i have the following problem:
Outlook updates mails with delays from 2minutes to 4hours (send/receive).
2 Exchange Server 2010 Servicepack 1 Rollup1 (Windows 2008 R2 Ent)
400 Outlook 2010 Clients in Cached Mode
- one server is not affected
- mails appear in owa (no exchange transport problem)
- Outlook in online mode works
- gpo settings Microsoft Outlook 2010/Account Settings/Exchange/Exchange-Cache-Mode --> wait time in seconds (i don´t think thats the problem because 200 Clients on the other server are working)
- virusscan on the server and client disabled (for the test)
- Exchange perforance troubleshouter shows no errors (one warning: the top6 user account for 30% of the mapi cpu usage on the server)
- if you press send/receive in outlook --> mails appear
- if you press reconnect in outlook connection state --> mails appear
- if i move the mailbox to the other server --> it works
1. Capture a netmon trace to check if there are network issues.
2. Run the profmon to capture the performance of the problematic server:
Comparing the counters to see if there are any bottlenecks.
thanks for your answer.
1. can´t find any network issues
2. ran Exchange_2010_Perfwiz-MBX.xml:
Description / Check 1 (very short) / Check 2 (1 hour) / Microsoft threshold
Memory\Transition Pages Repurposed/sec 3466/54/spikes less than 1000
Memory\Page Reads/sec 134/9,7/less than 100avg
Memory\Pages/Sec 3.483/116/less than 1000avg
Memory\Pages Input/sec 3.257/55/less than 1000avg
the other values seem to be ok. do you think i´ve a memory problem? 8GB RAM for 250 User ?
The symptoms seem to be strange.
Since the issue does not occur in Online Mode, it means the emails has been delivered to the user’s mailbox properly without delay. The issue is only with the cached mode feature in Outlook and it seems it does not retrieve or sync with the Exchange Server. However, according the other symptom that the issue does not occur when moving the user to another MBX server, it indicates a Exchange side issue.
How about creating a new mailbox database on the problematic MBX server and create a test user in the new database to check if the issue persist.
Which server did you change? CAS/HT or Mailbox? I've got the same issue. I looked at my resource utilization and memory and CPU are very low, but we don't have any VM Resource reservations. Just wondering.
BTW, we've tested every aspect, it just seems to be Cached Mode Client receive, sending messages is instantaneous. Could it be the receive queue?