https://mariadb.com/kb/en/mariadb/gtid/
| Status | Subtype | Assigned | Task | ||
|---|---|---|---|---|---|
| Resolved | aaron | T88445 MediaWiki active/active datacenter investigation and work (tracking) | |||
| Resolved | aaron | T135027 MASTER_POS_WAIT() alternative that works cross-DC | |||
| Open | None | T119626 Eliminate SPOF at the main database infrastructure | |||
| Resolved | jcrespo | T133386 Change dbstore1001 delayed slave to be a direct slave of the eqiad masters | |||
| Resolved | jcrespo | T85266 Look into Maria 10 parallel-replication | |||
| Resolved | jcrespo | T133385 Implement GTID replication on MariaDB 10 servers |
Main theoretical reasons:
Mentioned in SAL [2016-05-10T18:10:27Z] <jynus> testing GTID replication on es2019 T133385 T130702
Approximately half of the servers are using GTID, all the codfw slaves, all the external storage ones and some of the recently reimaged/under maintenance servers on eqiad.
GTID rolled in on all production coredb servers. Resolving now, although it will still be applied to other servers, slowly, due to not being yet upgraded to 10.0, having non-InnoDB engines or having multisource replication and/or local writes.