What information is needed to troubleshoot AccuRev performance?


What information should I collect to troubleshoot AccuRev performance issues?

When dealing with any AccuRev performance issue, you always want to collect the following information to help AccuRev support.

1. Has performance always been poor? If not, what changed in the environment?
2. Exact detail on what the user is trying to do? Does performance problem happen to other users on different client hosts? Or is it a common issue on the same platform?
3. Always confirm that if the operation actually finishes? A hang or lock-up can often be mistaken as performance problem.
4. AccuRev version on problem clients and server host?
5. Hardware specs on problem clients and server host?
6. Get complete AccuRev server logs. Default location is .../accurev/storage/site_slice/logs
7a. On the server host, get <accurev diag> output from depot storage directory.
7b. On the client host, get <accurev diag> output from user's work space directory.
7c. It is important to run <accurev diag> at least twice on each affected machine.
8. are you using any triggers? Send output from <accurev show triggers>. Also, the server_admin_trig (server_all_trig pre v3.5) will not be listed in the above output. To verify this trigger is not being used look in .../accurev/storage/site_slice. If it exists you will see a triggers directory under site_slice and the server_admin_trig (server_all_trig pre v3.5) will be in the triggers directory.
9. Collect tasklist.xml output; please see FAQ #106 for more information.
10. Is the performance better from thh cli vs the gui (or vise versa)?
11. Find out where the users home directory is located. If it is stored on a remote system, move it local to the client.

For more information contact AccuRev Support

Comment List
Related Discussions