| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul-jobs master: upload-logs: fix indentation https://review.openstack.org/500339 | 01:10 |
|---|---|---|
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul-jobs master: upload-logs/bindep: fix indentation https://review.openstack.org/500339 | 01:12 |
| dmsimard | mordred, Shrews, clarkb, jeblair: Still looking for reviews on a linters job for devstack-gate, it'd allow us to get rid of the zuul v2 bashate job too. | 01:54 |
| dmsimard | https://review.openstack.org/#/c/498140/ | 01:54 |
| pabelanger | dmsimard: left comments | 02:29 |
| dmsimard | pabelanger: cool, will address tomorrow, trying to wrap up this disgusting ara patch before taking a break :D | 02:30 |
| dmsimard | 15 files changed, 803 insertions(+), 1547 deletions(-) | 02:59 |
| dmsimard | all unit tests | 02:59 |
| * dmsimard throws up | 02:59 | |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/zuul feature/zuulv3: WIP: web: add jobs controller https://review.openstack.org/466561 | 03:01 |
| *** pabelanger has quit IRC | 04:14 | |
| *** pabelanger has joined #zuul | 04:16 | |
| jeblair | i just saw an ansible-playbook segfault; afaict we're still running the patched python, so there may be another bug lurking in there. | 05:48 |
| *** maxamillion has quit IRC | 06:09 | |
| *** maxamillion has joined #zuul | 06:09 | |
| *** lennyb has quit IRC | 06:56 | |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool feature/zuulv3: Implement a static driver for Nodepool https://review.openstack.org/468624 | 10:32 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool feature/zuulv3: Implement an OpenContainer driver https://review.openstack.org/468753 | 10:32 |
| openstackgerrit | Tristan Cacqueray proposed openstack-infra/nodepool feature/zuulv3: oci: replace paramiko with ansible playbooks https://review.openstack.org/500357 | 10:32 |
| mordred | jeblair: awesome | 12:29 |
| *** yolanda has quit IRC | 13:17 | |
| *** lennyb has joined #zuul | 14:17 | |
| dmsimard | mordred: so, somewhat of a silly question | 14:42 |
| dmsimard | mordred: with in-repo zuul.yaml files, what prevents me from knowingly or unknowingly DDoS'ing Zuul/Nodepool with junk jobs ? | 14:43 |
| mordred | dmsimard: social pressure :) | 14:43 |
| mordred | dmsimard: I mean, you can already DDOS the system just by submitting 10k patches to the devstack-gate repo | 14:44 |
| dmsimard | mordred: I guess, zuul.yaml lowers the barrier to entry that's all :p | 14:45 |
| dmsimard | could make one 50 node job or something | 14:45 |
| mordred | yah- I believe we've tlaked about adding a control to limit the number of nodes a job can request | 14:45 |
| mordred | yah | 14:45 |
| dmsimard | I see at least three vectors that should have some amount of safeguards around: concurrency (rate limiting), size (amount of nodes/job), length (timeout). These should probably be enforced by default and then probably configurable/whitelistable at the trusted/system level | 14:48 |
| dmsimard | Otherwise the noisy neighbor issue is real in a multi-tenancy/multi-project environment | 14:49 |
| * dmsimard has managed a multi-tenant public cloud before T_T | 14:50 | |
| tristanC | mordred: max-nodes-per-job is actually live :-) | 15:00 |
| mordred | tristanC: on neat! I can't keep up with it all ;) | 15:00 |
| tristanC | quick question, is it needed to restart executor when the main.yaml change? | 15:01 |
| mordred | not the executor - the scheduler needs to be told to re-read its config - which currently is done via SIGHUP (right jeblair?) | 15:01 |
| mordred | we have a todo list item to make a "zuul-scheduler reconfigure" command like other things have | 15:02 |
| tristanC | it seems like the scheduler is doing reconfigure on sighup indeed | 15:02 |
| mordred | yah | 15:02 |
| tristanC | and there are "TODOv3 implement" note on the executor graceful action | 15:04 |
| tristanC | well the scheduler seems to send everything the executor need, just making sure it doesn't need to be manually restarted or something | 15:04 |
| jeblair | mordred, tristanC: yes you're both right about everything! | 15:06 |
| mordred | tristanC: yah - executor only needs to be restarted on code changes | 15:07 |
| dmsimard | tristanC, mordred: are we deploying sql reporter and the web UI on the rollout next week or is that going to wait for a while still ? | 15:18 |
| jeblair | dmsimard: sql reporter yes (already done); webui no | 15:19 |
| tristanC | dmsimard: i actually dropped the webui part, my javascript-fu is not good... | 15:21 |
| dmsimard | tristanC: us both then | 15:21 |
| jeblair | we'll add the webui as soon as we can :) | 15:21 |
| dmsimard | tristanC: people say I've done a good job with the ara web ui but really it's all copypasta from patternfly.org | 15:21 |
| tristanC | though i kept the backend and the zuul cli so that sql reporter data can still be easily consumed | 15:22 |
| tristanC | i mean kept in the review, that is https://review.openstack.org/466561 ftr | 15:23 |
| jeblair | tristanC: once we get v3 running, will you help us get up to speed with the dashboard work you did so we can continue work on it? | 15:24 |
| jeblair | i really likedi t | 15:25 |
| jeblair | liked it | 15:25 |
| tristanC | jeblair: yes sure, I hope to gather use-cases at the ptg | 15:26 |
| tristanC | on a similar topic, I proposed some new route to the nodepool webapp to be able to query available labels | 15:27 |
| tristanC | which i think would be a nice addition to the zuul "dashboard", it's https://review.openstack.org/500254 | 15:28 |
