2018-05-23 20:32 GMT+02:00 Andres Freund <and...@anarazel.de>: > On 2018-05-22 16:39:58 -0700, Andres Freund wrote: > > Hi, > > > > On 2018-05-23 00:04:26 +0200, Paolo Crosato wrote: > > > I managed to recover the log of the first time we run into the issue, > the > > > error was the same but on template1: > > > > > > May 8 11:26:46 xxx postgres[32543]: [1154-1] user=,db=,client= ERROR: > > > found xmin 2600758304 from before relfrozenxid 400011439 > > > May 8 11:26:46 xxx postgres[32543]: [1154-2] user=,db=,client= > CONTEXT: > > > automatic vacuum of table "template1.pg_catalog.pg_authid" > > > > pg_authid (along with a few other tables) is shared between > > databases. So that's just hte same error. At which rate are you > > creating / updating database users / roles? > > Other questions: > - did you ever use VACUUM FULL or CLUSTER on pg_authid (or just on all > tables)? > - Did you have any failovers? > - Do you use logical replication? >
1) VACUUM FULL was issued after the first time the error occurred, and a couple of times later. CLUSTER was never run. 2) Several failovers tests were perfomed before the cluster was moved to production. However, before the move, the whole cluster was wiped, including all the application and monitoring users. After the db was moved to production, a couple of users were added without any problem. 3) No, even if the replication level is set to logical in postgresql.conf, we only use streaming replication. Best Regards, Paolo Crosato
