reapprove | 21:41 |
| anteaya | tonyb: I don't know what happened | 21:41 |
| tonyb | anteaya: okay thanks. | 21:41 |
| *** apuimedo has quit IRC | 21:41 | |
| *** ZZelle has joined #openstack-infra | 21:41 | |
| *** apuimedo has joined #openstack-infra | 21:41 | |
| anteaya | np | 21:42 |
| *** vgridnev has quit IRC | 21:42 | |
| fungi | 2015-12-21 09:33:46,632 DEBUG zuul.DependentPipelineManager: Reported change <Change 0x7f2ea9e0b410 259830,1> failed tests or failed to merge | 21:43 |
| *** dims has quit IRC | 21:43 | |
| *** zeih has joined #openstack-infra | 21:43 | |
| fungi | aha, i have a traceback, just a sec | 21:43 |
| tmcpeak | fungi: the implicit 0x in front of it makes it look more intimidating :P | 21:44 |
| *** nikhil_k has joined #openstack-infra | 21:44 | |
| tmcpeak | hex scares me, I just like representations that magically fall into the 0-F range | 21:44 |
| fungi | http://paste.openstack.org/show/482454/ | 21:45 |
| fungi | no idea what's up with the KeyError: 'LOG_PATH' there, but i'm working from the kitchen and semi-distracted at the moment | 21:46 |
| fungi | though it looks like it was going to fail anyway... | 21:47 |
| fungi | 2015-12-21 09:33:39,056 DEBUG zuul.DependentPipelineManager: <QueueItem 0x7f2da26af6d0 for <Change 0x7f2ea9e0b410 259830,1> in gate> is a failing item because ['it has a merge conflict', 'at least one job failed'] | 21:47 |
| openstackgerrit | Ian Wienand proposed openstack/diskimage-builder: yum-minimal : install selinux policy packages https://review.openstack.org/259867 | 21:48 |
| tonyb | fungi: Thanks. | 21:48 |
| *** nikhil has quit IRC | 21:48 | |
| anteaya | fungi: thanks | 21:48 |
| anteaya | tonyb: do you want to rebase? | 21:48 |
| tonyb | fungi: I assume if your're in the kitchen you're cooking so get back to that :D | 21:48 |
| tonyb | anteaya: yup doign it now. | 21:49 |
| *** irenab has joined #openstack-infra | 21:49 | |
| *** dims has joined #openstack-infra | 21:49 | |
| jhesketh | fungi: that looks like a zuul bug.. I'll take a dig into it | 21:50 |
| anteaya | jhesketh: thanks | 21:50 |
| openstackgerrit | Ramy Asselin proposed openstack-infra/elastic-recheck: Include query results in graph https://review.openstack.org/260188 | 21:51 |
| anteaya | fungi: when you are finished cooking/dining, the chef-<jobs>.yaml file were all on precise since they don't work on trusty as we discussed above, shouldn't the cookbook repos also be on precise then? https://review.openstack.org/#/c/260213/1/jenkins/jobs/projects.yaml starting at line 662 | 21:52 |
| openstackgerrit | Tony Breeds proposed openstack-infra/project-config: Barbican does not support python34 in kilo https://review.openstack.org/259830 | 21:53 |
| tonyb | anteaya: it was a merge conflict with another of my changes. | 21:54 |
| *** apuimedo_ has quit IRC | 21:55 | |
| *** apuimedo has quit IRC | 21:55 | |
| *** apuimedo_ has joined #openstack-infra | 21:56 | |
| *** apuimedo has joined #openstack-infra | 21:56 | |
| anteaya | you | 21:56 |
| fungi | anteaya: they have other jobs besides those in that same job group which use bare-trusty instead, so it was easiest to just change the precise-based ones | 21:56 |
| anteaya | glad you figured it out | 21:56 |
| anteaya | fungi: okey dokey | 21:57 |
| fungi | also makes it a little more obvious what's going to break when we get rid of precise soonish | 21:57 |
| openstackgerrit | Merged openstack-infra/elastic-recheck: Add query for bug 1527483 https://review.openstack.org/260119 | 21:57 |
| * anteaya refortifies before heading into the wilderness of fuel | 21:57 | |
| openstack | bug 1527483 in neutron "VPNaaS - No providers specified for 'VPN' service" [Critical,Confirmed] https://launchpad.net/bugs/1527483 - Assigned to Martin Hickey (martin-hickey) | 21:57 |
