Re: Help - corruption issue?

2011/4/18 Phoenix Kiula <phoenix(dot)kiula(at)gmail(dot)com>:
> Thanks Filip.
>
> I know which table it is. It's my largest table with over 125 million rows.
>
> All the others are less than 100,000 rows. Most are in fact less than 25,000.
>
> Now, which specific part of the table is corrupted -- if it is row
> data, then can I dump specific parts of that table? How? Pg_dumpall
> does not seem to have an option to have a "WHERE" clause?
>
> If the lead index is corrupt, then issuing a reindex should work. So I
> disconnected all other users. The DB was doing nothing. And then I
> started a psql session and issued the command "reindex database MYDB".
> After 3 hours, I see this error:
>
>
>
> [QUOTE]
> server closed the connection unexpectedly
>        This probably means the server terminated abnormally
>        before or while processing the request.
> The connection to the server was lost. Attempting reset: WARNING:
> terminating connection because of crash of another server process
> DETAIL:  The postmaster has commanded this server process to roll back
> the current transaction and exit, because another server process
> exited abnormally and possibly corrupted shared memory.
> HINT:  In a moment you should be able to reconnect to the database and
> repeat your command.
> Failed.
> !>
> [/UNQUOTE]
>
>
> What am I to do now? Even reindex is not working. I can try to drop
> indexes and create them again. Will that help?

it might. take a full file system backup first and drop the indexes.
before recreating them, take a regular dump (with pg_dump) and if it
goes through, you're golden, rebuild the indexes, *update the
postmaster to latest 8.2*, and you can go back online.

merllin