Analyze SqlServer 7.0 trace output

In one of our SqlServer 7.0 databases we get the following
error message:
Database AllWork: Transaction Log Backup...
Destination:
[e:\SQL7Data\BACKUP\AllWork\AllWork_tlog_200403290 853.TRN]
[Microsoft SQL-DMO (ODBC SQLState: 42000)] Error 4213:
[Microsoft][ODBC SQL Server Driver][SQL Server]Cannot
allow BACKUP LOG because file 'AllWork' has been subjected
to nonlogged updates and cannot be rolled forward. Perform
a full database, or differential database, backup.
[Microsoft][ODBC SQL Server Driver][SQL Server]Backup or
restore operation terminating abnormally.
After a full database backup, everything is fine for some
time, but then the problem reoccurs.
I managed to capture a full trace using Profiler. At 7:40
the transaction log backup succeeded without errors and at
7:45 the transaction log backup failed with the
errormessage shown above. Allthough the trace is only 5
minutes, it still is over 1.100.000 records. I already
looked for SELECT INTO, WRITETEXT and UPDATETEXT
statements but I can't find any.
How can I determine what causes this error message?
What should I look for in the trace?
Any help would be appreciated,
TIA,
Gerrit van Ham
Gerrit,
Is 'select into/bulk copy' switched on? If so, there may be a bcp fast load taking place. I can't remember if this gets shown in the profiler trace in SQL7 or not - I don't think it does, but you may need to check this. It sounds like a bcp operation is t
aking place with select into/bulk copy switched on.
Mark Allison, SQL Server MVP
http://www.markallison.co.uk