| anteaya | fungi: yep | 21:57 |
| fungi | anteaya: tonyb: i think i've seen that keyerror for log_path on reporting in the past when we've got a custom report string which is trying to use that parameter for a job where it doesn't exist/isn't defined | 21:58 |
| *** thorst_ has quit IRC | 21:59 | |
| anteaya | ah ha, thank you | 22:01 |
| tmcpeak | AJaeger, fungi: you guys fighting other fires or is it a decent time to merge that 'linters' change? | 22:02 |
| tmcpeak | oh, AJaeger is gone :) | 22:03 |
| *** apuimedo has quit IRC | 22:03 | |
| *** apuimedo has joined #openstack-infra | 22:05 | |
| *** apuimedo_ has quit IRC | 22:05 | |
| *** eil397 has joined #openstack-infra | 22:05 | |
| *** zeih has quit IRC | 22:05 | |
| *** apuimedo_ has joined #openstack-infra | 22:06 | |
| *** dims_ has joined #openstack-infra | 22:08 | |
| *** dims has quit IRC | 22:08 | |
| *** apuimedo has quit IRC | 22:09 | |
| *** piet has quit IRC | 22:10 | |
| *** apuimedo has joined #openstack-infra | 22:11 | |
| docaedo | where are requirements for jenkins tests declared? I had assumed requestsexceptions would be in the jenkins slave environment but guess not: http://logs.openstack.org/98/260198/3/experimental/propose-app-catalog-update/c037fc8/console.html | 22:12 |
| *** shardy has quit IRC | 22:12 | |
| *** dims has joined #openstack-infra | 22:12 | |
| mordred | docaedo: for that slave, it's all managed in puppet | 22:13 |
| *** lindsayk has joined #openstack-infra | 22:13 | |
| *** dims_ has quit IRC | 22:13 | |
| mordred | docaedo: openstack-infra/system-config:modules/openstack_project/manifests/proposal_slave.pp | 22:13 |
| zaro | I've got exim working, sends emails locally but can't get it to work with external connection. anybody know how to setup exim to accept external connections? | 22:14 |
| *** dcentral has joined #openstack-infra | 22:14 | |
| docaedo | mordred: ah wow ok, thanks :) | 22:14 |
| *** apuimedo has quit IRC | 22:15 | |
| *** apuimedo has joined #openstack-infra | 22:16 | |
| *** apuimedo_ has quit IRC | 22:16 | |
| *** apuimedo_ has joined #openstack-infra | 22:16 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/system-config: Add requestsexceptions to proposal slave https://review.openstack.org/260230 | 22:17 |
| mordred | docaedo: ^^ | 22:17 |
| *** dkranz has quit IRC | 22:17 | |
| docaedo | mordred: yikes fast! I was about to do that, thank you so much! | 22:18 |
| mordred | docaedo: :) | 22:18 |
| mordred | docaedo: I type clickyclicky | 22:18 |
| openstackgerrit | Ian Wienand proposed openstack-infra/glean: Fixup invalid hostnames https://review.openstack.org/259357 | 22:19 |
| docaedo | mordred: no kidding! | 22:19 |
| *** apuimedo has quit IRC | 22:20 | |
| *** sdake has quit IRC | 22:21 | |
| anteaya | mordred: do you say clickyclicky while you are typing? | 22:21 |
| anteaya | please say yes, please say yes | 22:21 |
| *** apuimedo has joined #openstack-infra | 22:22 | |
| *** apuimedo_ has quit IRC | 22:23 | |
| *** apuimedo_ has joined #openstack-infra | 22:24 | |
| pleia2 | hehe | 22:24 |
| bkero | Haha | 22:24 |
| bkero | mordred: Do you do fancy keyboard clickyclicky? | 22:25 |
| *** ZZelle has quit IRC | 22:26 | |
| mordred | anteaya: yes | 22:27 |
| mordred | bkero: always | 22:27 |
| pleia2 | this is why we all have to work from home | 22:28 |
| bkero | mordred: I'm still waiting for a laptop with a clicky keyboard :< | 22:28 |
| *** apuimedo_ has quit IRC | 22:28 | |
| *** ZZelle has joined #openstack-infra | 22:28 | |
| mordred | bkero: you are in good company in this channel then :) | 22:28 |
| bkero | Haha | 22:29 |
| *** dims has quit IRC | 22:29 | |
| *** apuimedo_ has joined #openstack-infra | 22:29 | |
| bkero | :) | 22:29 |
