Fabien Fischer <F.fischer@coneco-gmbh.de> writes:
> DEBUG: database system was interrupted being in recovery at 2002-11-07
> 14:12:46 CET
> This probably means that some data blocks are corrupted
> and you will have to use the last backup for recovery.
> DEBUG: invalid resource manager id in primary checkpoint record
> DEBUG: using previous checkpoint record at 0/7039D94
> DEBUG: redo record is at 0/7039D94; undo record is at 0/0; shutdown TRUE
> DEBUG: next transaction id: 473535; next oid: 166062
> DEBUG: database system was not properly shut down; automatic recovery in
> progress
> DEBUG: redo starts at 0/7039DD4
> FATAL 2: btree_insert_redo: failed to add item
> DEBUG: startup process (pid 1805) exited with exit code 2
> DEBUG: aborting startup due to startup process failure
Looks like a mess :-( What led up to this? System crash maybe?
It would be interesting to study your WAL logs and try to understand
what happened, but unless you want to let someone into your machine to
poke around, or donate a tarball of your whole $PGDATA tree for study,
we pnobably can't learn much.
As far as getting back on the air is concerned: I'd suggest running
contrib/pg_resetxlog and then doing a pg_dumpall, initdb, restore.
> My Linux-version is :
> -Linux version 2.4.2-2 (root@porky.devel.redhat.com)
> (gcc version 2.96 20000731 (Red Hat Linux 7.1 2.96-79))
> My Postgresql version is the binary:
> - postgresql 7.2
I'd suggest updating to PG 7.2.3. RHL 7.1 is a tad obsolete as well.
regards, tom lane
