> The problem has occurred a couple more times. Here is the information
> from the error log. I spent some time looking at the trace file, but I
> didn't see anything that looked like a serious problem. I would like to
> send along the trace file, but it is too large. Is there anyway to upload
> it?
>
> 2006-03-29 13:55:14 25088 ERR 54111 MOVE UpdMap.1: 1[11]->387[395]
> len:11
> 2006-03-29 13:55:14 25088 ERR 8 Admin ERROR 'log_error' CAUSED
> EMERGENCY SHUTDOWN
>
>
> Thanks,
> Ryan
>
> > -----Original Message-----
> > From: Ryan Warnick [mailto:[EMAIL PROTECTED]
> > Sent: Wednesday, March 29, 2006 6:10 AM
> > To: [EMAIL PROTECTED]
> > Subject: EMERGENCY SHUTDOWN
> >
> > In the last several hours, our database has shutdown twice.
> > Here is all of
> > the information that I can find in the log files:
> >
> > 2006-03-28 17:20:29 7773 ERR 54111 MOVE UpdMap.1: 1[17]->371[385]
> > len:17
> > 2006-03-28 17:20:29 7773 ERR 8 Admin ERROR 'log_error' CAUSED
> > EMERGENCY SHUTDOWN
> > 2006-03-28 17:21:38 ___ Stopping GMT
> > 2006-03-29
> > 00:21:38 7.5.0 Build 031-121-104-253
> > 2006-03-28 17:29:17 --- Starting GMT
> > 2006-03-29
> > 00:29:17 7.5.0 Build 031-121-104-253
> > 2006-03-28 17:53:42 ___ Stopping GMT
> > 2006-03-29
> > 00:53:42 7.5.0 Build 031-121-104-253
> > 2006-03-28 17:53:45 --- Starting GMT
> > 2006-03-29
> > 00:53:45 7.5.0 Build 031-121-104-253
> > 2006-03-28 19:45:15 19151 ERR 54111 MOVE UpdMap.1: 1[17]->371[385]
> > len:17
> > 2006-03-28 19:45:15 19151 ERR 8 Admin ERROR 'log_error' CAUSED
> > EMERGENCY SHUTDOWN
> >
> > It looks like it was the same problem in both cases. Is
> > there anything that
> > I can do to collect more information on what is happening?
>
>
> Hi Ryan,
>
> the error message means that some memory shall be copied into a region
> without sufficient size. I suppose that this happens while writing
> a undo or redo log entry.
>
> Please generate a kernel trace protocol (either with the DBMGui or the
> dbmcli command "trace_prot abkx". The generated testfile is located
> in the rundirectory of your database and is named with the name of
> your database and the ending ".prt". Please check if some additional
> information has been written into this file.
>
> To collect some more information at the point of the next error it
> would be fine if you could switch two trace-options on. The dbmcli
> commands are as follows:
> trace_on default
> trace_on TOPICLOGTRANS 6
> When the error occurs again please generate again a kernel trace protocol
> in order to get the additional information.
>
> Kind regards, Martin
>
