[10:24:54] s.ukhe: those are two nodes with dead disks that we have no spares for and are being gradually drained prior to decom [17:53:49] as FYI the dump-cloud-ip-ranges timer/unit on puppetservers now works fine, it seems that Microsoft fixed their CDN [22:35:27] !incidents [22:35:27] 5586 (UNACKED) db2212 (paged)/MariaDB Replica SQL: s1 (paged) [22:35:28] 5587 (UNACKED) db2212 (paged)/MariaDB Replica Lag: s1 (paged) [22:42:48] Amir1: do you think it's worth to depool db2212? [23:03:03] depooled db2212 and downtimed for 72h [23:03:32] followd procedure on https://wikitech.wikimedia.org/wiki/MariaDB/Troubleshooting#Depooling_a_replica [23:05:27] Amir1: for when you'll read: I've updated the Index corruption doc as described in the "Temporary incident response steps" doc [23:08:49] !incidents [23:08:49] 5586 (UNACKED) db2212 (paged)/MariaDB Replica SQL: s1 (paged) [23:08:49] 5587 (UNACKED) db2212 (paged)/MariaDB Replica Lag: s1 (paged) [23:10:15] thanks fabfur [23:10:27] (ack the page too) [23:11:22] can I do that from here correct? [23:11:35] `!ack `? [23:13:14] !ack 5586 [23:13:14] 5586 (ACKED) db2212 (paged)/MariaDB Replica SQL: s1 (paged) [23:13:20] !ack 5587 [23:13:32] !incidents [23:13:32] 5586 (ACKED) db2212 (paged)/MariaDB Replica SQL: s1 (paged) [23:13:32] 5587 (UNACKED) db2212 (paged)/MariaDB Replica Lag: s1 (paged) [23:13:37] !ack 5587 [23:13:38] 5587 (ACKED) db2212 (paged)/MariaDB Replica Lag: s1 (paged) [23:13:43] !incidents [23:13:43] 5586 (ACKED) db2212 (paged)/MariaDB Replica SQL: s1 (paged) [23:13:44] 5587 (ACKED) db2212 (paged)/MariaDB Replica Lag: s1 (paged) [23:13:49] well, going to sleep [23:13:49] Could not ack the alert. Please check the parameters. [23:14:09] !incidents [23:14:09] 5586 (ACKED) db2212 (paged)/MariaDB Replica SQL: s1 (paged) [23:14:09] 5587 (ACKED) db2212 (paged)/MariaDB Replica Lag: s1 (paged) [23:38:58] Thanks. Let me take a look [23:44:59] I started the rebuild for now