For historical reasons we maintain a legacy queue on volume sys of a
server which is a node in a 2-node cluster (nw6.5 sp5). Because of
returning abends, once a week or so, all of them about connmgr, I moved
all resources to the 2nd node and node 1 only serviced the queue. It
kept abending, to the point that I thought it a defective memory issue.
So, I have another 2-nodes cluster, nw6.5 sp7 and I moved the queue to
one of the nodes. And lo and behold, that node abended too with the same

Server IMPALA2 halted Tuesday, 5 February 2008 7:40:44.684
Abend 1 on P00: Server-5.70.07: Page Fault Processor Exception (Error
code 00000000)

CS = 0008 DS = 0023 ES = 0023 FS = 0023 GS = 0023 SS = 0010
EAX = 9B4C0140 EBX = 93C69900 ECX = 00000000 EDX = 00000000
ESI = 00000075 EDI = 00000000 EBP = AA23EF04 ESP = AA23ECCC
EIP = 8A351E4F FLAGS = 00010246
8A351E4F 8B38 MOV EDI, [EAX]=?
EIP in CONNMGR.NLM at code start +00003E4Fh
Access Location: 0x9B4C0140

Additional Information:
The CPU encountered a problem executing code in CONNMGR.NLM. The
problem may be in that module or in data passed to that module by a
process owned by SERVER.NLM.

Any help will be appreciated.

Limor Kosovsky
The Hebrew University of Jerusalem