| -@gerrit:opendev.org- Zuul merged on behalf of James E. Blair https://matrix.to/#/@jim:acmegating.com: [zuul/zuul] 839639: Fix updating change when emulating gerrit topic support https://review.opendev.org/c/zuul/zuul/+/839639 | 05:28 | |
| -@gerrit:opendev.org- Zuul merged on behalf of James E. Blair https://matrix.to/#/@jim:acmegating.com: | 05:58 | |
| - [zuul/nodepool] 840241: Measure rate limits from start times https://review.opendev.org/c/zuul/nodepool/+/840241 | ||
| - [zuul/nodepool] 840242: Fix object refresh in the AWS adapter https://review.opendev.org/c/zuul/nodepool/+/840242 | ||
| -@gerrit:opendev.org- Albin Vass proposed: | 07:55 | |
| - [zuul/zuul-jobs] 839225: mirror-workspace-git: urlencode src_dir https://review.opendev.org/c/zuul/zuul-jobs/+/839225 | ||
| - [zuul/zuul-jobs] 841376: Re-enable tests for prepare-workspace-git https://review.opendev.org/c/zuul/zuul-jobs/+/841376 | ||
| -@gerrit:opendev.org- likui proposed: [zuul/nodepool] 804837: setup.cfg: Replace dashes with underscores https://review.opendev.org/c/zuul/nodepool/+/804837 | 09:03 | |
| -@gerrit:opendev.org- Tristan Cacqueray proposed: [zuul/zuul] 841418: command: register the stdout when no_log is set https://review.opendev.org/c/zuul/zuul/+/841418 | 12:19 | |
| @tristanc_:matrix.org | Hello folks, it seems like Zuul discards the output of tasks running with no_log, not sure if this is on purpose, so here is a change to fix that ^ | 12:21 |
|---|---|---|
| -@gerrit:opendev.org- Tristan Cacqueray proposed: [zuul/zuul] 841418: command: register the stdout when no_log is set https://review.opendev.org/c/zuul/zuul/+/841418 | 13:13 | |
| -@gerrit:opendev.org- Dominik Jaeger proposed: [zuul/zuul] 839550: Add spec for fileset feature https://review.opendev.org/c/zuul/zuul/+/839550 | 13:23 | |
| -@gerrit:opendev.org- Dominik Jaeger proposed: [zuul/zuul] 839550: Add spec for fileset feature https://review.opendev.org/c/zuul/zuul/+/839550 | 13:25 | |
| -@gerrit:opendev.org- Simon Westphahl proposed: [zuul/zuul] 841430: Don't clear branch cache on reconfig https://review.opendev.org/c/zuul/zuul/+/841430 | 13:38 | |
| -@gerrit:opendev.org- Simon Westphahl proposed: [zuul/zuul] 841430: Don't clear branch cache on reconfig https://review.opendev.org/c/zuul/zuul/+/841430 | 14:38 | |
| @fzzfh:matrix.org | hello folks. I use nodepool connect openstack provider occur ssh connect error. my openstack cloud configure public and prviate netowrk. instance create by nodepool wouldn't assign public float-ip automatically. but nodepool would ssh to private network. so ssh failed. log like follow. what should do let nodepool create float-ip and attach to instance. | 14:46 |
| on openstck, I create instance manually, it wouldn't assign float-ip automatically. I need generate float-ip and attach to insance, then I can ssh to instance. and when I want create independent float-ip. after a moment, it would be automatic deleted by openstack, is there have config to set? | ||
| ``` | ||
| 2022-05-10 06:41:42,886 INFO nodepool.NodeLauncher: [e: 337d45649dcb4bf5aa3e41fd92cd9813] [node_request: 900-0000012233] [node: 0000010250] Creating server with hostname ubuntu-focal-openstack-provider-0000010250 in openstack-provider from image ubuntu-focal | ||
| 2022-05-10 06:43:56,722 ERROR nodepool.NodeLauncher: [e: 337d45649dcb4bf5aa3e41fd92cd9813] [node_request: 900-0000012233] [node: 0000010250] Launch attempt 1/3 failed: | ||
| Traceback (most recent call last): | ||
| File "/usr/lib/python3.6/site-packages/nodepool/driver/openstack/handler.py", line 247, in launch | ||
| self._launchNode() | ||
| File "/usr/lib/python3.6/site-packages/nodepool/driver/openstack/handler.py", line 231, in _launchNode | ||
| port=connection_port) | ||
