[Pgsql-ayuda] Base de datos corrupta
Víctor R. Ruiz
rvr@infoastro.com
Tue, 23 Sep 2003 14:58:48 +0200
Hola:
El Martes, 23 de Septiembre de 2003 16:44, Guillermo Schulman tecleó:
> Existe la posibilidad de tener una base de datos corrupta en postgres? En
> tal caso, cómo se puede detectar?
Pues justamente hoy tengo esto en mis logs:
Sep 23 07:26:18 postgres[22572]: [1-1] LOG: database system was interrupted
being in recovery at 2003-09-23 07:25:22 PDT
Sep 23 07:26:18 postgres[22572]: [1-2] ^IThis probably means that some data
blocks are corrupted
Sep 23 07:26:18 postgres[22572]: [1-3] ^Iand you will have to use the last
backup for recovery.
Sep 23 07:26:18 postgres[22572]: [2] LOG: checkpoint record is at 0/304C595C
Sep 23 07:26:18 postgres[22572]: [3] LOG: redo record is at 0/304BF58C; undo
record is at 0/0; shutdown FALSE
Sep 23 07:26:18 postgres[22572]: [4] LOG: next transaction id: 117016205;
next oid: 126474
Sep 23 07:26:18 postgres[22572]: [5] LOG: database system was not properly
shut down; automatic recovery in progress
Sep 23 07:26:18 postgres[22572]: [6] PANIC: ReadRecord: record with zero
length at 0/304BF58C
Sep 23 07:26:18 postgres[22569]: [1] LOG: startup process (pid 22572) was
terminated by signal 6
Sep 23 07:26:18 postgres[22569]: [2] LOG: aborting startup due to startup
process failure
Y documentándome estoy a ver cómo lo recupero :)
Saludetes,
--
Víctor R. Ruiz <rvr@infoastro.com> | - Todos estos momentos se perderán
Centro de Astrobiología CSIC-INTA | como lágrimas en la lluvia