[jibi]

Okay, so I've been working with RIM for a bit now on an issue I'm having. Basically, only one user out of 15 total on the Test server are receiving state updates. This includes items that are deleted from Outlook as well as items that are marked as read in Outlook - neither are replicating the changes to the handhelds for the other 14 users.

Wireless Calendar synchronization and Wireless email redirection both work, as do Enterprise Activation, Wireless Contact Synchronization, etc.

In the MAGT logs, every 12 minutes, it shows a Stopping and Starting of the 14 affected users, due to a change in their mailbox. When you run 'HandheldCleanup.exe -u', it receives the following for those 14 users (I have bolded what I changed for anonymity):

Change detected for:
Current DisplayName: Last, First
Previous Display Name: Last, First
Current Server DN: /o=Enterprise/ou=Domain/cn=Configuration/cn=Servers/cn=
SERVERNAME/cn=Microsoft Private MDB
Previous Server DN: /o=Enterprise/ou=Domain/cn=Configuration/cn=Servers/cn=SERVERNAME/cn=Microsoft Private MDB

The actual log entry in the MAGT log is something like:

[30046] (03/25 00:54:20):{0x1008} ScanGAL():Detected a mailbox move for user: Last, First
[40448] (03/25 00:54:20):{0x1008} Adding a new stopper thread to have 6 in total
[40000] (03/25 00:54:20):{0x500} [DIAG] EVENT=Register_thread, THREADID=0x500, THREADNAME="StopperThread"
[30045] (03/25 00:54:20):{0x500} Stopping handheld for Last, First
[30000] (03/25 00:54:20):{0x10EC} {mailbox@domain.com} Sending data to Dispatcher, size=23, Tag=143
[40000] (03/25 00:54:20):{0x10EC} {mailbox@domain.com} SendToDispatcher, Tag=143
[30070] (03/25 00:54:20):{0x10EC} User mailbox@domain.com shutting down...
[40000] (03/25 00:54:20):{0x1040} [BIPP] Send data, Tag=143
[40000] (03/25 00:54:20):{0x103C} [BIPP] Received status DELIVERED, Tag=143
[40645] (03/25 00:54:20):{0x103C} [BIPP] Failed to find user for received status, UserId=39, Tag=143
[30041] (03/25 00:54:24):{0x1008} Starting handheld for Last, First
[40704] (03/25 00:54:24):{0x1008} MAPIMailbox::MAPIMailbox(2) ServerDN=/o=Enterprise/ou=Domain/cn=Configuration/cn=Servers/cn=SERVERNAME/cn=Microsoft Private MDB, MailboxDN=/o=Enterprise/ou=Domain/cn=Recipients/cn=mailbox
[40739] (03/25 00:54:24):{0x1008} MAPIMailbox::MAPIMailbox(2) - ResolveName[1] success for DisplayName='/o=Enterprise/ou=Domain/cn=Recipients/cn=mailbox'.
[40430] (03/25 00:54:24):{0x1008} {mailbox@domain.com} MAPIMailbox::MAPIMailbox(2) - Mailbox opened on session 0x013b2678 (SERVERNAME)
[30068] (03/25 00:54:24):{0x1008} User mailbox@domain.com starting up...
[30069] (03/25 00:54:24):{0x1008} {mailbox@domain.com} Mailbox name is mailbox
[40221] (03/25 00:54:24):{0x1008} {mailbox@domain.com} In Cradle timer is 705000

So far, I have dropped the database and added a new one, adding back the users (but keeping their information to prevent recradling). I have also completely removed myself from the server and readded without saving my Exchange information (so I reactivated wirelessly) and still having the issue. I have uninstalled Exchange System Manager and reinstalled it to the C:\ drive (was installed on D:\, along with BES). I have uninstalled and reinstalled BES 4.0 w/o HotFixes (although this problem has existed since I installed BES 4.0), during the same time of uninstalling/reinstalling ESM.

I'm not really sure what else can be done. RIM says they've never seen the issue before. The ScanGAL function is detecting a change for something thats not changing (as seen in the above log of HandheldCleanup.exe). The HHC.exe also shows that this is happening at ALL times not just during the 12 minute server-side ScanGAL interval.

*Note: Forgot to add, when a user that is affected from the above symptons does a 'Reconcile Now' on their handheld, the state of the messages are updated as they should.

Any ideas?

[jibi]

Okay, in case anyone else ever has this issue in the future...

We went from Exchange 5.5 to Exchange 2000 about 9 months ago. We used Aelita's Exchange Migration utility to handle the migration. A few months after this, we also set a Global Address List policy from nothing (which left the Display Names in a 'First Last' form) to 'Last, First' (which added a space at the end of the Display Name as well). During one of the changes, either the migration or policy, the Display Name became, what seemed, encrypted (if viewed through an LDIFDE dump of the AD; it was visible as normal within the AD properties though).

I went into AD and removed the space at the end of the Display Name, which then made it readable in the LDIFDE dump and removed it from the HandheldCleanup read-out. All users are now receiving state updates/sychronizes within the wherever-defined 20 minute interval.

[bfrye]

Wow, I was just reading some knowledge center articles on the tsupport site and I saw a known issue with names with spaces in them.

[jibi]

i am so glad that i check those articles (which one was it?).

...but that was honestly the LAST thing i would think to check when troubleshooting this sort of issue.

Also, a few more things. Our production environment is still BES 3.6, so I had moved myself to it as part of the troubleshooting. The message state synchronization worked fine (within a minute on my tests). This just added to the confusion and differences between 3.6 and 4.0.

p.s-the tech said that he had tested with spaces and had no issues. odd.

p.p.s-go check the other site, if you get a minute. i feel like i'm talking to a wall.