| bkero | mordred: It already exists, but we should stop complaining and just use it despite flaws. http://e5c351ecddc2f880ef72-57d6ff1fc59ab172ec418789d348b0c1.r69.cf1.rackcdn.com/images/Z1jgrRQVIWfS.878x0.Z-Z96KYq.jpg | 22:29 |
| *** mriedem is now known as mriedem_away | 22:30 | |
| *** apoorvad has quit IRC | 22:30 | |
| *** zeih has joined #openstack-infra | 22:30 | |
| mordred | bkero: :) | 22:31 |
| bkero | I'd love to see you try to use that in an airplane coach seat | 22:31 |
| anteaya | mordred: oh good | 22:31 |
| anteaya | fungi: I'm fine with https://review.openstack.org/#/c/260213/1 and like the idea of you self-approving when you are ready | 22:31 |
| fungi | anteaya: okay, if clarkb is around enough to at least +1 the basic concept in that pair of patches i'm happy to press forward and babysit it into the evening | 22:32 |
| anteaya | fungi: understood | 22:33 |
| anteaya | fungi: and you had said you wanted clarkb to take a peek so I figured I'd wait and you are best to press the go button anyway | 22:33 |
| clarkb | fungi anteaya sorry got movie day today | 22:34 |
| anteaya | clarkb: oh yes that's right, STAR WARS! | 22:35 |
| *** e0ne has quit IRC | 22:35 | |
| anteaya | clarkb: have fun | 22:35 |
| anteaya | fungi: their is a sentence in the commit message on 260214 I don't parse | 22:36 |
| anteaya | there* | 22:36 |
| *** apuimedo has quit IRC | 22:36 | |
| anteaya | "and let it trusty that any worker registering the job name" | 22:36 |
| *** otter768 has joined #openstack-infra | 22:36 | |
| anteaya | it lost me | 22:36 |
| *** apuimedo has joined #openstack-infra | 22:37 | |
| *** admin0 has joined #openstack-infra | 22:37 | |
| openstackgerrit | Merged openstack-infra/project-config: Barbican does not support python34 in kilo https://review.openstack.org/259830 | 22:39 |
| jhesketh | tonyb, fungi: hmm, so it worked this time around ^ | 22:39 |
| jhesketh | I'm also confused because I can't see any jobs on project-config that use the LOG_PATH in their success_pattern | 22:39 |
| tonyb | jhesketh: yeah there was race/merge conflict the first time | 22:39 |
| tonyb | jhesketh: nothing specific in the change | 22:39 |
| jhesketh | I'll add some hardening so that next time it happens we'll have better logging | 22:39 |
| jhesketh | hmm okay, must have missed that part of the conversation | 22:40 |
| *** otter768 has quit IRC | 22:41 | |
| jhesketh | not sure why a merge conflict would cause the LOG_PATH to not exist though | 22:41 |
| fungi | i don't know that it was a race/merge conflict | 22:42 |
| *** dimtruck is now known as zz_dimtruck | 22:42 | |
| fungi | the message said it was being dequeued *either* because of a failed job or a merge conflict | 22:42 |
| fungi | i didn't dig back far enough in the debug log to find out which | 22:42 |
| anteaya | okay so this is a big file: https://review.openstack.org/#/c/260213/1/jenkins/jobs/projects.yaml | 22:43 |
| anteaya | I used to be able to control-f for the page | 22:43 |
| anteaya | now control-f is taken by either the left diff or the right | 22:43 |
| anteaya | and ends up in some js controlled search | 22:43 |
| anteaya | which stalls because the file is very big | 22:43 |
| anteaya | so I can't search that file | 22:43 |
| jhesketh | hmm, so the dequeue event caused a report, but formatting the report failed | 22:43 |
| anteaya | which makes me sad | 22:43 |
| fungi | anteaya: hah at typo in the commit message. s/trusty/trust/ there. i'll fix ;) | 22:43 |
| jhesketh | I wonder if we're missing all of the merge conflict reports? | 22:44 |
| anteaya | fungi: was hoping that was it | 22:44 |
| anteaya | fungi: my brain is a little jelly from 213 | 22:44 |
| fungi | anteaya: as for your ctrl-f woes, you can checkout the change locally and just `git show jenkins/jobs/projects.yaml` and use / and the n key to search in ne | 22:45 |
