[00:22:48] FIRING: PuppetFailure: Puppet has failed on aqs1021:9100 - https://puppetboard.wikimedia.org/nodes?status=failed - https://grafana.wikimedia.org/d/yOxVDGvWk/puppet - https://alerts.wikimedia.org/?q=alertname%3DPuppetFailure [00:37:48] RESOLVED: PuppetFailure: Puppet has failed on aqs1021:9100 - https://puppetboard.wikimedia.org/nodes?status=failed - https://grafana.wikimedia.org/d/yOxVDGvWk/puppet - https://alerts.wikimedia.org/?q=alertname%3DPuppetFailure [04:49:38] I think the check_data_report for checking new wikis isn't working, we've not received any emails about the new wiki at https://phabricator.wikimedia.org/T390710 [05:01:59] It seems to be working, looks like I started way too early to work today and the cron didn't start yet XD [05:40:24] switched es1 master in eqiad [07:38:18] * Emperor still looking for reviews of https://gerrit.wikimedia.org/r/c/operations/puppet/+/1138830 and its 2 related changes, please (though I won't deploy 'til Monday now) [07:38:42] done [07:46:45] thanks [08:24:52] switched es1 master in codfw [12:55:27] federico3: https://phabricator.wikimedia.org/T390710#10767298 if we don't get to this today it will be alerting the whole weekend [12:56:21] uh? [12:56:45] federico3: Didn't you see today's emails? [12:56:48] The private data ones [12:57:45] no, checking now [12:58:20] federico3: Please create a proper filter for those, they are important [13:01:43] filter on what parameter? [13:03:53] federico3: Check the email and whatever parameters works for you, but please, address the task and later the filters [13:10:58] https://phabricator.wikimedia.org/T390710#10767298 here are you referring to running cookbook sre.mysql.sanitize-wiki --wiki nupwiki --check-only --task T390710 ? (and later on without --check-only ? ) [13:10:58] T390710: Prepare and check storage layer for nupwiki - https://phabricator.wikimedia.org/T390710 [13:11:30] federico3: Yes, the redaction cookbook you merged [13:12:05] federico3: https://phabricator.wikimedia.org/T390710#10699405 [13:12:41] hm , it fails to start [13:17:24] mysql.get_dbs is detecting 2 hosts but the mysql module only handles one [13:18:18] There are 5 hosts to be sanitized, 2 sanitariums (1 per dc) and 3 hosts in eqiad, 2 cloudb* and an-redactteddb1001 [13:18:20] before I tweak the cookbook: is it expected to have db2186.codfw.wmnet and db1154.eqiad.wmnet here? [13:19:08] Of course [13:19:41] The whole process is at https://phabricator.wikimedia.org/T366146 [13:19:45] we are filtering by A:db-sanitarium and A:db-section-s5 here [13:27:21] ok, it's moving forward with a little tweak [13:27:30] good [13:34:23] federico3: Once done, can you update https://wikitech.wikimedia.org/wiki/MariaDB/PII with how to run the cookbook and all that? [13:34:34] ok [13:35:11] I'm seeing an empty set of values in Non-public databases that are present , Non-public tables that are present, Unfiltered columns that are present [13:35:37] What? [13:41:16] I'm referring to https://phabricator.wikimedia.org/P75464 - plus I'm seeing a glitch similar to the one in the related PR [13:42:27] s8 shouldn't have any private data [13:42:30] only s5 [13:42:44] (evidently it was tested on 1-host sets before) [13:52:25] now it failed to run the GRANT SELECT... @Amir1 perhaps you have more context about this? [13:53:01] federico3: which command are you running? [13:53:26] marostegui: you mean how I'm calling the runbook or what query is the runbook running? [13:53:28] The exact command is the one I pasted above [13:53:33] the GRANT SELECT [13:53:39] spicerack.mysql.MysqlError: Failed to run 'GRANT SELECT, SHOW VIEW ON `nupwiki_p`.* TO `labsdbuser`;' on an-redacteddb1001.eqiad.wmnet [13:53:59] Maybe some escaping or something? what is the more detailed error? [13:54:27] sadly it's not logging the error :-/ [13:57:32] I'm a bit puzzled by this syntax Instance(host, name=self.section) [13:57:57] or the original Instance(self.clouddb_hosts.remote_hosts, name=self.section) [14:05:30] Can I get a +1 for https://gerrit.wikimedia.org/r/c/operations/puppet/+/1139046 please, to fix an earlier error? ms-be2089 isn't able to rsync to other nodes ATM [14:07:29] (this does need applying today) [14:09:00] TY :) [14:12:49] whoops, A.mir1 beat me to it [16:49:48] marostegui: the cookbook requires more work, I suggest we make the create table/grant select by hand for now [16:55:27] I am gone for the day , can you do it? [16:57:11] I'm trying and not able to connect to an-redacteddb1001.eqiad.wmnet