Re: archivelog leads to crashes...?

On Mon, 15 Mar 2004 19:51:00 +0100, Andrew Metcalfe wrote (in message <a2588ec3.0403151051.749478bf_at_posting.google.com>):

> Last night I started a project to get hot backups working on my 8.1.7
> production server.
>
> After setting the database to archivelog, the instance would crash and
> refuse connections. I found the following in my alert.log file:
>
> alter database mount
> Mon Mar 15 00:16:39 2004
> Successful mount of redo thread 1, with mount id 1284634983.
> Mon Mar 15 00:16:39 2004
> Database mounted in Exclusive Mode.
> Completed: alter database mount
> Mon Mar 15 00:17:04 2004
> alter database archivelog
>
> Mon Mar 15 00:17:04 2004
> Completed: alter database archivelog
> Mon Mar 15 00:17:14 2004
> alter database open
>
> Mon Mar 15 00:17:15 2004
> Thread 1 opened at log sequence 47258
> Current log# 2 seq# 47258 mem# 0: E:\ORACLE\ORADATA\DB\REDO02.LOG
> Successful open of redo thread 1.
> Mon Mar 15 00:17:15 2004
> SMON: enabling cache recovery
> SMON: enabling tx recovery
> Mon Mar 15 00:17:23 2004
> Completed: alter database open
> Mon Mar 15 00:18:11 2004
> Thread 1 advanced to log sequence 47259
> Current log# 3 seq# 47259 mem# 0: E:\ORACLE\ORADATA\DB\REDO03.LOG
> Mon Mar 15 00:19:05 2004
> Thread 1 advanced to log sequence 47260
> Current log# 1 seq# 47260 mem# 0: E:\ORACLE\ORADATA\DB\REDO01.LOG
> Mon Mar 15 00:19:56 2004
> Thread 1 cannot allocate new log, sequence 47261
> All online logs needed archiving
> Current log# 1 seq# 47260 mem# 0: E:\ORACLE\ORADATA\DB\REDO01.LOG
> Mon Mar 15 00:39:12 2004
> Shutting down instance (immediate)
>
> Looking online, the culprit here is usually disk space. I have plenty
> of disk space on all disks on this server... so its not disk space.
> E: is a single fast large physical disk without any RAID.
>
> These are the steps I used to enable archivelog.
>
> svrmgrl
>
> connect internal
>
> shutdown immediate
>
> startup mount
>
> alter database archivelog;
>
> alter database open;
>
> Everything appeared to work properly. Upon completion, I could
> immediately connect, but after a very brief period I could no longer
> connect and my applications would hang.
>
> My intention is to use the Management Console to schedule nightly
> backups, but I haven't done that yet.
>
> _Am

How about
alter system archivelog start;

-- 
With kind regards,
Ronald
http://ronr.nl/unix-dba