[00:45:12] <springle>	 so, masters older than db1050 need to be rotated too, due to warranty and disk supply
[00:45:49] <springle>	 but db1058 T105627 should be ok; we'll see that cmj says
[01:50:45] <springle>	 set labsdb1002 slave_parallel_threads=0 and slave_exec_mode=idempotent to get replication moving, but this is not a fix. s5 and s7 showed row update failures with =strict
[01:51:18] <springle>	 eww
[01:51:44] <springle>	 tendril is a missing .htaccess. commit to come
[06:22:57] <jynus>	 the replication errors are expected after the crash
[06:24:09] <jynus>	 are you saying that with the default configuration labsdb1002 replication fails?
[16:18:53] <mutante>	 what does not work about tendril? it seems to work for me
[16:19:02] <mutante>	 for example https://tendril.wikimedia.org/host/view/db1015.eqiad.wmnet/3306
[16:20:31] <mutante>	 btw, the change to add the mising config.php was not merged yet on purpose
[16:20:38] <mutante>	 https://gerrit.wikimedia.org/r/#/c/224205/
[16:23:13] <JohnFLewis>	 mutante: why? :)
[16:25:15] <mutante>	 JohnFLewis: to avoid exactly this "tendril doesnt work since your changes" on the weeked
[16:25:38] <JohnFLewis>	 oh, I thought it was a different reason, perhaps technical than that response
[16:25:49] <mutante>	 it's always technical
[16:26:00] <JohnFLewis>	 well, different technical :)
[16:26:17] <mutante>	 technical: because it had no reviews :)
[16:26:42] <JohnFLewis>	 til now :) (/few hours ago)
[23:26:15] <mutante>	 springle: can we make the last change now.. won't touch it for a while after that and declare it done:)