instance goes down when archiver cannot keep up

From: prunoki <hegyvari_at_ardents.hu>
Date: 19 Oct 2005 06:21:24 -0700
Message-ID: <1129728084.246874.291670@f14g2000cwb.googlegroups.com>

Hi,

I have a 8.1.7.4 database on Windows 2000. 8 redo logs, size 10M each. When I issue an

alter table add new_column number(10) default 0 not null

on a table with more then 10M rows, I notice that the archiver cannot keep up with the redo generated. Oracle starts new archiver processes and after a minute or so the instance terminates with

error 473 detected in background process ORA-00473: ARCH process terminated with error

in several trace files (ckpt, lgwr, ...) I could not read any further information from anywhere else. From my understanding the database should have been frozen until one redo log became free again.

Any ideas why this happens and how to avoid it?

Regards,

Hegyvari Krisztian Received on Wed Oct 19 2005 - 08:21:24 CDT