From: | Björn Metzdorf <bm(at)turtle-entertainment(dot)de> |
---|---|
To: | "Tom Lane" <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
Cc: | <pgsql-admin(at)postgresql(dot)org> |
Subject: | Re: Uninitialized page - fixing |
Date: | 2003-06-08 23:49:44 |
Message-ID: | 00a901c32e18$a3b189c0$0564a8c0@shock |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-admin |
>> recently we are suffering from repeated
>> WARNING: Rel pg_toast_77798203: Uninitialized page 25489 - fixing
>
> Hmm. There is a known mechanism for that to happen in 7.2 and before,
> but not in 7.3. Can you see any pattern in the messages?
Here is another example:
2003-06-05 21:27:08 WARNING: Rel psessions: Uninitialized page 185 - fixing
psessions is often read from and written to, but it is not very big
(thousands, perhaps tenthousands of rows) and it is regularly vacuumed
(every 5 to 15 minutes). These messages appear under heavy load only (more
than 700 concurrent connections to the backend).
Since then (2003-06-05) we didn't get this message again.
I cannot see any pattern there, but at least with our old server, those
messages came just ahead some hardware problems (pg_clog problems, tables
were gone). But our new server right now is running very smoothly.
I will upgrade to 7.3.3 as soon as I find some time.
Regards,
Bjoern
From | Date | Subject | |
---|---|---|---|
Next Message | Bruce Momjian | 2003-06-09 01:21:56 | Re: Diagnose Database Corruption |
Previous Message | Sean Chittenden | 2003-06-08 18:14:46 | Re: No acceptable C compiler found |