cancel
Showing results for 
Search instead for 
Did you mean: 

Event 3419 and 2217, EV9 + Exchange 2010

FreKac2
Level 6
Partner Accredited Certified

Hi all,

 

Has anyone seen the 3419 event in combination with 2217 ?

These events pop up most of the time every minute (kind of logical with the dispenser sleep time), sometimes it could be 5-10minutes in between two of these events.

 

Event ID: 3419; Archive Task

There was a problem accessing an Exchange 2010 or later server. The dispenser will re-queue the current item and sleep for 1 minute(s).

Task: Exchange Mailbox Archiving Task for <Customer Exchange Server> 

 

Event ID: 2217; Archive Task

Message dispenser will resume processing

Task: Exchange Mailbox Archiving Task for <Customer Exchange Server>

 

Did a dtrace with of the archive tasks, used trigger/watch, let me know if you want to see more of the log.

As far as I know there haven't been any DAG failover but can't really check it myself right now.

 

70589 18:39:57.586  [2392] (ArchiveTask) <3484> EV:M MAPI error detected for an Exchange 2010 or later target (possible DAG failover) - 0x8004011D
70590 18:39:57.586  [2392] (ArchiveTask) <3484> EV:H :AgentMessageDispenser::ActivateObject() |Exiting routine at point A |
70591 18:39:57.586  [2392] (ArchiveTask) <3484> EV:M :AgentMessageDispenser::ProcessNextMessage() |ActivateObject has returned failure |
70592 18:39:57.586  [2392] (ArchiveTask) <3484> EV:M AgentMessageDispenser::ProcessNextMessage - AGENTS_E_DISPSLEEPNOREPOST, agent will sleep without reposting the current message
70593 18:39:57.633  [2392] (ArchiveTask) <3484> EV:M AgentMessageDispenser: 1 of 1 Agent threads is pausing|
70594 18:39:57.633  [2392] (ArchiveTask) <3484> EV:M AgentMessageDispenser::LogPauseReason - First Dispenser Thread to Sleep
70595 18:39:57.633  [2392] (ArchiveTask) <3484> EV~W Event ID: 3419 There was a problem accessing an Exchange 2010 or later server.  The dispenser will re-queue the current item and sleep for 1 minute(s). |Task: Exchange Mailbox Archiving Task for CMNMS10 |

1 ACCEPTED SOLUTION

Accepted Solutions

JesusWept3
Level 6
Partner Accredited Certified

Well it depends, are you really really seeing that error on the CAS servers? because if you're not, then its not a throttling issue but something else (possibly GC connectivity or something else in the background)

https://www.linkedin.com/in/alex-allen-turl-07370146

View solution in original post

23 REPLIES 23

JesusWept3
Level 6
Partner Accredited Certified

interesting
suggest giving support a call as this was meant to be fixed
What i'm assuming has happened is that someone was on Exchange A and enabled for EV, a DAG failover occured and put them on Exchange B, and they tried to archive or restore something and it went to Exchange Server A's MSMQ

 

Any chance you could post parts of the dtrace from before this so we can see where they're coming from? would suggest having a look in MSMQ and see if anything is stuck in A2 or R2

When a DAG failover occurs you should be running a prov task and resync

https://www.linkedin.com/in/alex-allen-turl-07370146

FreKac2
Level 6
Partner Accredited Certified

This was during the scheduled archiving run and the Provisioning Task runs 1hr before the scheduled tasks. There were no entries in the A2 or R2 queues. Haven't received an answer yet on the DAG failover but will keep you posted (working remotely :) ).

FreKac2
Level 6
Partner Accredited Certified

Update:

Apparently the Exchange Admins updated content/hotfixes on the Exchange servers yesterday, a couple of reboots and failovers :) (during the archiving window).

So I'll put this thread as solved for the time being.

Cheers

Fredrik

FreKac2
Level 6
Partner Accredited Certified

Seems it was premature to put this as solved as the event's keep popping up, even when there haven't been any failovers.

Will create a case soon but have a couple of questions which may be related but maybe not as well.

If you use the permissions scripts for Exchange 2010 is there anything that will prevent the EV service account to create a outlook profile to one of the "ev service/system mailboxes" ?

Because when I try to do that in Outlook (manually creating a new profile), Outlook states that it can't logon. If I try to access the system mailboxes through the service account's mailbox (starting outlook->open->other users folder) then I can see the mailbox, no complaints from outlook.

If I try to create a profile to a mailbox on a 2007 server there are no issues whatsoever.

From time to time we also get event's in regard to issues loging on and contacting the GC.

The customer don't use any 2008 DC servers, in regard to e.g. NSPI settings ( if that could be the cause for this) but I found these MS technotes/help:

http://technet.microsoft.com/en-us/library/ee332317.aspx

If you look at the bottom of that technote it states that there is a default max of 50 sessions. Not sure if that is enough or if that setting could be an issue here.

http://technet.microsoft.com/en-us/library/ee332346.aspx

So it seems that it's the CAS servers in Exchange 2010 that will "act" as the logon server.

Again not sure if this is/could be an issue but I thought I should drop the questions to you guys, if you have any input.

 

Cheers

Fredrik

biosphere
Level 4
Partner

Just started seeing the same behavior after switching archiving and journaling tasks from old Exchange 2003 to new Exchange 2010 environment..

