We have been running Server 6.5.1/W2K for quite some time without any significant issues. Suddenly yesterday afternoon, we began getting a console message “PANIC: LookupHandle: handle out of range”. We restarted the server and experienced the same error and crash sequence several times. In looking through the NSD log file, we found that the Fatal Thread came from Amgr. We restarted the server without Amgr running and have had no crashes since. There are no Notes Log entries which we can recognize as a clue to which agent or database is the culprit. Is there a way to find it in the NSD log file? I can post the file, but it is rather large or can email it to anyone interested in looking. Thanks.
John Rowland
Murphy Oil Corporation
Subject: Domino 6.5.1 Amgr Crashes Server
John, send me the NSD log, I should have a solution after I look at the NSD. CHERNANDEZ@UAFC.COM
Subject: Domino 6.5.1 Amgr Crashes Server
Does it crash right away when you start agent manager, or does it crash only as it begins to process an agent.
If it doesn’t crash right away, run a “tell amgr show sched” to see your scheduled agents, then go turn them all off. Then restart Amgr and see if it crashes. If it doesn’t, enable those agents one at a time, letting each run to completion until you find the one that’s crashing the server.
Also try running “Tell Amgr Debug” with * or another parameter as listed in the Admin Help, to get some additional info.
Subject: RE: Domino 6.5.1 Amgr Crashes Server
Based on observation, it crashes when an agent executes. When we were restarting the server yesterday, the server would run anywhere from 3 to 5 minutes before crashing again. Unfortunately, every time it crashes, the server has to do consistency checks on several hundred mail files and we have users beating down the door. Consequently, we will probably not load Amgr and run the suggested console commands until a slow time, i.e. lunch time. I was in hopes that I might be able to spot something in the dump files that would help me do it non-invasively. Thanks.
John
Subject: RE: Domino 6.5.1 Amgr Crashes Server
I understand.
When you see the Panic error, latest text always the same?
Set “Log_AgentManager=1” to get AMgr starting to log its agent runs to the Notes Log.
You could run “start consolelog” to get more logging that may not yet be in the Notes Log due to crashing not allowing the last few entries to be written to Notes Log.
Subject: Domino 6.5.1 Amgr Crashes Server
John - this is the EXACT situation that is happening at my company. We, too are running W2K/6.5.1. I have been communicating with someone at IBM sending NSDs back and forth. We also have gone as far as rebuilding the entire server after looking at the Notes logs, OS logs and running diagnostics on the server. All appear to be fine.
I took out the Amgr parameter from the notes.ini after trying to start the server and stop Amgr. That’s where I received the same panic error as you. I restarted the server, let it run for an hour or so, and then took Greg’s advice and ran “tell amgr schedule”. I did find an agent that was scheduled on this server. Hopefully, this is the culprit.
Subject: RE: Domino 6.5.1 Amgr Crashes Server
I think I may have spoken too soon. My Domino Server is now down again and I think I am back to square 1.
This is the error I am receiving when starting Domino…
Thread=[0700:0004-060c]
Stack base=0x03CE0094, Stack size = 6652 bytes
PANIC: OSBBlockAddr: Bad BBlock handle
Any help would be greatly appreciated.