| File "/usr/lib/python3.6/site-packages/nodepool/nodeutils.py", line 79, in nodescan | ||
| "connection to %s on port %s" % (ip, port)): | ||
| File "/usr/lib/python3.6/site-packages/nodepool/nodeutils.py", line 39, in iterate_timeout | ||
| raise exc("Timeout waiting for %s" % purpose) | ||
| nodepool.exceptions.ConnectionTimeoutException: Timeout waiting for connection to 172.32.65.152 on port 22 | ||
| 2022-05-10 06:43:57,805 INFO nodepool.NodeLauncher: [e: 337d45649dcb4bf5aa3e41fd92cd9813] [node_request: 900-0000012233] [node: 0000010250] Node b07a474e-c834-4d21-8dc4-4cabb11a3463 scheduled for cleanup | ||
| ``` | ||
| nodepool openstack provider configure like | ||
| ``` | ||
| providers: | ||
| - name: openstack-provider | ||
| driver: openstack | ||
| cloud: packstack | ||
| region-name: 'RegionOne' | ||
| clean-floating-ips: true | ||
| image-name-format: '{image\_name}-{timestamp}' | ||
| boot-timeout: 120 | ||
| launch-timeout: 900 | ||
| rate: 1.0 | ||
| diskimages: | ||
| - name: ubuntu-focal | ||
| pools: | ||
| - name: main | ||
| max-servers: 1 | ||
| networks: | ||
| - in-net | ||
| security-groups: | ||
| - default | ||
| auto-floating-ip: True | ||
| labels: | ||
| - name: ubuntu-focal | ||
| min-ram: 4096 | ||
| flavor-name: m1.large | ||
| diskimage: ubuntu-focal | ||
| ``` | ||
| -@gerrit:opendev.org- Dominik Jaeger proposed: [zuul/zuul] 839550: Add spec for fileset feature https://review.opendev.org/c/zuul/zuul/+/839550 | 14:47 | |
| @fzzfh:matrix.org | clouds.yaml. what else should I configure | 14:48 |
| ``` | ||
| clouds: | ||
| packstack: | ||
| auth: | ||
| auth_url: http://172.0.37.47:5000/v3 | ||
| password: 4f34252f23084dfc | ||
| project_domain_name: Default | ||
| project_name: admin | ||
| user_domain_name: Default | ||
| username: admin | ||
| identity_api_version: '3' | ||
| region_name: RegionOne | ||
| volume_api_version: '3' | ||
| ``` | ||
| @clarkb:matrix.org | fzzf: I would attempt to manually boot an instance on that in-net network and try to attach a floating ip and verify that works | 14:50 |
| @clarkb:matrix.org | if it doesn't work then you should get errors of some sort to go off of. If it does work then you can try and replicate what you did through your configs | 14:50 |
| @fzzfh:matrix.org | Clark: I had boot an instance on in-net and attach a floating ip. then I ssh instance on nopool host. it suucess. | 14:53 |
| I don't know how to set these configs on clouds.yaml or openstack cloud. | ||
| @fzzfh:matrix.org | I mean I boot instance manually with in-net and attach float-ip,then ssh would success | 14:55 |
| @fzzfh:matrix.org | in-net is private net. I assign float-ip from ex-net(public net) | 14:58 |
| -@gerrit:opendev.org- Dominik Jaeger proposed: [zuul/zuul] 828125: Add support for filesets in job configuration https://review.opendev.org/c/zuul/zuul/+/828125 | 14:58 | |
| @clarkb:matrix.org | fzzf: you can try setting requires_floating_ip and floating_ip_source on your clouds.yaml config | 14:59 |
| -@gerrit:opendev.org- Dominik Jaeger proposed: [zuul/zuul] 839550: Add spec for fileset feature https://review.opendev.org/c/zuul/zuul/+/839550 | 15:04 | |
| -@gerrit:opendev.org- Joshua Watt proposed: [zuul/zuul] 838924: merger: Handle merges with cherry-pick merge-mode https://review.opendev.org/c/zuul/zuul/+/838924 | 15:06 | |
| -@gerrit:opendev.org- Dominik Jaeger proposed: [zuul/zuul] 839550: Add spec for fileset feature https://review.opendev.org/c/zuul/zuul/+/839550 | 15:06 | |
| @fzzfh:matrix.org | Clark: I'll try. I haven't found this in the SDK document yet. https://docs.openstack.org/openstacksdk/latest/user/index.html | 15:10 |