JesusWept3
Level 6
Partner Accredited Certified
Remember that when setting permissions through the scripts, it must be run as the VSA account which must have it's mailbox on exchange 2010 (the VSA not just the system mailbox) But really being in a down situation like you are at the moment, would suggest calling support
https://www.linkedin.com/in/alex-allen-turl-07370146

FreKac2
Level 6
Partner Accredited Certified

It's not really down but very slow.

I did set the DS Server regkey on Friday to see if there is a change in performance.

From about 3000 mails during three hours, it went up to 90000.

But I just learned that the customer also moved the disks to new SAN storage, so at the moment I'm not sure if it's the regkey or the disks that caused the bad performance.

Will reset the regkey today and see if it goes back to the low figures again.

Also, at least if the manual is correct, the account the runs the permission scripts need to be in the "Organisation Management" role, it doesn't need to be the VSA.

So is the manual incorrect on that or ?

amankahlon
Level 3
Accredited

After every now and then we get an error "There was a problem accessing an Exchange 2010 or later server. The dispenser will re-queue the current item and sleep for 1 minute(s). "

 

3419 followed by 2217

also the queues freeze- especially A6

 

FreKac2
Level 6
Partner Accredited Certified

Forgot about this thread but no, the events still pop up.

The customer installed a new DC in their environment and I then targeted that DC with the "DS Server" regkey.

The problem seemed to have been resolved, but it wasn't.

How no idea why the events stopped but then reappearing a couple of days later.

Still have performance issues, not as severe as before but nowhere near to what it should.

Started a case, sent in traces but all that could be seen was the problems with connecting to the exchange server/GC.

Will open a new case but in regard to performance rather than connection issues.

If anyone want to share their hourly archiving rate I would appreciate it.

My current theory is that it's Outlook 2007 limitations that is the culprit but can't be sure, and since 2003 isn't an option with Exchange 2010 it's a bit problematic to verify.

Jon_S
Level 3

Anyone had any luck with this? Also having the same problem, I'm still testing but there are a couple of things which may fix the issue. There's this:

http://www.symantec.com/business/support/index?page=content&id=TECH139751

If you look at more information it says you can't use closest GC, but you can use DS Server to specify a CAS, not a DC.

And there's also this:

http://www.symantec.com/business/support/index?page=content&id=TECH129173

That says there's a new MS patch to increase the MAPI connection limit for Outlook 2007.

I've tried the 2nd and I don't think that's improved the situation and I've tried specifying the CAS server with the DS Server entry, but we also have one Exchange 2003 server for a remote site that the archiving failed on. I'll have to keep an eye on it and see if it's just that server that fails or if it does it at all again.

JesusWept3
Level 6
Partner Accredited Certified

please don't install the outlook 2007 performance hotfix, it is currently under review by symantec and microsoft and it is advised that you uninstall it as it causes more issues

https://www.linkedin.com/in/alex-allen-turl-07370146

Jon_S
Level 3

I noticed it says it causes issues with the Journal task, but if there are more issues that aren't specified or Symantec haven't tested it enough then it probably shouldn't be mentioned on the tech note at all.

JesusWept3
Level 6
Partner Accredited Certified

well lets be fair here, the performance hotfix was developed and released by Microsoft, not symantec, MS have their own QA, not every configuration however can be certified by Symantec , coupled with environmental issues, work loads and what not, its difficult to certify someone elses hotfix

https://www.linkedin.com/in/alex-allen-turl-07370146

Jon_S
Level 3

I'm not trying to point fingers for the issues it may cause. I'm not blaming them for putting it up as a resolution in the first place, but if it's doing more harm than good, then it should be taken off the symantec article.

FreKac2
Level 6
Partner Accredited Certified

Jon_S, if you look on your CAS server/servers, do you get any events stating that the EV throttling policy is over budget ?

Look at my post in regard to the throttling policy settings and see if you see the same (about the fifth post from the bottom).

https://www-secure.symantec.com/connect/forums/latest-outlook-2007-hotfix-might-it-fix

Jon_S
Level 3

You're right, I did notice that but didn't pursue it any further after seeing that the throttling policy was applied correctly. I've changed the others to $NULL, so I'll update after a few days with the results. I've removed the hotfix and DS Server reg key just so I can narrow down the solution.

Thanks for your help FreKac

Jon_S
Level 3

Tried removing the hotfix and DS Server key (with RPC thresholds changed) and it failed. I then tried to set the DS Server reg key and it still failed, so it only leaves me with the hotfix which isn't really fit for production by the sounds.

Josamito
Level 3
Partner

I had the same problem and i just did an upgrade to Enterprise Valt 9.0.1 and...  problem solved!

You can use the official guide in order to upgrade to 9.0.1 from 9.0

Merv
Level 6
Partner

i'm also have these same errors...Exchange 2010 with 9.02.brand new environment.Any more updates since feb?

possible fixes:

1. DS Server regedit

2. Outlook hotfix

3. EV throttling policy - however i did apply the 9.02 script

4. DAG failovers

Did I miss out anything?

 

<edit>

I just came across this TN which talks about load balancers being the cause of the problem:

http://www.symantec.com/business/support/index?page=content&id=TECH145331&key=50996&basecat=TROUBLES...

will have to investigate with the network team to look this up