Arcserve 7 Abended, now will not Merge

[domburi]

I had a problem with Arcserve not merging a few weeks ago. I deleted the database and replaced it with a blank one. Then I rebooted the server and this fixed the problem.

However, 4 days later, the Netware 6.0sp2 server Abended after a backup job while in the middle of running Prune/Purge. The Abend was:

Server SERVER1 halted Friday, December 27, 2002 4:58:05.101 am
Abend 1 on P00: Server-5.60c-1937: CPU Hog Detected by Timer
Running process: ASDB 1081 Process
Thread Owned by NLM: ASDB.NLM

I rebooted the server and now the backup runs but no merging takes place. I did a manual merge and it completed successfully without errors, but no detail files show up in Arcserve Manager. I assume my database is corrupted again.

How can I prevent this from happening again???

 

[ArcserveFreek]

Hi,
Please do the following.

1. Apply the patch QO24478 and update asdb modules.
2. Make the following changes in the file "sys:\system\bti.cfg"

[MicroKernel]
MaxFiles =200 (default is 20)
MaxCursors=200 (default is 60)
CacheSize=32768 (default is 1024)
MaxClients=120 (default is 30)
BackgroundThreads=25 (default is 4)

[Btrieve Communications Manager]
MaxWorkerThreads=25 (default is 3)
MaxClients=120 (default is 15)

After increasing these parameters you will need to restart the server, so that the changes will take effect.
After rebooting the server, start Arcserve and run test backups to observe the performance.

ArcserveFreek
"Patience, persistence and perspiration make an unbeatable combination for success."

 

[domburi]

I actually have the ASDB files from patch qo31736. Is that a problem? Should I revert back to the older ones???
I'll try upping the bti.cfg values and restart the server. I'm assuming I should start with a blank database as well?

Thanks!

 

[domburi]

I did all this and thought I was all set! Two backup jobs ran and merged fine. I took a previous tape to merge it in (needed to do a restore) and partway through the merge, the server ABENDED:

Server DDD halted Friday, January 10, 2003 10:22:51.863 am
Abend 1 on P00: Server-5.60c: Page Fault Processor Exception (Error code 00000000)

EIP in LIBC.NLM at code start +00052EF3h
Access Location: 0xBD82C000

Running process: APROCESS 1 Process
Thread Owned by NLM: APROCESS.NLM

Any ideas???

 

[skambraks]

Hi,
I have the same probleme with exactly the same background (NW6 SP2, AS7 with SP) and the abends:

Server SERVER halted Saturday, 18 January 2003 0.00.40,438
Abend 1 on P00: Server-5.60c-1937: CPU Hog Detected by Timer
Running process: ASDB 73 Process
Thread Owned by NLM: ASDB.NLM

I have found this problem in several other newsgroups and no answer. CA tells me the problem is unknown.
have you got fixed the problem?? If so, please give me a hint how to do it: e-mail: a.skambraks@kerks.de.
greetings,
alfred

 

[markshaw100]

Hi,

I'd guess, the usual suggestions which are to apply the latest patches and NLM updates.

Since upping BTI.CFG did you recreate the database set. It would be a good idea as its almost certainly corrupted. Just upping BTI on its own isn't likely to work.

Take care if you have other apps such as DHCP loaded on the server that also consume BTREIVE resouces away from the ArcServe database modules. You may have to further increase them over the standard values.

Mark