| @clarkb:matrix.org | Yes, I wonder if they removed documetnation about clouds.yaml there. There was in the past an entire page dedicated to the options I thought | 15:11 |
| @fzzfh:matrix.org | Clark: okay, get it. | 15:16 |
| -@gerrit:opendev.org- Zuul merged on behalf of James E. Blair https://matrix.to/#/@jim:acmegating.com: [zuul/nodepool] 840526: Improve AWS driver performance https://review.opendev.org/c/zuul/nodepool/+/840526 | 15:33 | |
| @clarkb:matrix.org | Zuulians https://review.opendev.org/c/zuul/zuul/+/840987 and https://review.opendev.org/c/zuul/nodepool/+/840990 and their children are semi important bookkeeping chagnes to update release notes for users udpating to zuulv6 | 16:07 |
| @fungicide:matrix.org | Clark: fzzf: yeah, i'm surprised this didn't get copied over when the sdk absorbed the os-client-config library: https://docs.openstack.org/os-client-config/latest/user/configuration.html#config-files | 16:20 |
| @jim:acmegating.com | zuul-maint: while we're doing cleanup, it would be nice to get https://review.opendev.org/840760 in; then i can send a reminder out to folks so there's plenty of time to handle that deprecation. | 16:49 |
| -@gerrit:opendev.org- Zuul merged on behalf of Clark Boylan: [zuul/nodepool] 840990: Update releasenotes to capture python3.8 requirement https://review.opendev.org/c/zuul/nodepool/+/840990 | 17:04 | |
| -@gerrit:opendev.org- Zuul merged on behalf of Clark Boylan: [zuul/nodepool] 840991: Set python metadata in setup.cfg to >=3.8 https://review.opendev.org/c/zuul/nodepool/+/840991 | 17:15 | |
| -@gerrit:opendev.org- Zuul merged on behalf of Clark Boylan: [zuul/zuul] 840987: Amend Zuul v6 releaes notes to include python3.8 min version bump https://review.opendev.org/c/zuul/zuul/+/840987 | 17:41 | |
| @clarkb:matrix.org | corvus: fungi: re the blog addition should we go ahead and fix the emdash error and squash that change down into the parent change to have actual ocntent then work towards landing some changes? I think that is pretty close to being publishable if we remove the intermediate step that I pushed to sketch stuff out | 18:27 |
| @jim:acmegating.com | Clark: sounds good to me; also i had a followup at https://review.opendev.org/840983 | 18:31 |
| @jim:acmegating.com | (see the lorem ipsum at https://review.opendev.org/840979 to see it in action) | 18:32 |
| @clarkb:matrix.org | yup +2 | 18:32 |
| @clarkb:matrix.org | fungi: do you want to fix the emdash problem and squash your change into mine or should I give that a go? | 18:33 |
| @fungicide:matrix.org | yeah, i can revise the reprint and reshuffle the order of the stack a little if it helps. i left the reprint wip because i wasn't sure if we wanted to just make that a backdated blog post or use it as a starting draft for a new article instead | 18:33 |
| @fungicide:matrix.org | just a sec | 18:33 |
| @clarkb:matrix.org | I'm happy to publish the reprint. It is good information and fits the prupose of a blog well then we can always add additional articles afterwards | 18:35 |
| @fungicide:matrix.org | wfm, will un-wip it in that case | 18:35 |
| @clarkb:matrix.org | once we land it I think it will be clear to people that they can push blog posts too and the format is pretty easy to copy pasta | 18:36 |
| @clarkb:matrix.org | Oh also today is two weeks after I announced the change t ostop adding deb src repos in configure mirrors | 18:39 |
| @clarkb:matrix.org | https://review.opendev.org/c/zuul/zuul-jobs/+/839593/1 should be landable now | 18:40 |
| -@gerrit:opendev.org- Jeremy Stanley https://matrix.to/#/@fungicide:matrix.org proposed wip: | 18:49 | |
| - [zuul/zuul-website] 840978: Backdate an initial blog article https://review.opendev.org/c/zuul/zuul-website/+/840978 | ||
