Tuesday, 2022-06-14

opendevreviewDavid Hill proposed openstack/diskimage-builder master: Add a warning in satellite configuration  https://review.opendev.org/c/openstack/diskimage-builder/+/84565000:40
opendevreviewDavid Hill proposed openstack/diskimage-builder master: Add a warning in satellite configuration  https://review.opendev.org/c/openstack/diskimage-builder/+/84565000:42
*** rlandy|bbl is now known as rlandy00:57
*** rlandy is now known as rlandy|out01:45
*** ysandeep|sick is now known as ysandeep04:37
*** jpena|off is now known as jpena07:35
*** ysandeep is now known as ysandeep|lunch07:45
*** ysandeep|lunch is now known as ysandeep09:14
*** soniya29|ruck is now known as soniya29|ruck|afk09:32
*** rlandy|out is now known as rlandy10:23
*** soniya29|ruck|afk is now known as soniya29|ruck10:23
BlaisePabon[m]1...probably not... 11:21
BlaisePabon[m]1is there a stray commit in the change?11:22
*** dviroel|out is now known as dviroel11:22
fungiBlaisePabon[m]1: instead of revising the first change to address the review comments, you pushed a separate change with the first change as its parent11:22
funginormally you'd just revise the first change by amending it and pushing again11:22
fungiremember that in a gerrit workflow (similar to the linux kernel mailing list workflow), you're iterating on improvements to a proposed patch as it will eventually merge, rather than creating a series of patches with each one correcting the mistakes in the previous one11:24
fungiso each revision replaces the original patch in its entirety11:25
fungigerrit will display the differences between each revision of the patch so reviewers can see how it has evolved11:25
*** rlandy is now known as rlandy|mtg11:55
*** ysandeep is now known as ysandeep|afk12:15
*** ysandeep|afk is now known as ysandeep12:30
opendevreviewDavid Hill proposed openstack/diskimage-builder master: Add a warning in satellite configuration  https://review.opendev.org/c/openstack/diskimage-builder/+/84565012:45
opendevreviewBlaise R Pabon proposed opendev/infra-manual master: Update the sandbox instructions Clarify the instructions, reduce ambiguity, increase consistency. Rather than change the legacy OpenStack  URLs, I changed the text to acknowledge that we refer to OpenStack.  https://review.opendev.org/c/opendev/infra-manual/+/84574312:46
opendevreviewBlaise R Pabon proposed opendev/infra-manual master: Update the sandbox instructions  https://review.opendev.org/c/opendev/infra-manual/+/84574412:52
fricklerBlaisePabon[m]1: maybe you should keep practicing in the sandbox for a bit, you keep submitting new patches instead of updating the original one12:53
*** gthiemon1e is now known as gthiemonge13:15
*** rlandy|mtg is now known as rlandy13:30
bjonglez_I want to propose changes to Neutron, so I followed https://docs.opendev.org/opendev/infra-manual/latest/gettingstarted.html13:43
bjonglez_I have an account on Gerrit (bjonglez), and I added my SSH key(s) to it yesterday13:43
bjonglez_however, "git review -s" doesn't work, my SSH key is not accepted13:44
bjonglez_ssh review.opendev.org13:44
bjonglez_bjonglez@review.opendev.org: Permission denied (publickey).13:44
bjonglez_I'm pretty sure I'm using the right key, I uploaded all of the default ones (id_rsa.pub, id_ecdsa.pub and id_ed25519.pub) to Gerrit13:45
bjonglez_ah, right, I'm using the wrong SSH port...13:47
bjonglez_so, SSH login works with port 29418, but "git review -s" still fais:13:48
bjonglez_The following command failed with exit code 255 "scp -P29418 bjonglez@review.opendev.org:hooks/commit-msg .git/hooks/commit-msg"13:48
bjonglez_running the scp manually:13:48
bjonglez_subsystem request failed on channel 013:48
bjonglez_scp: Connection closed13:48
BlaisePabon[m]1hi bjonglez_ I'm also getting started with Gerrit. Would you like to practice together?13:55
* BlaisePabon[m]1 sent a code block: https://matrix.org/_matrix/media/r0/download/matrix.org/vivpuayVnhPrFBwGdUDHxOnu13:57
BlaisePabon[m]1in your example you used `scp`14:00
jrosser_bjonglez_: which version of git-review do you have?14:08
bjonglez_jrosser_: the version from Debian bookworm, 2.2.0-114:09
bjonglez_I heard about scp being deprecated, maybe my SSH client or the Gerrit server stopped supporting it14:10
bjonglez_ok, I switched to a HTTPS-based remote URL for gerrit and "git review -s" downloaded the hook14:10
jrosser_this is probably related to you haveing openssh 9.x14:12
jrosser_later releases of git-review will work around this14:13
jrosser_i would recommend installing using pip (and maybe also a virtualenv) rather than using your OS package manager14:13
*** ysandeep is now known as ysandeep|out14:25
*** dasm|off is now known as dasm14:31
opendevreviewBlaise R Pabon proposed opendev/infra-manual master: Update the sandbox instructions  https://review.opendev.org/c/opendev/infra-manual/+/84559314:33
fungibjonglez_: yes, as jrosser_ points out, the scp subsystem deprecation override is in git-review 2.3.0 and later, so 2.2.0 is definitely too old14:38
