[01:38:07] <springle>	 jynus: ugh, sorry. looks like google doc share notification went into my cronspam folder
[01:38:10] <springle>	 looking...
[11:15:46] <jynus>	 *TokuDB* table crashed on labsdb1002; recreated, restarted replication
[11:27:11] <jynus>	 extra columns on "image" table on commons bumped the disk usage, probably more than it should. Some of them were done online, maybe they require a recreation.
[11:31:19] <jynus>	 -140GB of space on db1042, while the table is only 88GB. UNDO/tmp on ibdata?
[16:06:41] <jynus>	 in the last days there are some higher-than usual, but still low number of db timeouts
[16:06:51] <jynus>	 could be anything, really
[16:08:08] <jynus>	 timeouts on connection (check kibana), and with no specific differences per mw host, and the expected differences by db host by traffic