Video Screencast Help
Search Video Help Close Back
to help
Not able to make it to Vision this year? Get a sampling in the Best of Vision on Demand group.

Journal archiving very slow

Created: 27 Jan 2011 | 7 comments
Bekir's picture
0 0 Votes
Login to vote

Hello all,

We've started Journal archiving 2 weeks ago. Initially we had 500.000 mails in the journal mailbox. EV (9.0.1) kept archiving for 2 days until 140.000 mails left. Then it somewhat stopped. We couldn't get journaling going on with that speed again.

Then we decided to stop journaling to that mailbox and created a new journal mailbox. Now we are journaling to that new mailbox, and wierdly EV can archive that new journal without any problems (almost 50.000 mails a day).

Back to the problem, the old journal mailbox now has around 128.000 messages and it is being archived with a speed of 2 mails per minute (!?!) which is so slow...

Traces look normal. But the task itself is not creating any reports when it is run in report mode.

We looked at the message classes on an Outlook, and all the messages are in IPM.Node.EnterpriseVault.PendingArchive.part class.

What can we do to finish archiving this mailbox? What can be the problem?

Discussion Filed Under:

Comments

JesusWept2's picture
27
Jan
2011
0 Votes 0
Login to vote

have you gotten the recent

have you gotten the recent journaling hotfix for EV9 SP1?
Also what version of outlook is on the server?

WiTSend's picture
27
Jan
2011
0 Votes 0
Login to vote

To recap: 1st Journal mailbox

To recap:

1st Journal mailbox archiving very slowly

2nd Journal mailbox archiving normally.

Numerous items still in 1st mailbox, with "pending" archive shortcuts.

 

Resolution:

1)   check to see if any of the "pending" items are actually already archived:

      a) if they are delete the pending shortcuts

      b)  if not use the "Cancel operation" button and clear the pending status

2)  Stop the 1st Journal archive task, move the remaining items from the 1st journal mailbox to the 2nd, remove the 1st Journal archive task.

 

You could spend a lot of cycles trying to figure out "why", or you can resolve the issue since you have a archive task running within normal parameters with minimal impact.

JesusWept2's picture
27
Jan
2011
0 Votes 0
Login to vote
EV Director's picture
27
Jan
2011
0 Votes 0
Login to vote

Hi, I would suggest getting

Hi,

I would suggest getting dtrace of journaltask for same 10 minutes and posting it here.

Mike Bilsborough
Director,Enterprise Vault Engineering Support 

Bekir's picture
28
Jan
2011
0 Votes 0
Login to vote

Hi, thanks for the replies. I

Hi, thanks for the replies.

I applied the hotfix, it didn't help so far. So I've done some research and cleared the message class type of the messages in the mailbox with this script:

--------------------------------

Sub Item_Open

   ' Change the following line to your new Message Class
   ' NewMC = "IPM.Note.EnterpriseVault.PendingArchive.Part"
   NewMC = "IPM.Note"

   Set CurFolder = Application.ActiveExplorer.CurrentFolder
   Set AllItems = CurFolder.Items
   NumItems = CurFolder.Items.Count

   ' Loop through all of the items in the folder
   For I = 1 to NumItems

      Set CurItem = AllItems.Item(I)

      ' Test to see if the Message Class needs to be changed
      If CurItem.MessageClass <> NewMC Then

         ' Change the Message Class
         CurItem.MessageClass = NewMC

         ' Save the changed item
         CurItem.Save

      End If

   Next

   MsgBox "Done."

End Sub

--------------------------------

Then I ran the journal archiving task again. It started marking all the messages again. After that process EV archived another 30.000 messages within an hour.

However, it's gone back to that slow state again. What do you think about this?

Nothing is pending by the way. I'll run a trace and upload.

Best regards,
Bekir Burak Durmaz

Bekir's picture
28
Jan
2011
0 Votes 0
Login to vote

oulook version is 2007

oulook version is 2007

Best regards,
Bekir Burak Durmaz

Bekir's picture
02
Feb
2011
0 Votes 0
Login to vote

It looks like we have an MSMQ

It looks like we have an MSMQ pool size issue. Default 1GB was not enough, we increased it to 8GB.

We'll see if it resolves the issue.

Best regards,
Bekir Burak Durmaz