journaling mailbox recieving duplicate emails

[hospitalcase]

We have a third party mail archiving system which worked fine with exchange 2003, however we have moved to exchange 2010 and while journaling is still working we are getting random duplicate emails into the journaling mailbox. For instance if a mail is addressed to more than one individual within our organisation we get a seperate mail for each of these individuals instead of just one.
This happens randomly accross all users.
Does anyone know of anyway of stopping this so as only to get one email.
This is how the journaling mailbox is working
1. Open the Exchange 2010 Management Console
2. Navigate to “Organization Configuration” and expand view
3. Click Mailbox
4. In the results pane, click the “Database Management” tab and select the Mailbox database on which you want to enable journaling.
5. In the action pane, under the mailbox name, click “Properties” or right click and select “Properties”.
6. In the selected mailbox Properties, click the Maintenance tab.
7. Select the Journal Recipient check box and then click Browse. Locate your MailMeter journal mailbox and select it.
8. Click OK

 

[ntinlin]

You will get duplicates if the two recipients are in different databases.
Bifurcation occurs at the database level in this case so Exchange is doing what it is designed to.

Your archiving system should single instance that when it goes into its database / storage.

Neill

 

[hospitalcase]

Thanks for the reply.
We only have one mailbox database and one public folder database.

 

[ntinlin]

Hmm.

Remember that e-mails now leave the Exchange Store, heads out to the Hub Transport which then sends it back to all the individual addressees. There is no single instancing. Thus there will be multiple copies going to the journal when the messages hit the database.

Again, your archiving util if it is any good should be single instancing the e-mails when it pulls them out of the Journal mailbox so not really an issue.

You should also have your Journal mailbox in a separate database with circular logging enabled as best practise. You'll be generating a LOT of logs.

Neill