[08:24:12] jayme: I have another proxy update patch, if you (or someone else) has time for review and deployment: https://gerrit.wikimedia.org/r/c/operations/puppet/+/1182106 [08:26:12] kostajh: I'd like to leave the review to someone who speaks nginx :D [08:29:51] jayme: fair enough! [08:54:36] hello, getting an error on a homer diff check `Policy error: Policy kubedse_import referenced but not defined In [edit] (policy-options)` on the devices `cr2-codfw.wikimedia.org` and `cr1-codfw.wikimedia.org` [09:00:19] jayme: suggestions for who I could ask for reviews? [09:01:07] stevemunene: I'd suggest to contact #-traffic about that [09:01:37] kostajh: v.gutierrez is probably a good option, given he also did the review for the last patch [09:02:01] ok [10:26:06] vgutierrez: would you have time to review / deploy https://gerrit.wikimedia.org/r/c/operations/puppet/+/1182106 ? [10:28:34] * vgutierrez looking [10:41:37] looks like pcc is failing on an unrelated resource there [10:41:55] because of logic added in https://gerrit.wikimedia.org/r/c/operations/puppet/+/1182148 [10:44:14] tappof ^ [10:45:14] kostajh: happy to review it.. but deploying hcaptcha stuff is definitely out of my scope :) [10:45:34] we can do all required testing post deploy [10:47:21] mszabo: Could you share the link to the PCC compiler output? [10:47:32] tappof: https://puppet-compiler.wmflabs.org/output/1182106/7327/urldownloader1004.wikimedia.org/change.urldownloader1004.wikimedia.org.err [10:47:53] thx mszabo [10:51:58] vgutierrez: what mszabo said above, we can do the verification after deployment. the visible change is that a `__cf_bm` cookie should no longer be seen in the client, because the proxy will have removed it [10:57:03] mszabo: I’ve got a fix, sorry... I’ll submit a patch in a few minutes. [11:06:19] kostajh, mszabo looks good.. it's a pity that nginx doesn't let you filter response headers natively [11:07:18] vgutierrez: yeah its an issue with multiple headers specifically [11:07:28] $upstream_http_foo works if you expect only one [11:07:41] but set-cookie may legitimately appear more than once [11:07:46] yup [11:18:25] https://gerrit.wikimedia.org/r/c/operations/puppet/+/1182148 reverted mszabo [11:18:32] ty [11:35:03] vgutierrez: please let us know if you would like any other changes on the patch [11:41:20] looks good to me kostajh [11:44:34] cool. let's ship it? [12:41:10] kostajh: sure, let me know when [12:41:45] we're ready when you are [12:43:09] ack, let's go then [12:43:36] which servers are taking care of serving hcaptcha.wm.o? [12:43:46] urldownloader1004 alone? [12:44:09] hmm it looks like that [12:45:33] yep [12:46:24] running puppet on urldownloader1004 as we speak [12:47:37] https://www.irccloud.com/pastebin/hJbQcaUH/ [12:47:57] ok... /etc/nginx/lua isn't there? :) [12:49:32] hmm [12:50:17] yeah.. you need to create /etc/nginx/lua first [12:50:25] could you submit a CR? [12:50:28] seems to be referenced in a few other places but I guess not here https://codesearch.wmcloud.org/search/?q=etc%2Fnginx%2Flua&files=&excludeFiles=&repos= [12:51:54] I'm guessing we want something like https://gerrit.wikimedia.org/g/operations/puppet/+/6f691134a3c3c758816f57a661d1f3fe26717fd7/modules/dynamicproxy/manifests/init.pp#179, but not sure where we'd add that for the proxy [12:52:14] https://gerrit.wikimedia.org/r/c/operations/puppet/+/1182539/ [12:52:38] thanks! [12:53:05] vgutierrez: does that need `Hosts: urldownloader1004.wikimedia.org` ? [12:53:25] ah, you've added it [12:58:58] kostajh: looks good? [13:00:57] vgutierrez: seems like it's worth trying [13:03:00] merged and running puppet again in urldownloader1004 [13:03:29] Notice: /Stage[main]/Profile::Hcaptcha::Proxy/File[/etc/nginx/lua]/ensure: created [13:03:29] Notice: /Stage[main]/Profile::Hcaptcha::Proxy/File[/etc/nginx/lua/filter_set_cookie.lua]/ensure: defined content as '{sha256}ffaf1ffec362c7e5ec3102faba9ba0eb0716270fb958d97b4ef31aba0c496083' [13:03:36] ok, that's better [13:04:28] kostajh: your change is now live [13:05:18] vgutierrez: thanks! looking [13:07:44] vgutierrez: looks good, thank you! [15:46:29] 06serviceops, 13Patch-For-Review: Migrate the etcd main cluster to cfssl-based PKI - https://phabricator.wikimedia.org/T352245#11124927 (10Scott_French) [15:47:28] 06serviceops, 13Patch-For-Review: Migrate the etcd main cluster to cfssl-based PKI - https://phabricator.wikimedia.org/T352245#11124940 (10Scott_French) p:05Triage→03Medium [16:15:02] 06serviceops, 06MW-Interfaces-Team, 07OKR-Work: Route test2wiki rest.php APIs through rest-gateway - https://phabricator.wikimedia.org/T402412#11125147 (10Clement_Goubert) p:05Triage→03High [17:28:11] 06serviceops, 06DC-Ops, 10ops-codfw, 06SRE: Q1:rack/setup/install maps201[1-4] - https://phabricator.wikimedia.org/T400637#11125489 (10Jhancock.wm) a:03Jhancock.wm [17:32:48] 06serviceops: Migrate the etcd main cluster to cfssl-based PKI - https://phabricator.wikimedia.org/T352245#11125533 (10Scott_French) Alright, the next and final phase of this is to devise and execute a plan for migrating etcd itself to cfssl-PKI. As noted in the task description, the functionality is already pre... [17:45:55] 06serviceops, 10envoy, 06SRE, 06Traffic: Envoy config updates from v1.26 - https://phabricator.wikimedia.org/T403101 (10RLazarus) 03NEW [18:12:25] 06serviceops, 06MediaWiki-Platform-Team, 07Epic: Migrate Wikimedia production from PHP 8.1 to PHP 8.3 - https://phabricator.wikimedia.org/T360995#11125783 (10Scott_French) [18:25:01] 06serviceops, 06DC-Ops, 10ops-codfw, 06SRE: Q1:rack/setup/install maps201[1-4] - https://phabricator.wikimedia.org/T400637#11125886 (10ops-monitoring-bot) Cookbook cookbooks.sre.hosts.reimage was started by jhancock@cumin1003 for host maps2011.codfw.wmnet with OS bookworm [18:25:14] 06serviceops, 06DC-Ops, 10ops-codfw, 06SRE: Q1:rack/setup/install maps201[1-4] - https://phabricator.wikimedia.org/T400637#11125888 (10ops-monitoring-bot) Cookbook cookbooks.sre.hosts.reimage was started by jhancock@cumin1003 for host maps2012.codfw.wmnet with OS bookworm [18:25:22] 06serviceops, 06DC-Ops, 10ops-codfw, 06SRE: Q1:rack/setup/install maps201[1-4] - https://phabricator.wikimedia.org/T400637#11125889 (10ops-monitoring-bot) Cookbook cookbooks.sre.hosts.reimage was started by jhancock@cumin1003 for host maps2013.codfw.wmnet with OS bookworm [18:25:33] 06serviceops, 06DC-Ops, 10ops-codfw, 06SRE: Q1:rack/setup/install maps201[1-4] - https://phabricator.wikimedia.org/T400637#11125890 (10ops-monitoring-bot) Cookbook cookbooks.sre.hosts.reimage was started by jhancock@cumin1003 for host maps2014.codfw.wmnet with OS bookworm [18:40:34] 06serviceops, 10envoy, 06SRE: Envoy config updates from v1.26 - https://phabricator.wikimedia.org/T403101#11125946 (10ssingh) [18:41:03] 06serviceops, 10envoy, 06SRE: Upgrade Envoy to v1.26.8 and drop buster - https://phabricator.wikimedia.org/T402584#11125948 (10ssingh) [18:41:15] 06serviceops, 10envoy, 06SRE: Upgrade Envoy to >= 1.24 - https://phabricator.wikimedia.org/T380211#11125960 (10ssingh) [18:41:57] 06serviceops, 10envoy, 06SRE: Upgrade Envoy to >= 1.24 - https://phabricator.wikimedia.org/T380211#11125964 (10ssingh) For awareness: I checked with @RLazarus and removing the Traffic tag. We can add back later as required. [18:42:38] 06serviceops, 06DC-Ops, 10ops-codfw, 06SRE: Q1:rack/setup/install maps201[1-4] - https://phabricator.wikimedia.org/T400637#11125967 (10ops-monitoring-bot) Cookbook cookbooks.sre.hosts.reimage started by jhancock@cumin1003 for host maps2011.codfw.wmnet with OS bookworm executed with errors: - maps2011 (**FA... [19:06:31] 06serviceops: Identify areas covered by the Production Readiness checklist - https://phabricator.wikimedia.org/T400476#11126116 (10bking) [20:31:47] 06serviceops, 06DC-Ops, 10ops-eqiad, 06SRE: Q1:rack/setup/install maps101[1-4] - https://phabricator.wikimedia.org/T400638#11126375 (10VRiley-WMF) a:03VRiley-WMF [20:37:03] 06serviceops, 06DC-Ops, 10ops-codfw, 06SRE: Q1:rack/setup/install maps201[1-4] - https://phabricator.wikimedia.org/T400637#11126381 (10Jhancock.wm) [20:37:51] 06serviceops, 06DC-Ops, 10ops-codfw, 06SRE: Q1:rack/setup/install maps201[1-4] - https://phabricator.wikimedia.org/T400637#11126384 (10Jhancock.wm) ran into an issue where the installer thinks that there is a raid when the config does not call for one. need to go back and compare and check for issues. [20:56:35] 06serviceops, 13Patch-For-Review: Create dedicated changeprop-jobqueue rule for CategoryCountUpdateJob - https://phabricator.wikimedia.org/T402873#11126430 (10Scott_French) @Ladsgroup - Thanks for the follow-up! > I don't think so. These stuff are not considered canonical data. The small counters have self-he... [21:49:37] 06serviceops: Migrate the etcd main cluster to cfssl-based PKI - https://phabricator.wikimedia.org/T352245#11126653 (10Scott_French) [23:21:32] 06serviceops, 13Patch-For-Review: Migrate the etcd main cluster to cfssl-based PKI - https://phabricator.wikimedia.org/T352245#11126899 (10Scott_French) Looking at what the PCC diffs on [[ https://gerrit.wikimedia.org/r/1182658 | r/1182658 ]] look like, this looks promising in terms of requiring zero tweaks to...