Discussion Locked

Exchange blues

0 helpful
3 replies
Closed to new repliesThis discussion is closed. It stays readable, but it takes no new replies.

I am running Exchange 5.5 SP/4 on NT 4.0 SP/6
Ten days ago I had to rebuild the private IS.
I used Exmerge to dump the mailboxes to .psts
and then after starting IS with a new private
store, I ran Exmerge to pull the .psts back into the IS. All the online defrags etc. ran
without a hitch for 10 days. The first sign of a problem was Event 1104 'Background cleanup' error on an attachment. Since that
event, the online defrag generates Event 184
with an error that it terminated prematurelywith a error -1019. These same sorts of problems is what initially led me to rebuild
the private IS, and run Exmerge.

What I would like to know is, what typically
causes the reoccurence of these kinds of problems? Is it an indication that the server
hardware is experiencing problems (nothing
in the event log is indicating that), or
could it be some fault in the Exchange
installation, or some other possibility?

The first time a serious problem was
encountered was mid-December of 2000, then
it reoccured late March of 2001, and now
just ten days later. Each time we had to
remove the priv.edb and allow Exchange to
recreate it, and then use Exmerge to export
and import.

Your insight would be greatly appreciated.

3 replies

by mckaytech

The -1019 error indicates an uninitialized page in the database. It's an interesting error in that NTBackup will still back up a database with this error, but defragging will choke on it.

Generally, my experience has been that the scenario you've painted has two possible causes: either hardware trouble with your disk subsystem or a software conflict with a low-level program that does direct disk reads (virus checking and open-file backup agents are the two prime suspects in that category).
In one recent experience with a similar problem on a Compaq server, we found that the SNMP agents on the storage system were reporting anomalies with two of the three drives in the RAID array and since those drives were replaced, there has been nofurther problem. But there was nothing in the Event Viewer or anywhere other than the hardware level SNMP agents to give a clue.

From your description, I'm guessing that it might be a similar hardware problem in your case.

paul

This discussion is closed to new replies.