fungizigo is one of the maintainers for that package in debian, so may want to be aware. in particular, you could file a bug report against the git-review package in debian asking for it to be updated to latest upstream, since that version is no longer compatible with the openssh version in sid and bookworm14:39
zigoYes, please file a bug, and I'll do the work.14:40
fungibjonglez_: if you want to hand-patch your local version with just that fix, you can also just apply the diff for git_review/cmd.py in https://review.opendev.org/823413 (it's really just four additional lines of python)14:43
bjonglez_fungi: right, thanks a lot! I will open a debian bug report then14:54
Clark[m]Side note rsa keys won't work with Gerrit until 3.6 if using newer openssh14:57
BlaisePabon[m]1ed25519 FTW15:00
*** soniya29|ruck is now known as soniya29|out15:21
*** dviroel is now known as dviroel|lunch15:58
opendevreviewMerged openstack/project-config master: Remove lower-constraints and tox-py36 from Neutron Grafana  https://review.opendev.org/c/openstack/project-config/+/84425415:58
opendevreviewMerged openstack/project-config master: Add py38/py39 jobs to Grafana  https://review.opendev.org/c/openstack/project-config/+/84559615:58
*** marios is now known as marios|out16:00
admin1https://gist.githubusercontent.com/a1git/451de74df71977c971455abb806eede9/raw/193defa9d156d04507373f77b2808514e9252034/gistfile1.txt  -- this is how I am creating the 8080 via user_variables override 16:05
fungiadmin1: sorry, i must be missing some context since i have no idea what that is... did you mean to post in a different channel maybe?16:35
*** jpena is now known as jpena|off17:04
admin1fungi .. sorry .. it was for a diff channel 17:08
fungino problem! just making sure i wasn't missing something important ;)17:13
*** dviroel|lunch is now known as dviroel17:17
*** rlandy is now known as rlandy|mtg18:15
*** rlandy|mtg is now known as rlandy19:01
ianwit looks like since i last looked at grafana alerting it's all basically completely changed anyway -- https://grafana.com/docs/grafana/latest/alerting/alerting-rules/create-grafana-managed-rule/19:52
ianwthe whole rsa key thing is definitely a big carrot for 3.620:01
ianwreturning to a prior issue -> https://grafana.opendev.org/d/1f6dfd6769/opendev-load-balancer?orgId=1&from=now-7d&to=now&viewPanel=520:05
ianwit seems like we had 4 days of really big spikes at periodic job time (it seemed to be causing responsiveness issues for gitea)20:06
ianwbut things have settled down now?  20:06
ianwtheories on this welcome :)20:06
fungimaybe there was something else going on in the background? maybe some other environment also has a cronjob which was pulling from gitea in addition to ours?20:15
ianwit could have been/be external, i had missed the actual period of high load and didn't do a retroactive analysis of the traffic logs to see where it was coming from20:20
clarkbzuul jobs should never talk to gitea20:25
clarkbI think that rule is broken by constraints but otherwise is generally true20:26
ianwmaybe one periodic job is doing something odd * number of opendev repositories or something?20:28
clarkbianw: left a note on the docs update with a -1 since it impacts one of the goals of the change20:34
*** dviroel is now known as dviroel|afk20:46
fungithe openstack/releases jobs also pull from gitea because they can't predict in advance which openstack repos they'll need20:50
*** dasm is now known as dasm|off21:02
*** rlandy is now known as rlandy|bbl21:58
opendevreviewIan Wienand proposed opendev/system-config master: gerrit docs: cleanup and use shell-session  https://review.opendev.org/c/opendev/system-config/+/84506622:01
ianwclarkb: ^ thanks22:02
ianwclarkb: i think where i got stuck was that i needed to remove the email address from the external-id too.  in the instructions it calls delete on the identity, do you think we need to note you might need to delete the email too?22:03
clarkbianw: I'm not sure I understand the question.22:06
clarkboh I see22:06
fungithat's what one of the account cleanup scripts doea22:06
clarkbidentity in this case being the openid as described in the doc22:07
clarkbyes the email has to be remove dtoo to remove the conflict if that is the issue22:07
clarkbif the user never added extra emails that won't exist22:07
clarkbso its only in the case where they have added emails on top of what the openid specified22:07
ianwok, so that is an expected possible outcome.  i might add a note22:08
fungiin short, one of the newer gerrit behaviors is that you can't use an e-mail address in a new account even if it's only associated with an inactive account22:08
opendevreviewIan Wienand proposed opendev/system-config master: gerrit docs: add note that duplicate user may have email addresses to remove  https://review.opendev.org/c/opendev/system-config/+/84585322:12
*** rlandy|bbl is now known as rlandy23:59

Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!