PostgreSQL Administration - Disaster recovery: pg

Author

Disaster recovery: pg_class

Enrico

2006-10-25, 8:24 am

After a system crash, I have this error:

SELECT * from mytable;
WARNING: could not write block 14 of 1663/1743558/1259
DETAIL: Multiple failures --- write error may be permanent.
ERROR: xlog flush request 4/D43E0A00 is not satisfied --- flushed only to 4/D42700F0
CONTEXT: writing block 14 of relation 1663/1743558/1

I searched for 1259 oid, and I found that 1259 links to pg_class table.
Is there any way to restore my pg_class table?

Thanks in advantage.

Enrico

--
If Bill Gates had a penny for everytime Windows crashed,he'd be a multi-billionaire by now .......oh look, he already is !!!!
scotty@linuxtime.it - Skype:sscotty71
http://www.linuxtime.it/enricopirozzi

---------------------------(end of broadcast)---------------------------
TIP 9: In versions below 8.0, the planner will ignore your desire to
choose an index scan if your joining column's datatypes do not
match

Tom Lane

2006-10-25, 8:24 am

Enrico <scotty@linuxtime.it> writes:
> After a system crash, I have this error:
> SELECT * from mytable;
> WARNING: could not write block 14 of 1663/1743558/1259
> DETAIL: Multiple failures --- write error may be permanent.
> ERROR: xlog flush request 4/D43E0A00 is not satisfied --- flushed only to 4/D42700F0
> CONTEXT: writing block 14 of relation 1663/1743558/1

I'm not convinced there's anything much there you need to fix. You
should look into why the filesystem seems to have lost part of your WAL
file --- that strongly suggests that the disk or kernel is lying about
write ordering, which will cause you problems eventually. But this
message per se will go away as soon as the WAL endpoint moves another
1.5Mb. Unless there's other corruption elsewhere you'll be fine.

regards, tom lane

---------------------------(end of broadcast)---------------------------
TIP 9: In versions below 8.0, the planner will ignore your desire to
choose an index scan if your joining column's datatypes do not
match

Enrico

2006-10-25, 8:24 am

> I'm not convinced there's anything much there you need to fix. You
> should look into why the filesystem seems to have lost part of your WAL
> file --- that strongly suggests that the disk or kernel is lying about
> write ordering, which will cause you problems eventually. But this
> message per se will go away as soon as the WAL endpoint moves another
> 1.5Mb. Unless there's other corruption elsewhere you'll be fine.
>
> regards, tom lane

Hi Tom,
I think that the system crash is due to a ram failure,
my server doesn't start if is present a specific bank of ram,
and without it, it works fine.

I watched logs and in the /var/mail/postgres and I've seen this error that matchs in time with system crash.

pg_dump: ERROR: invalid memory alloc request size 18446744073709551613

regards
Enrico

--
If Bill Gates had a penny for everytime Windows crashed,he'd be a multi-billionaire by now .......oh look, he already is !!!!
scotty@linuxtime.it - Skype:sscotty71
http://www.linuxtime.it/enricopirozzi

---------------------------(end of broadcast)---------------------------
TIP 4: Have you searched our list archives?

http://archives.postgresql.org

Sponsored Links