| - [zuul/zuul-website] 840979: Ten years of Zuul article https://review.opendev.org/c/zuul/zuul-website/+/840979 | ||
| -@gerrit:opendev.org- Jeremy Stanley https://matrix.to/#/@fungicide:matrix.org proposed on behalf of James E. Blair https://matrix.to/#/@jim:acmegating.com: [zuul/zuul-website] 840983: Add space between blog entries https://review.opendev.org/c/zuul/zuul-website/+/840983 | 18:49 | |
| -@gerrit:opendev.org- Jeremy Stanley https://matrix.to/#/@fungicide:matrix.org marked as active: [zuul/zuul-website] 840978: Backdate an initial blog article https://review.opendev.org/c/zuul/zuul-website/+/840978 | 18:50 | |
| @clarkb:matrix.org | fungi: do we want to squash things down so we don't end up with a half finished blog for a period of time until that last one lands? | 18:51 |
| @fungicide:matrix.org | i guess there's the risk that they don't merge at roughly the same time? sure I can squash them if it's preferable | 18:52 |
| @fungicide:matrix.org | just remember that afs updates get batched every 5 minutes, so the promotes would have to land on either side of the cronjob for that to actually happen | 18:53 |
| @clarkb:matrix.org | ya I'm not sure how likely that is to happen, but if it does happen might be awkward | 18:54 |
| -@gerrit:opendev.org- Jeremy Stanley https://matrix.to/#/@fungicide:matrix.org proposed on behalf of Clark Boylan: [zuul/zuul-website] 840952: Add a Blog system to the zuul website https://review.opendev.org/c/zuul/zuul-website/+/840952 | 18:59 | |
| -@gerrit:opendev.org- Jeremy Stanley https://matrix.to/#/@fungicide:matrix.org proposed wip: [zuul/zuul-website] 840979: Ten years of Zuul article https://review.opendev.org/c/zuul/zuul-website/+/840979 | 18:59 | |
| @fungicide:matrix.org | that squashes out the placeholder entries from your framework change; i'll abandon the two child changes that orphans | 19:00 |
| @clarkb:matrix.org | corvus: ^ maybe you want to check how 840952 renders and approve if it looks good? (there are a few parent changes that need review too | 19:05 |
| @jim:acmegating.com | will do | 19:15 |
| @jim:acmegating.com | done | 20:18 |
| -@gerrit:opendev.org- Zuul merged on behalf of Clark Boylan: [zuul/zuul-website] 840951: Use relative paths in docs/index.html https://review.opendev.org/c/zuul/zuul-website/+/840951 | 20:22 | |
| -@gerrit:opendev.org- Zuul merged on behalf of Jeremy Stanley https://matrix.to/#/@fungicide:matrix.org: [zuul/zuul-website] 840980: Style heading levels 4-6 https://review.opendev.org/c/zuul/zuul-website/+/840980 | 20:23 | |
| -@gerrit:opendev.org- Zuul merged on behalf of Clark Boylan: [zuul/zuul-website] 840952: Add a Blog system to the zuul website https://review.opendev.org/c/zuul/zuul-website/+/840952 | 20:23 | |
| @clarkb:matrix.org | thanks! | 20:50 |
| -@gerrit:opendev.org- Zuul merged on behalf of James E. Blair https://matrix.to/#/@jim:acmegating.com: [zuul/zuul] 840760: Add a script to find deprecated queue syntax https://review.opendev.org/c/zuul/zuul/+/840760 | 21:07 | |
| -@gerrit:opendev.org- Zuul merged on behalf of Clark Boylan: [zuul/zuul] 840992: Set python-requires to >=3.8 https://review.opendev.org/c/zuul/zuul/+/840992 | 21:18 | |
| @y2kenny:matrix.org | what is the scalability effect of zookeeper to zuul? (i.e., what can be the differences between having 1 zk vs 3 zk vs 10 zk?) Or is it more of a reliability/availability thing? | 21:33 |
| @avass:vassast.org | Kenny Ho: i think it's mostly for reliability | 21:35 |
| @avass:vassast.org | Iirc reads scale with more replicas but writes doesn't really scale with more zks | 21:36 |
| @y2kenny:matrix.org | ok thanks. I am still trying to figure out how to unfreeze zuul. The same dev push the same 30 patchset again and zuul is basically halted for 2 hrs. The only scalable thing that I haven't scaled up is zk and I am wondering if that would help. | 21:37 |