| dmsimard | jeblair, mordred: What's your top 3 request for ara ? I'll priorize them after I'm done with the current API work I can hopefully show them at the ptg. | 15:33 |
| dmsimard | Playbook naming ? Playbook grouping ? | 15:33 |
| tristanC | i made some good progress on the zuulv3 integration in software-factory, there is now a "--upstream-zuul-jobs" option to auto configure openstack-infra/zuul-jobs as a untrusted-project | 15:34 |
| jeblair | dmsimard: 1) (highest priority) customized playbook name; 2) un-reverse playbook order | 15:36 |
| dmsimard | tristanC: huh, that reminds me.. is there anything new we should be aware of from a third party CI perspective ? I guess our zuul is connected to gerrit, not to zuul, so it should be transparent | 15:36 |
| dmsimard | jeblair: ah, yes, ordering. | 15:36 |
| mordred | dmsimard, jeblair: I'd put 'ability to indicate job/phase grouping' in there too | 15:38 |
| dmsimard | okay, so we got 1 2 3 :D | 15:39 |
| tristanC | dmsimard: yes, rdo consumes the gerrit event-stream which isn't going to change | 15:39 |
| dmsimard | playbook phase/groups will be last since it will be the one involving the most work of the three | 15:39 |
| dmsimard | thanks. | 15:39 |
| mordred | dmsimard: I had half a patch written the other day when musing about grouping- I pushed it up just for the fun of it | 15:51 |
| dmsimard | mordred: ack, will take a look | 15:51 |
| mordred | it's not an exciting patch :) | 15:51 |
| dmsimard | jeblair: is it currently possible for a job to dynamically define a success/failure message ? Say, to highlight errors that could have occurred during the job, for example. | 15:52 |
| dmsimard | mordred: just looked at the patch ಠ_ಠ | 15:53 |
| dmsimard | mordred: the hard part is more than likely the UI/UX lol | 15:54 |
| dmsimard | but I have some idas | 15:54 |
| dmsimard | ideas* | 15:54 |
| *** jkilpatr has quit IRC | 17:09 | |
| *** olaph has quit IRC | 18:09 | |
| *** olaph has joined #zuul | 18:11 | |
| *** olaph1 has joined #zuul | 18:15 | |
| *** olaph has quit IRC | 18:17 | |
| *** olaph1 is now known as olaph | 18:26 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul-jobs master: Override tox requirments with zuul git repos https://review.openstack.org/489719 | 19:07 |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul-jobs master: Override tox requirments with zuul git repos https://review.openstack.org/489719 | 19:14 |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul-jobs master: Add UPPER_CONSTRAINTS_FILE file if it exists https://review.openstack.org/500320 | 19:14 |
| *** yolanda has joined #zuul | 19:21 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul-jobs master: Override tox requirments with zuul git repos https://review.openstack.org/489719 | 19:22 |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul-jobs master: Override tox requirments with zuul git repos https://review.openstack.org/489719 | 19:37 |
| openstackgerrit | Monty Taylor proposed openstack-infra/nodepool feature/zuulv3: Fix nodepool's cdocs about config-drive https://review.openstack.org/500391 | 19:44 |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul-jobs master: Override tox requirments with zuul git repos https://review.openstack.org/489719 | 19:57 |
| *** jesusaur has quit IRC | 20:40 | |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul feature/zuulv3: Only run zuul-stream testing on callback changes https://review.openstack.org/500168 | 20:40 |
| *** jesusaur has joined #zuul | 20:46 | |
| dmsimard | mordred: bah there's some value in your patch for groups. I hadn't considered two layers, just one. I guess it makes sense. | 21:25 |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul-jobs master: Override tox requirments with zuul git repos https://review.openstack.org/489719 | 21:29 |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul-jobs master: Override tox requirments with zuul git repos https://review.openstack.org/489719 | 21:39 |
| jeblair | dmsimard: https://docs.openstack.org/infra/zuul/feature/zuulv3/user/config.html#attr-job.success-message | 22:03 |
| dmsimard | jeblair: right, I've seen that, but that's static in the zuul.yaml file | 22:04 |
| jeblair | dmsimard: ah, yeah, nothing now but we can add some more dynamic stuff with zuul_return later | 22:05 |
| openstackgerrit | Paul Belanger proposed openstack-infra/zuul feature/zuulv3: Support IPv6 with zuul_stream https://review.openstack.org/500401 | 22:31 |
| openstackgerrit | Paul Belanger proposed openstack-infra/zuul feature/zuulv3: Support IPv6 with zuul_stream https://review.openstack.org/500401 | 22:55 |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul feature/zuulv3: Move ara output generation to post playbook https://review.openstack.org/500403 | 23:05 |
| openstackgerrit | Monty Taylor proposed openstack-infra/zuul feature/zuulv3: Move ara output generation to post playbook https://review.openstack.org/500403 | 23:12 |
| openstackgerrit | Merged openstack-infra/zuul feature/zuulv3: Only run zuul-stream testing on callback changes https://review.openstack.org/500168 | 23:22 |
| openstackgerrit | Paul Belanger proposed openstack-infra/zuul feature/zuulv3: Use afs_publisher_target for afs-docs https://review.openstack.org/500406 | 23:32 |
| openstackgerrit | Monty Taylor proposed openstack-infra/nodepool feature/zuulv3: Fix nodepool's docs about config-drive https://review.openstack.org/500391 | 23:55 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!