| anteaya | true I can do that | 22:45 |
| fungi | also gertty has search now | 22:45 |
| anteaya | I'm thinking of giving gertty another shot after I get christmas over with | 22:45 |
| anteaya | currently doing battle with my cat over who has say where my wrist goes | 22:46 |
| openstackgerrit | Jeremy Stanley proposed openstack-infra/project-config: Remove precise/trusty label mapping in Zuul config https://review.openstack.org/260214 | 22:46 |
| *** markvoelker has joined #openstack-infra | 22:46 | |
| fungi | commit message typo addressed ^ | 22:46 |
| anteaya | at least she has given me my keyboard back | 22:47 |
| anteaya | fungi: thank you | 22:47 |
| *** lykinsbd has quit IRC | 22:47 | |
| fungi | thanks for reviewing it | 22:48 |
| *** moravec has quit IRC | 22:48 | |
| anteaya | fungi: if I read the comments on removed in this file correctly: https://review.openstack.org/#/c/260214/1/zuul/layout.yaml the infra puppet repos should be running on precise nodes, yes? | 22:48 |
| anteaya | fungi: welcome, thanks for doing the work | 22:49 |
| *** tzn has quit IRC | 22:49 | |
| fungi | anteaya: the system-config one | 22:49 |
| fungi | gate-system-config-puppet.* | 22:49 |
| anteaya | oh okay thanks | 22:50 |
| * anteaya looks again | 22:50 | |
| fungi | which i couldn't find. i should look again just to be sure | 22:50 |
| fungi | ahh, gate-system-config-puppet-syntax may need to get forced to bare-precise | 22:50 |
| fungi | checking now | 22:50 |
| anteaya | that looks like a job that should be defined in zuul/layout.yaml | 22:51 |
| *** pvaneck has quit IRC | 22:51 | |
| *** lykinsbd has joined #openstack-infra | 22:51 | |
| *** markvoelker has quit IRC | 22:51 | |
| fungi | though i'm also okay with starting to run that one on bare-trusty instead, i agree that should probably be a separate change | 22:51 |
| anteaya | okay as long as you have a plan | 22:53 |
| anteaya | I'm fine with whatever it is | 22:53 |
| fungi | huh, i'm not having a lot of luck finding where that job is actually defined | 22:54 |
| nibalizer | maybe in puppet.yaml? | 22:54 |
| nibalizer | we (infra) have some but not perfect overlap with the puppet modules team | 22:54 |
| fungi | looks like it's in the puppet-check-jobs job group, though we pass in a node: override there to bare-trusty and i haven't changed that | 22:55 |
| fungi | so sort of strange that it was working before this | 22:55 |
| anteaya | fungi: it is possible it is an old comment from when all the puppet modules were in system-config | 22:55 |
| *** smarcet has quit IRC | 22:55 | |
| anteaya | fungi: and I have posted a comment on ps1 of 214 | 22:55 |
| fungi | thanks | 22:56 |
| fungi | i'm still pondering this | 22:56 |
| anteaya | welcome | 22:56 |
| anteaya | yup | 22:56 |
| *** irenab has joined #openstack-infra | 22:57 | |
| *** irenab has quit IRC | 22:57 | |
| *** apoorvad has joined #openstack-infra | 22:58 | |
| *** ajmiller has quit IRC | 22:58 | |
| *** ajmiller_ has joined #openstack-infra | 22:58 | |
| fungi | yeah, if you look in jenkins/jobs/projects.yaml you'll see we add the puppet-check-jobs group to the system-config project | 22:58 |
| fungi | and we pass a node: bare-trusty parameter to that group | 22:58 |
| fungi | that job-group includes the gate-{name}-puppet-syntax template | 22:59 |
| fungi | (in the jenkins/jobs/puppet-module-jobs.yaml file) | 22:59 |
| anteaya | I see that | 22:59 |
| anteaya | I think perhaps that comment is an artifact | 23:00 |
| fungi | and that job-template has a node: '{node}' parameterization | 23:00 |
| anteaya | of configurations gone by | 23:00 |
| fungi | well, except that we're matching ^gate-system-config-puppet.*$ and calling the set_node_options_default_precise() function on that | 23:01 |