| @y2kenny:matrix.org | So looking at the log again here are the pattern I can see: | 21:54 |
| @y2kenny:matrix.org | the scheduler log is basically filled with pages and pages of "Updating <Change...> and Adding change to queue in pipeline | 21:55 |
| @y2kenny:matrix.org | and then ending with "Scheduling global repo state for item" | 21:56 |
| @y2kenny:matrix.org | at the beginning those 3 things seems to be happening fairly close to each other. (One event have "Updating" at 19:49 and "Adding change" at 19:54. But later on in the log, there is an event that has "Updating" at 20:45 but "Adding change" only happens at 21:04, which is 20 minutes apart. | 22:01 |
| @y2kenny:matrix.org | I am guessing this type of wait is happens for every pipeline but I can't tell what the wait is. In between "Updating" and "Adding change" for an event, it just logs for the same thing for other events. (Also updating change, but the same patchset but under a different event id) | 22:04 |
| @clarkb:matrix.org | Kenny Ho: is it potentially waiting on a merger to become free? maybe check merger logs for those ids? | 22:33 |
| @y2kenny:matrix.org | um... let me see. | 22:33 |
| @y2kenny:matrix.org | merger events seems to happen after the "Adding change" at 21:14 | 22:38 |
| @y2kenny:matrix.org | and the merge appears to complete its work within 3~4 minutes | 22:39 |
| @y2kenny:matrix.org | ending with removing request at 21:17 | 22:40 |
| @y2kenny:matrix.org | so "Updating" @ 20.45, "Adding change" @ 21.04, "Merging for change" @21:13 - 21:17, "Scheduling global repo state" @ 21:25... oh wait... the scheduling global repo state is for a different pipeline | 22:43 |
| @y2kenny:matrix.org | I am not sure why the merge is attempting to merge though given that the repo is under the include: [] in the tenant config. Also, if I understand the log correctly, I basically have ~30(# of commits in series)*30(# of commits in series)* # of pipeline of merge attempts? | 22:47 |
| @clarkb:matrix.org | Zuul needs to merge the change in order to run jobs against it. But it shouldn't need to do it early to determine the config if include is [] | 22:50 |
| @y2kenny:matrix.org | 30X30X# pipeline I mean | 22:50 |
| @clarkb:matrix.org | I wonder if it is just doing the merges anyway and then discarding the results | 22:51 |
| @clarkb:matrix.org | Kenny Ho: I think it may be closer to sum(1..30) rather than 30*30 | 22:51 |
| @y2kenny:matrix.org | oh right yes... earlier commit wouldn't need to do all 30 | 22:52 |
| @clarkb:matrix.org | the base change ahs no parents and is singular. Then the next one does the base change and it. Then the one after that is base change, previous, and current | 22:52 |
| @y2kenny:matrix.org | I guess the other thing is, this patch series has no job for it (it's a developer private branch)... I am guessing the scheduler is not using the branch information from the stream event? or may be it's the other way around... the scheduler is aware of the branch info from stream event but it is not aware of the no-job situation because something else is responsible for the job filtering? | 22:57 |
| @y2kenny:matrix.org | I guess in the opendev typical use case, the branch/job filtering config is in the repo itself so merger needs to attempt merging first | 22:59 |
| @clarkb:matrix.org | yes for the vast majority of our changes zuul needs to merge the change to determine what jobs are appropriate for it | 23:02 |
| @clarkb:matrix.org | The scheduelr is aware of the events to that branch via the event stream. It needs to process the events sufficiently to determine no further action should be taken | 23:03 |
