Hi, One of my test databases "crashed" over the weekend after one of the sysadmin's in the team shut down the systemd container before stopping the database. My question is, how bad is this :) Some background : I have an application that uses Postgres as it's database. Both the application, and postgres will be upgraded to newer versions in the coming weeks. As part of a test-run I've upgraded the application to ensure that works, before I upgrade the database. The application is undergoing some UAT testing prior to having the database upgraded. Currently I'm running 9.1.10, and will be upgrading to the latest release of 9.3.x (not my choice of version, but the application currently doesn't support 9.5.x) I've taken a full file-system backup of the database directories (with the database shut down), and a pg_dumpall. Having restarted the database everything seems okay until the logs start showing these errors : [2016-05-05 20:08:55 UTC] LOCATION: AbortBufferIO, bufmgr.c:2799 [2016-05-05 20:08:56 UTC] ERROR: XX000: xlog flush request 36D0/E8301358 is not satisfied --- flushed only to 36CA/94E12DC0 [2016-05-05 20:08:56 UTC] CONTEXT: writing block 0 of relation pg_tblspc/16395/PG_9.1_201105231/16400/33044487_vm [2016-05-05 20:08:56 UTC] LOCATION: XLogFlush, xlog.c:2171 [2016-05-05 20:08:56 UTC] WARNING: 58030: could not write block 0 of pg_tblspc/16395/PG_9.1_201105231/16400/33044487_vm [2016-05-05 20:08:56 UTC] DETAIL: Multiple failures --- write error might be permanent. [2016-05-05 20:08:56 UTC] LOCATION: AbortBufferIO, bufmgr.c:2799 [2016-05-05 20:08:57 UTC] ERROR: XX000: xlog flush request 36D0/E8301358 is not satisfied --- flushed only to 36CA/94E12DC0 [2016-05-05 20:08:57 UTC] CONTEXT: writing block 0 of relation pg_tblspc/16395/PG_9.1_201105231/16400/33044487_vm [2016-05-05 20:08:57 UTC] LOCATION: XLogFlush, xlog.c:2171 [2016-05-05 20:08:57 UTC] WARNING: 58030: could not write block 0 of pg_tblspc/16395/PG_9.1_201105231/16400/33044487_vm [2016-05-05 20:08:57 UTC] DETAIL: Multiple failures --- write error might be permanent. [2016-05-05 20:08:57 UTC] LOCATION: AbortBufferIO, bufmgr.c:2799 [2016-05-05 20:08:58 UTC] ERROR: XX000: xlog flush request 36D0/E8301358 is not satisfied --- flushed only to 36CA/94E12DC0 [2016-05-05 20:08:58 UTC] CONTEXT: writing block 0 of relation pg_tblspc/16395/PG_9.1_201105231/16400/33044487_vm [2016-05-05 20:08:58 UTC] LOCATION: XLogFlush, xlog.c:2171 [2016-05-05 20:08:58 UTC] WARNING: 58030: could not write block 0 of pg_tblspc/16395/PG_9.1_201105231/16400/33044487_vm [2016-05-05 20:08:58 UTC] DETAIL: Multiple failures --- write error might be permanent. [2016-05-05 20:08:58 UTC] LOCATION: AbortBufferIO, bufmgr.c:2799 [2016-05-05 20:08:59 UTC] ERROR: XX000: xlog flush request 36D0/E8301358 is not satisfied --- flushed only to 36CA/94E12DC0 [2016-05-05 20:08:59 UTC] CONTEXT: writing block 0 of relation pg_tblspc/16395/PG_9.1_201105231/16400/33044487_vm [2016-05-05 20:08:59 UTC] LOCATION: XLogFlush, xlog.c:2171 [2016-05-05 20:08:59 UTC] WARNING: 58030: could not write block 0 of pg_tblspc/16395/PG_9.1_201105231/16400/33044487_vm [2016-05-05 20:08:59 UTC] DETAIL: Multiple failures --- write error might be permanent. [2016-05-05 20:08:59 UTC] LOCATION: AbortBufferIO, bufmgr.c:2799 [2016-05-05 20:09:00 UTC] ERROR: XX000: xlog flush request 36D0/E8301358 is not satisfied --- flushed only to 36CA/94E12DC0 [2016-05-05 20:09:00 UTC] CONTEXT: writing block 0 of relation pg_tblspc/16395/PG_9.1_201105231/16400/33044487_vm [2016-05-05 20:09:00 UTC] LOCATION: XLogFlush, xlog.c:2171 [2016-05-05 20:09:00 UTC] WARNING: 58030: could not write block 0 of pg_tblspc/16395/PG_9.1_201105231/16400/33044487_vm [2016-05-05 20:09:00 UTC] DETAIL: Multiple failures --- write error might be permanent. [2016-05-05 20:09:00 UTC] LOCATION: AbortBufferIO, bufmgr.c:2799 Cheers, Rob. -- Sent via pgsql-admin mailing list (pgsql-admin@xxxxxxxxxxxxxx) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-admin