| fungi | which should be switching it to bare-precise from zuul's perspective, i think | 23:01 |
| *** Guest17831 is now known as ruagair | 23:01 | |
| *** ruagair is now known as Guest10064 | 23:02 | |
| fungi | i mean, there's a simple way to see what that's doing. can look at the layout job | 23:02 |
| *** apuimedo_ has quit IRC | 23:02 | |
| fungi | on the first change and then the second | 23:02 |
| *** apuimedo_ has joined #openstack-infra | 23:02 | |
| fungi | or look at a recent job and see where it ran | 23:03 |
| *** irenab has joined #openstack-infra | 23:03 | |
| *** mtanino has quit IRC | 23:03 | |
| fungi | https://jenkins01.openstack.org/job/gate-system-config-puppet-syntax/261/ | 23:04 |
| fungi | bare-trusty | 23:04 |
| fungi | so yeah, whatever that was supposed to be matching, it never worked | 23:04 |
| *** irenab has quit IRC | 23:04 | |
| *** apuimedo has quit IRC | 23:05 | |
| anteaya | woooo | 23:05 |
| *** amuller has quit IRC | 23:05 | |
| *** apuimedo has joined #openstack-infra | 23:05 | |
| *** hashar has quit IRC | 23:05 | |
| *** woodster_ has quit IRC | 23:06 | |
| *** irenab has joined #openstack-infra | 23:07 | |
| fungi | anteaya: and yeah, good catch on the dev param functions file | 23:07 |
| fungi | i deleted one line too many in it | 23:08 |
| anteaya | oh good | 23:08 |
| * anteaya was helpful today! | 23:08 | |
| *** irenab has quit IRC | 23:09 | |
| *** apuimedo_ has quit IRC | 23:09 | |
| *** moravec has joined #openstack-infra | 23:09 | |
| *** apuimedo has quit IRC | 23:10 | |
| openstackgerrit | Jeremy Stanley proposed openstack-infra/project-config: Remove precise/trusty label mapping in Zuul config https://review.openstack.org/260214 | 23:10 |
| fungi | anteaya: ^ thanks! | 23:10 |
| *** apuimedo has joined #openstack-infra | 23:11 | |
| anteaya | welcome | 23:11 |
| *** irenab has joined #openstack-infra | 23:12 | |
| *** apuimedo_ has joined #openstack-infra | 23:12 | |
| *** irenab has quit IRC | 23:13 | |
| *** darvon has quit IRC | 23:15 | |
| *** apuimedo has quit IRC | 23:15 | |
| *** apuimedo_ has quit IRC | 23:16 | |
| *** apuimedo has joined #openstack-infra | 23:16 | |
| *** irenab has joined #openstack-infra | 23:16 | |
| *** apuimedo_ has joined #openstack-infra | 23:17 | |
| *** irenab has quit IRC | 23:17 | |
| mordred | infra-root: "I have the upgrade for the CPUs scheduled for tomorrow (12.22.15) at 5pm PST." | 23:18 |
| mordred | infra-root: that's re: bluebox cloud | 23:18 |
| nibalizer | for bluebox? | 23:18 |
| *** darvon has joined #openstack-infra | 23:18 | |
| nibalizer | awesome | 23:18 |
| mordred | yah | 23:18 |
| fungi | yay! | 23:18 |
| anteaya | mordred: nice | 23:18 |
| pleia2 | cool | 23:18 |
| nibalizer | i hope the people doing the upgrade get to take the rest of the week off | 23:18 |
| mordred | so we should probably bleed it down to 0 if it's not already and/or quiesce the mirror? | 23:18 |
| anteaya | I think it is still 0 | 23:19 |
| fungi | i think we didn't bring it back up above 1 max yet | 23:20 |
| anteaya | sorry no, it was 1 on Friday | 23:20 |
| anteaya | yeah | 23:20 |
| fungi | so pretty easy to do (or ignore) before then | 23:20 |
| anteaya | http://git.openstack.org/cgit/openstack-infra/project-config/tree/nodepool/nodepool.yaml#n925 | 23:20 |
| anteaya | do we want it at 0 or -1? | 23:21 |
| *** apuimedo has quit IRC | 23:21 | |
| *** dims has joined #openstack-infra | 23:21 | |
| *** apuimedo_ has quit IRC | 23:21 | |
| *** smarcet has joined #openstack-infra | 23:22 | |
| *** apuimedo has joined #openstack-infra | 23:22 | |
| *** irenab has joined #openstack-infra | 23:22 | |
| *** apuimedo_ has joined #openstack-infra | 23:22 | |
| *** irenab has quit IRC | 23:23 | |
| *** tzn has joined #openstack-infra | 23:24 | |
| *** baoli has joined #openstack-infra | 23:25 | |
