We are having random abend problems with a NetWare 6.5 server running on VMware ESXi 5.0.

In July, we moved the NetWare VMware guest from an older VMware ESXi 3.x host server to the new VMware ESXi 5.0 host server.

Rarely had abend problems with the server on the old ESXi 3.x host.

Since moving the server in July, we've had at least 5 "abends". I put abends in quotes because some of the "abends" never created an ABEND.LOG file. When the server abends, all we have is a black console screen in the ESXi vSphere Client console for the NetWare guest. All network communication with NetWare server ceases. We've tried waiting for 15-20 minutes to see if it becomes responsive again, but it never does. We then have to just power it off, and bring it back up.

The abends that we have logs for are similar to the log snippet below, except that the Interrupted Process is different each abend.

Server SHS halted Tuesday, August 14, 2012 6:08:16.503 pm
Abend 1 on P00: Server-5.70.08-0: Deadlock detected waiting for spinlock currently owned by CPU 00

CS = 0008 DS = 0023 ES = 0023 FS = 0023 GS = 0023 SS = 0010
EAX = 00000030 EBX = 91711518 ECX = 915889D0 EDX = 00000030
ESI = 00132D03 EDI = 0000000A EBP = 00000000 ESP = 48990AC4
EIP = 00110C9C FLAGS = 00000006
00110C9C 83C404 ADD ESP, 00000004
EIP in UNKNOWN memory area

The violation occurred while processing the following instruction:
00110C9C 83C404 ADD ESP, 00000004
00110CA1 7413 JZ 00110CB6
00110CA3 0F31 RDTSC
00110CA5 8BF8 MOV EDI, EAX
00110CA7 23C3 AND EAX, EBX
00110CA9 83C807 OR EAX, 00000007
00110CAE 03F8 ADD EDI, EAX
00110CB0 0F31 RDTSC

Running process: Interrupt service routine (nested count 1)
Interrupted process: NWTS 19123 Process
Thread Owned by NLM: DTS.NLM
Stack pointer: 4898B9A0
OS Stack limit: 489817A0
CPU 0 (Thread 9C3D4420) is in a NO SLEEP state
Scheduling priority: 67371008
Wait state: 5050190 Blocked on a kernel CV
Stack: -00132CC5 (LOADER.NLM|AlternateKeyboardHandler+101)
--00000000 (LOADER.NLM|KernelAddressSpace+0)
--00001B09 (LOADER.NLM|KernelAddressSpace+1B09)
--00000000 (LOADER.NLM|KernelAddressSpace+0)
--91711518 ?
00110ABD (LOADER.NLM|kspinlock_patch+76)
9155A58F (TCP.NLM|CreateIPv4AddressFromMapped+BFB)
-915889D0 (TCP.NLM|IP6DeregisterFunc+1270)

-- Server Hardware Info
Dell PowerEdge T610
Intel Xeon X5690 (x2)
48GB Memory

-- VMWare Info
VMware ESXi 5.0 build 504890
One 2.86TB Datastore
Three running guest machines
1. NetWare 6.5 sp8
2. Windows 2003 Standard Server
3. Windows XP Pro - SP3

-- NetWare Info
NetWare 6.5 SP8 with all post SP8 patches I can find.
eDirectory SMP
One processor assigned
4GB memory assigned
One 1.1TB hard drive assigned (SCSI)
One network adapter assigned (E1000)

-- Other Products Running on NetWare Server
NDPS Manager
NDPS Broker
GroupWise MTA 7.0.3
GroupWise POA 7.0.3
ZENworks Desktop 7
DNS Server
VMware Tools 1.01
Apache & TomCat

There are multiple NetWare servers running in the environment at different locations. Some Running as VMware guests and a couple physical servers yet.

This is a school district and I need to get this fixed before school starts in September.

Does anyone have any ideas? Something I missed on the ESXi server config?
If more information is needed please let me know.

Steven Reich