List of symptoms:
- Online users complain of slow client performance while browsing through the folders
- Online users complain of slow client performance while searching & selecting addresses through the address book
- Online users complain that they cannot login in timely fashion
- Caching and remote users complain that updates do not occur or take a long time.
On POA screen:
1. C/S (TCPHandler) Threads will increase and stay there for a long period of time
2. C/S (TCPHandler) Threads may max out at 99
3. Message Worker Threads will increase and stay there for a long period of time
4. Message Worker Threads may max out at 30
5. File Queues will climb rapidly and process slowly
For performance issues, focus on 820e (Cannot Lock File) errors. It is not necessary that this error occurs. However, they do indicate that the POA has a problem writing to a message store database.
- Bring down the POA and Copy SDD\CLIENT\OFVIEWS from the to the Post Office directory. After copying the OFVIEWS, Rebuild the Post Office Database (do take a backup copy before a rebuild) by connecting to the Domain under which this Post Office resides.
- For 820e errors in the POA log file. Verify the following:
A. Backup is not running
B. Virus scanner is NOT scanning GroupWise Post Office Directories. Virusscanners should never scan GroupWise Post Office directories. This will cause database corruption and poor performance.
- On the Server console---Monitor console---Server Parameters---NCP---Level 2 OpLocks Enabled. It should be set to "OFFâ
- On the Server console---Monitor console---Server Parameters---NCP---Client File Caching Enabled. It should be set to "OFFâ
- In ConsoleOne---POA properties---GroupWise---Agent Setting. Check the Thread settings.
Ideal settings: (for around 1000 users)
Message Handler Threads: 15
TCP Handler Threads: 30
Max. Physical Connections: 2048
Max. App. Connections: 4084
Max. Thread Usage for Priming & Moves: 80 (default value is 20 but if most of the users are using caching/remote mail box than increase this value to 80)
6. In ConsoleOne---POA properties---GroupWise---QuickFinder. Check the QuickFinder Interval, it should be during night hours and once in 24 hours.
- If needed, apply the latest TCP & WinSock modules on the server.
- If the above does not resolve the issue than check the following on the server:
;--- Undocumented TCP settings to improve GroupWise large file attachments ---
# For a GW server that supports a POA and Windows clients, set Minshall, Nagle and Delayed ACK to ON. TCH
Set TCP minshall algorithm = on
Set TCP nagle algorithm = on
# For a GW Webaccess or GWIA server or X-Plat clients, set Minshall, Nagle and Delayed ACK to off. TCH
;--- KB 10013542 increases small ECBS pool to avoid TCP hang ---
Set TCP IP maximum small ECBS = 65534
;---- Files and connections for Concurrent connections ----
Set maximum concurrent opens per file and connection = 1000000
;--- Common File System Parameter Category ---
Set maximum concurrent directory cache writes = 500
;--- Communications Parameter Category ---
Set maximum pending TCP connection requests = 4096
Set TCP defend land attacks = on
Set maximum packet receive buffers = 10000
Set minimum packet receive buffers = 3000
Set new packet receive buffer wait time = 0.1
Set Maximum Interrupt Events = 50
;--- This section is specific for Novell GroupWise (any version) ---
; There are 4 states a connection can be in: FIN_WAIT1, FIN_WAIT2, CLOSING,
; and TIME_WAIT, this setting will force close a connection if there are
; more than 20 in any of these 4 states.
; Reducing this setting will limit exposure to TCP Fin attacks.
; The default is 0, which means unlimited.
; Also fixes GWINTER connections to get cleared.
; This is a NW 6.x only setting
; See TIDs 10071321 and 10073428
Set maximum wait states = 20
;--- Directory Services Parameter Category ---
# If this server is a busy GW server, Novell recommend that it does not host DS replicas.