We have a cluster (HP DL380 G4, MSA 500) running NW 6.0, SP5. Each
server has 4G of memory. One server runs the backups with Arcserve 11.1
SP1. A backup was schedule last night, starting at 11:00 PM. This
morning there was the following message on the console from 12:04 AM:
"Cache memory allocator out of available memory." The backup hung not
long after the message was written. I tried to end the job and stop
Arcserve, which then led to an abend. The server didn't write much into
the abend log before freezing up, but what it wrote is below. Any ideas
appreciated. Thanks. Phil

Server WR2 halted Sunday, September 17, 2006 10:35:31.036 am
Abend 1 on P00: Server-5.60.05: Page Fault Processor Exception (Error
code 00000000)

Registers:
CS = 0008 DS = 0010 ES = 0010 FS = 0010 GS = 0010 SS = 0010
EAX = B7DE51E0 EBX = 00000000 ECX = 01A37300 EDX = 002488C8
ESI = 00000000 EDI = 0001000B EBP = 08E2B974 ESP = 08E2B928
EIP = 42836AD1 FLAGS = 00010246
42836AD1 83B94C01000000 CMP [ECX+0000014C]=?,00000000
EIP in FSTAPE.NLM at code start +00002AD1h
Access Location: 0x01A3744C

The violation occurred while processing the following instruction:
42836AD1 83B94C01000000 CMP [ECX+0000014C],00000000
42836AD8 75E6 JNZ 42836AC0
42836ADA 68D0070000 PUSH 000007D0
42836ADF E89C01238B CALL THREADS.NLM|delay
42836AE4 59 POP ECX
42836AE5 31C0 XOR EAX,EAX
42836AE7 8D65F8 LEA ESP,[EBP-08]
42836AEA 5E POP ESI
42836AEB 5B POP EBX
42836AEC 5D POP EBP



Running process: FSTAPE 2 Process
Thread Owned by NLM: FSTAPE.NLM
Stack pointer: 8E2B8F8
OS Stack limit: 8E1D500
Scheduling priority: 67371008
Wait state: 3030070 Yielded CPU
Stack: -15290014 ?
--08E2B93C ?
-1529000C ?
--08E2B954 ?
--00000000 ?
--38343732 ?
--34323039 ?
--00000000 ?
--00000000 ?
--00000000 ?
--00000000 ?