[05:51:26] <wikibugs>	 10DBA, 10Patch-For-Review: Test upgrading sanitarium hosts to Buster + 10.4 - https://phabricator.wikimedia.org/T268742 (10Marostegui)
[06:28:13] <wikibugs>	 10DBA, 10Patch-For-Review: Test upgrading sanitarium hosts to Buster + 10.4 - https://phabricator.wikimedia.org/T268742 (10Marostegui)
[06:35:48] <wikibugs>	 10DBA, 10Patch-For-Review: Test upgrading sanitarium hosts to Buster + 10.4 - https://phabricator.wikimedia.org/T268742 (10Marostegui) So far no errors on s1, s3, s5 and s8 on db1154 since they were re-cloned from the sanitarium masters. Triggers seem to be working fine, no PII there reported by the private da...
[06:39:25] <Amir1>	 marostegui: Hey, I have a problem that I don't know what to do, I looked everywhere and couldn't find a solution
[06:39:26] <Amir1>	 https://phabricator.wikimedia.org/T270032#6687299
[06:39:37] <marostegui>	 Amir1: checking
[06:40:41] <Amir1>	 basically, the column's datatype is datetime and modifying it binary(14) fails because it does a literal conversion to "2020-12-13 17:02:26" which is bigger than 14
[06:40:57] <Amir1>	 not to mention deploying this on production is going to be fun
[06:41:21] <marostegui>	 Amir1: An itermediate solution could be to let it be just varbinary20
[06:42:23] <Amir1>	 yeah, it seems mysql doesn't have anything for it
[06:42:24] <marostegui>	 Amir1: But essentially, the only way to do it would be what you said: varbinary 20 -> script to change values -> varbinary 14
[06:42:28] <marostegui>	 but is it worth?
[06:43:11] <Amir1>	 on its own, it doesn't 
[06:43:33] <Amir1>	 but it's more to clean up the timestamp fields in core schema
[06:43:51] <Amir1>	 have them all be binary(14) (or varbinary(14))
[06:44:11] <Amir1>	 also this is going to make categorylinks bigger I assume which is not fun in commons
[06:44:17] <marostegui>	 yeah
[06:44:39] <marostegui>	 But the transition between having both values will be ok?
[06:44:49] <marostegui>	 so values on varbinary 20 and some others being changed by the script
[06:44:58] <marostegui>	 is that ok from MW point of view?
[06:45:02] <Amir1>	 it seems mediawiki understands both without issues
[06:45:18] <Amir1>	 I haven't tested it from top to bottom but the code looks agnostic
[06:45:39] <Amir1>	 basically it takes the value, finds the format, returns
[06:46:00] <marostegui>	 Ah ok, then it should have no problem, but this clean up can take months 
[06:46:17] <marostegui>	 changing it to varbinary 20, then running the script and then changing it to varbinary 14
[06:46:48] <Amir1>	 yeah :(((((
[06:47:08] <marostegui>	 I am ok with that as a task, but if we are 100% sure that MW can handle both values at the same time
[06:47:21] <marostegui>	 And obviously knowing that it can take loooooong to do it
[06:47:35] <Amir1>	 I'll double check everything
[06:47:45] <marostegui>	 Amir1: if you want me to do a change on a codfw host or something so you can test, let me know
[06:47:52] <Amir1>	 sure
[06:48:03] <Amir1>	 thanks
[06:48:16] <marostegui>	 thank you for dealing with those things :)
[06:48:34] <Amir1>	 the fun thing is that the column is basically useless, it's used in some places but not much
[06:49:36] <marostegui>	 hahaha
[06:49:39] <marostegui>	 lovely
[06:50:33] <wikibugs>	 10DBA, 10Data-Persistence-Backup: Drop table profiling from WMF wiki mariadb servers - https://phabricator.wikimedia.org/T266125 (10Marostegui) a:03Marostegui I will take care of this after the holidays
[06:51:08] <Amir1>	 indeed
[07:24:26] <Amir1>	 marostegui: since you're around, This is needed for upgrading to puppet 6 https://gerrit.wikimedia.org/r/c/operations/puppet/+/642649
[07:24:33] <Amir1>	 but pcc is acting weird 
[07:24:55] <Amir1>	 Unable to find fact file for: dbmonitor1001.eqiad.wmnet
[07:26:23] <marostegui>	 Amir1: jbond42 did some changes to pcc I think, maybe we can check with him
[07:28:56] <Amir1>	 added him
[07:28:57] <marostegui>	 Ah wait
[07:29:00] <marostegui>	 I think I know what it is
[07:29:11] <Amir1>	 :D
[07:30:44] <marostegui>	 Amir1: can you spot the difference? https://puppet-compiler.wmflabs.org/compiler1002/27261/  \o/
[07:31:01] <Amir1>	 aaaaah
[07:31:09] <Amir1>	 Thanks
[07:31:30] <marostegui>	 Just commented on the patch
[07:31:58] <Amir1>	 Shall we merge it then
[07:32:05] <Amir1>	 🥺
[07:32:29] <marostegui>	 Not sure if we should merge today or after the holidays break
[07:32:35] <marostegui>	 https://phabricator.wikimedia.org/macro/view/62/ haha
[07:33:03] <Amir1>	 whatever you prefer 
[07:33:20] <marostegui>	 If it is not urgent, let's merge in january I think
[07:33:52] <Amir1>	 yeah, it's not urgent, it's been sitting there for a while
[07:33:58] <Amir1>	 it's just there's a million of them
[07:34:03] <marostegui>	 hahaha
[07:34:08] <marostegui>	 Let's wait then yeah
[07:34:16] <marostegui>	 Pretty much everyone is out today already
[09:21:32] <wikibugs>	 10DBA, 10Patch-For-Review: Productionize x2 databases - https://phabricator.wikimedia.org/T269324 (10Marostegui) These hosts are ready. Pending steps should be quick to do, and will be done after the holidays. To avoid introducing more variables and potential noise creators during the break:  - Enable notifica...
[09:23:04] <wikibugs>	 10DBA, 10Patch-For-Review: Productionize x2 databases - https://phabricator.wikimedia.org/T269324 (10Marostegui)
[17:01:42] <wikibugs>	 10DBA, 10DC-Ops, 10Operations, 10ops-eqiad: (Need By: 2020-11-29) rack/setup/install db11[51-76] - https://phabricator.wikimedia.org/T267043 (10Cmjohnson) a:05Cmjohnson→03Jclark-ctr These are racked and netbox updated but not connected to the switches yet.  @Jclark-ctr could you please cable these and...