| *** smarcet has quit IRC | 23:26 | |
| *** asselin_ has joined #openstack-infra | 23:26 | |
| *** apuimedo has quit IRC | 23:26 | |
| openstackgerrit | Merged openstack/diskimage-builder: Fix unmount/remove race in cleanup_build_dir https://review.openstack.org/259586 | 23:27 |
| *** tmcpeak has left #openstack-infra | 23:27 | |
| *** apuimedo has joined #openstack-infra | 23:27 | |
| *** asselin__ has quit IRC | 23:27 | |
| openstackgerrit | Joshua Hesketh proposed openstack-infra/zuul: Add extra hardening on success-pattern url https://review.openstack.org/260243 | 23:28 |
| *** baoli has quit IRC | 23:28 | |
| jhesketh | fungi: ^ that will make things a little better | 23:28 |
| fungi | anteaya: ideally 0 if they're taking it offline for a while | 23:29 |
| fungi | jhesketh: thanks! | 23:29 |
| anteaya | fungi: okay I'll offer a patch | 23:30 |
| *** tzn has quit IRC | 23:30 | |
| *** dims has quit IRC | 23:30 | |
| *** sabeen has joined #openstack-infra | 23:32 | |
| *** krtaylor has quit IRC | 23:32 | |
| fungi | sounds good | 23:32 |
| *** dims has joined #openstack-infra | 23:32 | |
| *** irenab has joined #openstack-infra | 23:33 | |
| *** ZZelle has quit IRC | 23:33 | |
| openstackgerrit | Anita Kuno proposed openstack-infra/project-config: Turn bluebox servers off in nodepool. https://review.openstack.org/260247 | 23:35 |
| *** ZZelle has joined #openstack-infra | 23:36 | |
| boris-42 | fungi: hi | 23:36 |
| openstackgerrit | Merged openstack/requirements: Bump minimum tempest-lib version https://review.openstack.org/259266 | 23:36 |
| boris-42 | fungi: so recently https://review.openstack.org/#/c/252590/ one of jobs was removed from Rally project | 23:36 |
| anteaya | mordred: 260247 | 23:36 |
| *** apuimedo_ has quit IRC | 23:36 | |
| boris-42 | fungi: unfortunatelly nobody notify rally team about this... | 23:36 |
| boris-42 | fungi: however is it possible to add it back? or at least add centos7 instead? | 23:37 |
| *** apuimedo has quit IRC | 23:37 | |
| *** otter768 has joined #openstack-infra | 23:37 | |
| anteaya | boris-42: there were many patches like that: https://review.openstack.org/#/q/topic:juno-eol | 23:37 |
| *** apuimedo has joined #openstack-infra | 23:38 | |
| anteaya | boris-42: I tried to get a hold of you, you weren't in channel, it would have been a courtesy ping anyway, I wasn't going to try to find you on the mailing list | 23:38 |
| boris-42 | anteaya: it would be nice if at least one of core members or PTL of project will be notified | 23:38 |
| anteaya | I tried | 23:38 |
| anteaya | you weren't in channel as I said | 23:38 |
| *** apuimedo_ has joined #openstack-infra | 23:38 | |
| anteaya | and I wasn't going to try to find you on the mailing list | 23:38 |
| anteaya | as the job was going to be removed | 23:38 |
| boris-42 | anteaya: ok | 23:39 |
| boris-42 | anteaya: in any case is there centos7 job? | 23:39 |
| fungi | boris-42: yeah, i didn't know that one was done without the knowledge of the rally team, though there was quite a bit of discussion on the -dev ml about how we were dropping centos 6.x testing once we no longer needed it to support python 2.6 | 23:39 |
| boris-42 | fungi: sure no worries | 23:39 |
| boris-42 | fungi: so is there now centos 7? | 23:39 |
| fungi | readding a centos6 job worker is not going to happen at this point, but if centos 7 will work for you then that sounds like a fine alternative. we have them | 23:39 |
| boris-42 | fungi: great so same name just centos7? | 23:40 |
| fungi | we have devstack-centos7 nodes (don't let the name fool you, they don't have devstack running, they just avoid installing extra packages so that frameworks like devstack can declare what they want to install instead) | 23:41 |
| *** gordc has quit IRC | 23:42 | |
