Re: Help with database recovery ...

On 02/05/11 03:32, Iztok Stotl wrote:
> My database crashed and server won't start ...
>
> ------------------------------------------------------
> LOG: database system was interrupted while in recovery at 2011-05-01
> 19:31:37 CEST
> HINT: This probably means that some data is corrupted and you will have
> to use the last backup for recovery.
> LOG: checkpoint record is at 2/D41C1350
> LOG: redo record is at 2/D41C1350; undo record is at 0/0; shutdown
> FALSE
> LOG: next transaction ID: 752151177; next OID: 50716
> LOG: next MultiXactId: 1; next MultiXactOffset: 0
> LOG: database system was not properly shut down; automatic recovery in
> progress
> LOG: redo starts at 2/D41C1394
> PANIC: could not access status of transaction 752150401
> DETAIL: could not read from file "pg_clog/02CD" at offset 73728:
> Success
> LOG: startup process (PID 3504) was terminated by signal 6
> LOG: aborting startup due to startup process failure
> LOG: logger shutting down
> ---------------------------------------------------------------
> What can I do ?

This shouldn't happen unless something broke badly on your system. Check
your disks and file systems.

Before trying to fix anything, make a complete copy of the entire
postgresql directory, the one that contains the 'pg_xlog', 'pg_clog'
folders etc. Store that copy on a removable hard drive or some other
media that's not connected to the system you're recovering.

This message is very weird: "could not read from file "pg_clog/02CD" at
offset 73728: Success". What does 'ls -l pg_clog/' say?

--
Craig Ringer