| @y2kenny:matrix.org | ok. But does the scheduler also know that branch has no job? but I guess that goes back to the earlier point about scheduler wouldn't know until merge is attempted. Is it possible add some kind of short circuting for project that is under include: []? | 23:06 |
| @y2kenny:matrix.org | basically, if the repo is under empty include and there isn't any job for it, there won't be any job for it | 23:07 |
| @y2kenny:matrix.org | (not sure if I am phrasing this correctly...) | 23:08 |
| @y2kenny:matrix.org | whereas, in zuul typical use case, there isn't any job for this repo currently, but there might be job for this repo after I try merging it and look into it | 23:09 |
| @clarkb:matrix.org | Yes, it is possible that zuul could optimize when include: [] is listed by checking its current config pre merge before proceeding | 23:09 |
| @clarkb:matrix.org | It is important to note that include: [] doesn | 23:09 |
| @clarkb:matrix.org | * It is important to note that include: \[\] doesn't imply no jobs. So zuul would still need to check taht condition. | 23:09 |
| @y2kenny:matrix.org | right... it doesn't imply no jobs defined in other repo... just that in the current known universe, if there is no jobs already, there wouldn't be a possibility | 23:11 |
| @y2kenny:matrix.org | whereas in the typical use case, even when there is no jobs for the repo in the current known universe, it's possible that a job is coming after merge | 23:12 |
| @y2kenny:matrix.org | or may be I am still confused about what zuul is known when... | 23:13 |
| @y2kenny:matrix.org | my assumption is that if there's a 3rd party repo defining jobs for this repo, the job is known | 23:13 |
| @y2kenny:matrix.org | by the scheduler | 23:14 |
| @y2kenny:matrix.org | I think using example would be easier for me to express what I mean | 23:16 |
| @y2kenny:matrix.org | so let say repo A is listed under include:[]. There could be repo B that defines job for repo A | 23:16 |
| @y2kenny:matrix.org | but if repo B does not define a job for branch X for repo A, and given that repo A is under include: [], this should imply no jobs for repo A branch X (the current event being processed)? | 23:18 |
| @y2kenny:matrix.org | and whether repo B or any other repo outside of repo A define jobs for repo A is fixed prior to the processing of current event so there shouldn't be checking needed? | 23:20 |
| @clarkb:matrix.org | Yes, I think there is a special case that could potentially be optimized for there. But I'm also pretty sure that Zuul doesn't do that today | 23:23 |
| @clarkb:matrix.org | It would require Zuul be updated to do that | 23:23 |
| @y2kenny:matrix.org | ok | 23:23 |
| @iwienand:matrix.org | Clark: do you have a sec for https://review.opendev.org/c/zuul/zuul-jobs/+/840566 ; i'd like to see if those warnings go away on upload | 23:51 |
| @clarkb:matrix.org | ianw: the :6: appended to the key id is the trust level interpreted by the next command in the pipe? | 23:53 |
| @iwienand:matrix.org | yep, which is "5" in the menu if you do it interactively because ... gpg | 23:54 |
| @iwienand:matrix.org | we already do similar in https://opendev.org/zuul/zuul-jobs/src/branch/master/roles/encrypt-file/tasks/import-key.yaml#L37 | 23:54 |
| @clarkb:matrix.org | ya the only thing I'm trying to sort out is if --show-keys can expose private info | 23:57 |
| @clarkb:matrix.org | would using --fingerprint be more appropriate to match that other code? | 23:57 |
| @clarkb:matrix.org | the man page isn't very verbose on this | 23:57 |
| @clarkb:matrix.org | I'm sure gpg expects you to just know based on some convention or other, but that is all magic to me | 23:58 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!