| openstackgerrit | Boris Pavlovic proposed openstack-infra/project-config: Add experimental job to check Rally installation on Centos 7 https://review.openstack.org/260258 | 23:43 |
| *** sputnik1_ has joined #openstack-infra | 23:43 | |
| boris-42 | fungi: so it's not bare-centos7? | 23:43 |
| boris-42 | fungi: so it will be better to Rally team to have just really not modifed centos image | 23:44 |
| openstackgerrit | Diana Whitten proposed openstack-infra/project-config: Add openstack-ux repo https://review.openstack.org/259239 | 23:44 |
| *** krtaylor has joined #openstack-infra | 23:44 | |
| *** apuimedo has quit IRC | 23:45 | |
| *** sdake has joined #openstack-infra | 23:45 | |
| *** apuimedo_ has quit IRC | 23:45 | |
| fungi | boris-42: yes, the bare-.* node labels are also misleading. they're not actually "bare" but rather have a lot of packages preinstalled while the devstack-.* nodes preinstall as little as possible | 23:45 |
| *** zz_dimtruck is now known as dimtruck | 23:46 | |
| boris-42 | fungi: so i should deffintelly use devstack- one? | 23:46 |
| *** apuimedo has joined #openstack-infra | 23:46 | |
| boris-42 | definitely* | 23:46 |
| fungi | we're working toward dropping all the bare-.* types and running jobs exclusively on the devstack-.* types (though we're planning to rename the latter to just indicate the distro-release and drop the terms "bare" and "devstack") | 23:46 |
| fungi | yes, try it on devstack-centos7 for now | 23:46 |
| *** apuimedo_ has joined #openstack-infra | 23:46 | |
| openstackgerrit | Boris Pavlovic proposed openstack-infra/project-config: Add experimental job to check Rally installation on Centos 7 https://review.openstack.org/260258 | 23:46 |
| boris-42 | fungi: ^ fixed | 23:46 |
| fungi | boris-42: i think you missed restoring the lines which were dropped from jenkins/jobs/projects.yaml in the change you're sort of reverting | 23:48 |
| *** piet has joined #openstack-infra | 23:50 | |
| *** apuimedo has quit IRC | 23:50 | |
| *** baoli has joined #openstack-infra | 23:50 | |
| *** baoli has quit IRC | 23:51 | |
| *** apuimedo_ has quit IRC | 23:51 | |
| *** apuimedo has joined #openstack-infra | 23:52 | |
| *** ZZelle has quit IRC | 23:52 | |
| *** apuimedo_ has joined #openstack-infra | 23:52 | |
| *** gildub has joined #openstack-infra | 23:54 | |
| zaro | yo! anybody good with setting up smtp mail server? | 23:55 |
| fungi | zaro: pretty good. spent much of my life as a mailserver admin in at least some capacity, and still run some smtp servers even today | 23:56 |
| *** ildikov has quit IRC | 23:56 | |
| *** apuimedo has quit IRC | 23:56 | |
| fungi | what specifically are you trying to do? send messages out? receive and deliver messages locally? receive messages and forward them back out? | 23:56 |
| *** otter768 has quit IRC | 23:57 | |
| *** apuimedo_ has quit IRC | 23:57 | |
| *** irenab has joined #openstack-infra | 23:57 | |
| *** changbl has joined #openstack-infra | 23:57 | |
| *** apuimedo has joined #openstack-infra | 23:57 | |
| *** dimtruck is now known as zz_dimtruck | 23:57 | |
| zaro | fungi: i setup exim as outgoing mail server in a VM. got it to work from localhost. but can't seem to get it to list from external IPs | 23:57 |
| zaro | *list/listen | 23:58 |
| *** apuimedo_ has joined #openstack-infra | 23:58 | |
| fungi | aha, so you want it to receive messages from the internet? | 23:58 |
| *** irenab has quit IRC | 23:58 | |
| *** Guest10064 is now known as TongMaster | 23:58 | |
| fungi | zaro: in the typical ubuntu/debian packages of exim4 there should be a /etc/exim4/update-exim4.conf.conf file | 23:59 |
| zaro | fungi: basically i want to use it for testing gerrit emails. want to setup gerrit to connect to i | 23:59 |
| zaro | i/it | 23:59 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
