[10:18:45] 10serviceops, 10Continuous-Integration-Infrastructure (shipyard): Remove graphite data for nodepool - https://phabricator.wikimedia.org/T215172 (10hashar) Yes they are automatically garbage collected after 15 days: ` lang=puppet,name=modules/role/manifests/graphite/production.pp # Nodepool, which has sever... [13:33:48] 10serviceops, 10Operations, 10ops-codfw, 10User-jijiki: Degraded RAID on thumbor2002 - https://phabricator.wikimedia.org/T214813 (10jijiki) [13:52:13] 10serviceops, 10Operations, 10ops-codfw, 10User-jijiki: Degraded RAID on thumbor2002 - https://phabricator.wikimedia.org/T214813 (10jijiki) @Papaul At your convenience, please replace the drive again, the machine should boot. The server has been already depooled since last week. If the server fails to b... [14:27:03] 10serviceops, 10Continuous-Integration-Infrastructure (shipyard), 10Patch-For-Review: Remove graphite data for nodepool - https://phabricator.wikimedia.org/T215172 (10fgiunchedi) Indeed the directory/metrics aren't there anymore: ` graphite1004:~$ ls -la /srv/carbon/whisper/nodepool ls: cannot access '/srv/... [14:29:41] 10serviceops, 10Continuous-Integration-Infrastructure (shipyard), 10Patch-For-Review: Remove graphite data for nodepool - https://phabricator.wikimedia.org/T215172 (10fgiunchedi) 05Open→03Resolved a:03fgiunchedi Patch merged, thanks @hashar ! [16:37:14] Hi! im building helm 2.12.2 to upgrade the helm client [16:37:24] however at some point we should also upgrade tiller [16:49:43] <_joe_> the sooner the better? We'd need to have a place where we keep track of all of this software (k8s, etcd, calico, tiller...) and what needs upgrading [16:52:11] the tiller image is built from here AFAICT https://gerrit.wikimedia.org/r/plugins/gitiles/operations/docker-images/production-images/+/refs/heads/master/images/tiller/Dockerfile.template [16:52:27] so i guess updating the package and triggering a rebuild should be enough if im not wrong [16:52:48] dont know if we can add a LABEL or tag around there (the current one uses latest) [16:55:04] <_joe_> what do you mean? [16:55:49] let me rephrase [16:56:16] the current docker tiller image is built from that template, is built tagged as docker.registry...tiller:latest [16:56:37] so if i update the "tiller" debian package and rebuild it it will retag the "latest" image [16:56:47] i would like to add a 2.12.2 tag to that image [17:31:25] <_joe_> uhm you do that by adding an entry here [17:31:26] <_joe_> https://gerrit.wikimedia.org/r/plugins/gitiles/operations/docker-images/production-images/+/refs/heads/master/images/tiller/changelog [17:31:48] <_joe_> so we can adopt the version of the deb package from now on for the image [17:35:40] <_joe_> we can take a look tomorrow, I'm off for good now [17:36:27] nop thats enough [17:36:28] curl https://docker-registry.wikimedia.org/v2/tiller/tags/list (⎈ |helmmanagement:default) [17:36:28] {"name":"tiller","tags":["0.0.1-20190122","0.0.1","latest"]} [17:36:39] we just need to drop the latest usage and use the pinned tag [17:36:51] thanx _joe_ tty tomorrow [17:47:04] Is there any interest in having stashbot in this channel? The main use here would probably be Phabricator mentions, but I could also setup some !log routing too if that would be useful [17:47:18] https://wikitech.wikimedia.org/wiki/Tool:Stashbot [18:17:54] 10serviceops, 10Release-Engineering-Team, 10Release Pipeline (Blubber): Base Blubber policy file for CI - https://phabricator.wikimedia.org/T215319 (10thcipriani) [18:19:59] 10serviceops, 10Release-Engineering-Team, 10Release Pipeline (Blubber): Base Blubber policy file for CI - https://phabricator.wikimedia.org/T215319 (10thcipriani) p:05Triage→03Normal Clarification needed from #serviceops folks: is it only the base-image for the production variant we want to restrict? Fr... [18:36:21] 10serviceops, 10Performance-Team, 10Thumbor, 10Patch-For-Review: Failing Thumbor PDF build test (JPX) - https://phabricator.wikimedia.org/T215253 (10jijiki)