[07:33:17] 10DBA, 10Operations, 10Patch-For-Review, 10Puppet, 10User-jbond: Extend Puppet CA Expiry date - https://phabricator.wikimedia.org/T236277 (10Joe) One suggestion: shouldn't we keep the old CA cert around while transitioning? What I mean is we should keep the current CA cert around in our approved certif... [08:37:07] 10DBA: Investigate possible memory leak on db1115 - https://phabricator.wikimedia.org/T231769 (10jcrespo) Based on trending, it will start swapping around 18 december if not restarted. [09:49:04] 10DBA, 10Operations, 10Patch-For-Review, 10Puppet, 10User-jbond: Document all uses of the puppetCA certificate - https://phabricator.wikimedia.org/T237259 (10Joe) >>! In T237259#5641301, @jbond wrote: > @Eevans moritz mentioned there maybe some cassandra consideration to take into account and you could e... [10:16:51] 10DBA, 10Operations, 10Patch-For-Review, 10Puppet, 10User-jbond: Document all uses of the puppetCA certificate - https://phabricator.wikimedia.org/T237259 (10Joe) The calico/node service, and the kube-controller-manager service will need to be restarted on the kubernetes workers and masters respectively. [16:04:34] 10DBA, 10TechCom-RFC: MediaWiki database policy and/or guidelines (2019) - https://phabricator.wikimedia.org/T220056 (10Nikerabbit) I think the policy text quoted in the task description can be restored as-is. While reading it I had the following thoughts: * `Update the installer` sections explains the procedu... [16:07:45] 10DBA, 10TechCom-RFC: MediaWiki database policy and/or guidelines (2019) - https://phabricator.wikimedia.org/T220056 (10jcrespo) It is ok if it no longer an official policy- I just want something on Mediawiki.org, even if it is just a link to Wikitech.org for discoverability reasons. [16:46:07] "Boost minor version found is 67 we need 69" on mysql package [16:46:41] I wonder if I should try to package 1.69, patch mysql to allow 67, or other option? [16:47:56] even sid only has 1.67.0.1 [17:25:56] I have patched the config to allow compiling to 1.67, we'll see how/if it breaks on compile to at least know if it helps with compile size [18:36:01] boot is not trivial to backport as the packages are not co-installable and other packages might use it [18:36:37] is that mysql-8? I'd expect that they bundle a copy in the source tree, that might be an accetable option here [18:38:58] moritzm: I was preciselly trying to avoid that 1GB static binary [18:39:23] I had a working compilation, but was trying to link it dynamically [18:39:39] yeah, but if there's really a hard req on 1.69 it's probably the lesser of two evils [18:40:17] yeah, it is hard, it doesn't compile, and it is group replication, which is a core feature [18:40:39] :-/ [20:03:29] actually, I was able to compile with the older version [20:03:39] a different thing is if it works at all [20:08:20] and the mysqld size is still 1GB, despite setting everyhing I could dynamically linking [20:11:33] https://phabricator.wikimedia.org/P9555 [20:14:37] But: nm mysqld | grep boost | wc -l ~> 11886 [20:14:47] so I must be doing something wrong [20:42:35] oh, I finally got it [20:43:01] -rwxr-xr-x jynus/jynus 69545272 2019-11-07 21:40 percona-server-8.0.17-8-linux-x86_64/bin/mysqld