Re: Alert: Thread 1 cannot allocate new log

This is not a harmless message although it does not imply loss of data. What it points to is loss of performance. DBWR will suspend until the situation has resolved.

The log entries are being held up by the system not having finished with the next redo log in line.
This either means that ARCH process has not finished with it if in Archive Log Mode or the updates are occuring too fast and wrapping round the logs.

Simple solution would be to increase the size of you redo logs and/or increase their number.

Regards

--
Kevin A Lewis (BOCM PAULS LTD - Animal Feed Manufacturer - Ipswich England)

                        <KevinALewis_at_HotMail.com>

The views expressed herein by the author of this document are not necessarily those of BOCM PAULS Ltd. Bryan M. Kramer <kramer_at_techne.ca> wrote in message news:37B18C3B.F69C253A_at_techne.ca...
> I'm running oracle 8 on an NT server.
>
> I have a system that is mysteriously and not repeatably missing
> records in the database that we expect to see after a run of the
> complete application. Running the component individually seems to
> indicate
> that there is no problem with the application. The only anomoly is that
>
> Current log# 5 seq# 258490 mem# 0:
> /oracle06/oradata/dev2/redodev205.log
> Thread 1 cannot allocate new log, sequence 258491
> Checkpoint not complete
>
> messages are appearing in the alert log. A search of deja news shows
> some
> discussion on this topic which imply that the message is a harmless
> warning.
>
> Is this really the case? Is it possible that the inserts and updates are
> NOT being
> committed for some reason?
>
> Thanks
> --
> Bryan M. Kramer, Ph.D., Software Architect
> Techné Knowledge Systems Inc.
> 439 University Ave., Suite 900, Toronto ON M5G 1Y8
> 416-977-3960 fax 416-977-9902 www.techne.ca
>
>
Received on Wed Aug 11 1999 - 10:11:05 CDT