I am running Linux pigeon 2.6.10-1.760_FC3 on my home machine.

The machine has 2 GB of RAM and 4 GB of swap.

I need to run windows XP in a vmware session (for accessing a customer's network with a windows-only vpn).

I have allocated 1596 MB for my XP virtual machine. When I launch the guest OS,
after a short while, it gets killed.

Looking in /var/log/messages, I find that this is because of the oom-killer. This just sucks. I don't ever want the kernel to randomly kill my processes. After researching this, I found that
if I echo "2" >/proc/sys/vm/overcommit_memory, it is supposed to disable the oom-killer.

I've done this, and I *thought* this was working (I worked for a while day without it kicking in), but now, I'm sure the value is "2", but oom-killer is still killing my vm. I've seen notes on the net that say the "2" is not supported yet.

Does anyone know what I can do to stop this thing.....

Here is what I see in my /var/log/messages file:
--------------------
Feb 16 14:41:06 pigeon kernel: oom-killer: gfp_mask=0xd0
Feb 16 14:41:06 pigeon kernel: Mem-info:
Feb 16 14:41:06 pigeon kernel: DMA per-cpu:
Feb 16 14:41:06 pigeon kernel: cpu 0 hot: low 2, high 6, batch 1
Feb 16 14:41:06 pigeon kernel: cpu 0 cold: low 0, high 2, batch 1
Feb 16 14:41:06 pigeon kernel: Normal per-cpu:
Feb 16 14:41:06 pigeon kernel: cpu 0 hot: low 32, high 96, batch 16
Feb 16 14:41:06 pigeon kernel: cpu 0 cold: low 0, high 32, batch 16
Feb 16 14:41:06 pigeon kernel: HighMem per-cpu:
Feb 16 14:41:06 pigeon kernel: cpu 0 hot: low 32, high 96, batch 16
Feb 16 14:41:06 pigeon kernel: cpu 0 cold: low 0, high 32, batch 16
Feb 16 14:41:06 pigeon kernel:
Feb 16 14:41:06 pigeon kernel: Free pages: 5372kB (832kB HighMem)
Feb 16 14:41:08 pigeon kernel: Active:235963 inactive:265434 dirty:23 writeback:1 unstable:0 free:1343 slab:5777 mapped:134793 pagetables:1867
Feb 16 14:41:09 pigeon kernel: DMA free:68kB min:68kB low:84kB high:100kB active:12kB inactive:11836kB present:16384kB pages_scanned:83 all_unreclaimable? no
Feb 16 14:41:10 pigeon kernel: protections[]: 0 0 0
Feb 16 14:41:10 pigeon kernel: Normal free:4472kB min:3756kB low:4692kB high:5632kB active:680kB inactive:838588kB present:901120kB pages_scanned:45 all_unreclaimable? no
Feb 16 14:41:10 pigeon kernel: protections[]: 0 0 0
Feb 16 14:41:10 pigeon kernel: HighMem free:832kB min:512kB low:640kB high:768kB active:943160kB inactive:211312kB present:1179584kB pages_scanned:64 all_unreclaimable? no
Feb 16 14:41:10 pigeon kernel: protections[]: 0 0 0
Feb 16 14:41:10 pigeon kernel: DMA: 1*4kB 0*8kB 0*16kB 0*32kB 1*64kB 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 68kB
Feb 16 14:41:10 pigeon kernel: Normal: 174*4kB 2*8kB 1*16kB 1*32kB 2*64kB 0*128kB 0*256kB 1*512kB 1*1024kB 1*2048kB 0*4096kB = 4472kB
Feb 16 14:41:10 pigeon kernel: HighMem: 12*4kB 28*8kB 1*16kB 1*32kB 0*64kB 0*128kB 0*256kB 1*512kB 0*1024kB 0*2048kB 0*4096kB = 832kB
Feb 16 14:41:10 pigeon kernel: Swap cache: add 59013, delete 53995, find 14789/18211, race 0+0
Feb 16 14:41:10 pigeon kernel: Free swap: 3852568kB
Feb 16 14:41:10 pigeon kernel: 524272 pages of RAM
Feb 16 14:41:10 pigeon kernel: 294896 pages of HIGHMEM
Feb 16 14:41:10 pigeon kernel: 5388 reserved pages
Feb 16 14:41:10 pigeon kernel: 580493 pages shared
Feb 16 14:41:10 pigeon kernel: 5018 pages swap cached
Feb 16 14:41:10 pigeon kernel: Out of Memory: Killed process 7402 (vmware-vmx).
---------------------

I don't know if this is still relevant, but i think the proper setting for preventing processes being killed by oom_killer is 1 (=always overcommit), rather than 2.