VMware player and NotFred crashes

[Catalina588]

NotFred console says "Out of memory: kill process 986 (mpiexec) score 24984 or a child. Killed process 987 (FahCore_a2.exe)." At this point, FAHspy loses connection (stops running) and NotFred Diskless Folder can no longer connect. VMware-vmx.exe process cpu time drops from 50% (of 8 cores) to ~38%. Happens about every 15-20 minutes. Happens when no other folding or other jobs running. Workaround is VMware Player Ctrl-Alt-Delete, whch is a reboot. Note, on restart, Core is re-downloaded before restart from checkpoint.

The NotFred install was done after reading and following several install guides, including at Hard[OC].

Computer Configuration:
Core i7-865 (8 cores) Intel DX58SO mobo; 6 GB; stock clocks; nVidia 8600 GT and 9800 GT in multi-GPU setup; Windows 7 Ultimate x64.
NotFred set up as 2 cores x 2 instances. VM Player gets 2048 MB memory. Task Manager does not indicate any real memory problems ("506 MB Free").

NotFred Configuration:

Diskless Folder
Diskless Folding version 090706 by Nick Reilly
Processor type: x86_64
Found 4 processors
User: Catalina588
Team: 711
Passkey: 
Big Packets: big
Advmethods: yes
Memory: 2048 MB
Backup every 5 minutes
Core checkpoint every 5 minutes
Restart running with oneunit flag
Remote reboot: disabled
Install: yes
Console blank: disabled
Hostname: fold-C6AC 
IP address: 192.168.1.117 
Workgroup: DISKLESS
Running under VMWare, not using NTP to sync time
Hang check logfile
SMP instance per 2 CPUs
Folding CD Generator
Reconfigure this host and any USB drives.
Processor temperatures (Intel Core and AMD K8)
Instance 1
Folding Log File
Unit Info
Kill folding cores
Instance 2
Folding Log File
Unit Info
Kill folding cores

VMware Player Configuration:

#!/usr/bin/vmplayer
.encoding = "windows-1252"

# This is a Workstation 5 or 5.5 config file
# It can be used with Player
config.version = "8"
virtualHW.version = "4"

# Number of virtual CPUs. Your virtual machine will not
# work if this number is higher than the number of your physical CPUs
numvcpus = "4"

# Memory
memsize = "2048"

# Selected operating system for your virtual machine
guestOS = "other-64"

# displayName is your own name for the virtual machine
displayName = "VM1"

# These fields are free text description fields
annotation = "VM1 Folding"
guestinfo.vmware.product.long = "Diskless Folding by Nick Reilly"
guestinfo.vmware.product.url = "http://reilly.homeip.net"
guestinfo.vmware.product.class = "virtual machine"

# First serial port, physical COM1 is not available

# Optional second serial port, physical COM2 is not available

# First parallel port, physical LPT1 is not available

# Settings for physical floppy drive

# Logging
logging = "FALSE"

# These settings decides interaction between your
# computer and the virtual machine
isolation.tools.hgfs.disable = "FALSE"
isolation.tools.dnd.disable = "FALSE"
isolation.tools.copy.enable = "FALSE"
isolation.tools.paste.enabled = "FALSE"

# First network interface card
ethernet0.present = "TRUE"
ethernet0.virtualDev = "e1000"
ethernet0.connectionType = "bridged"
ethernet0.addressType = "generated"
ethernet0.generatedAddressOffset = "0"

ide0:0.present = "TRUE"
ide0:0.fileName = "folding.vmdk"

ide0:0.redo = ""

# Clock is in UTC
rtc.diffFromUTC = -5

extendedConfigFile = "folding.vmxf"
virtualHW.productCompatibility = "hosted"
serial0.present = "FALSE"
serial1.present = "FALSE"
parallel0.present = "FALSE"
floppy0.present = "FALSE"
checkpoint.vmState = "folding.vmss"
ethernet0.generatedAddress = "00:0c:29:4e:c6:ac"
uuid.location = "56 4d 07 39 44 31 19 bb-e1 b1 87 d3 6b 4e c6 ac"
uuid.bios = "56 4d 07 39 44 31 19 bb-e1 b1 87 d3 6b 4e c6 ac"
cleanShutdown = "FALSE"
replay.supported = "FALSE"
replay.filename = ""
vmotion.checkpointFBSize = "20512768"

#As per hardforum user guide
priority.grabbed ="idle"
priority.ungrabbed = "idle"

 

[flogge]

Other than folding for the wrong team , I see where you entered the priority settings in the vmx file you have "priority.grabbed ="idle". I think it should be "priority.grabbed = "idle" with the extra space after the =. You should be able to run all 8 logical cores under 1 VM using VMPlayer 3.0 There is no reason to split them anymore that I know of. Just change the numvcpus = "4" to "8" and only run 1 VM and change notfred to run an instance per 8 CPUs. If you have reason to keep 2 separate instances with 4 cores each you can set the radio button in notfred console to instance per 4 CPUs rather than per 8 CPUs

 

[T4rd]

Are you OC'ing? That error is similar to what I was receiving before I got my OC stable. If you are OC'ing, run Prime95 and see if it errors out after a few mins (usually if the OC is unstable) to verify its stability.

Also, just in case you didn't know, the i7s have 8 threads, and only 4 cores.

Edit: and to echo what flogge said, yeah, it's a bit more efficient on memory if you just run 1 VM with 8 cores. You'll get the same combined PPD then and be using much less RAM.

 

[Catalina588]

Flogge, your code review found the bug. One instance/8 threads/2048 MB memory at 99% CPU utilization. Thanks. Intel supplied the box, so they get the points. Rough justice.

T4rd, nope, the OCing got pulled a while ago. FAH is tougher than even Prime95. I do not recommend OCing before you get your serious folding up, running, and stable. Once I finish the remaining console units from the old OS, I'll try OCing utility with NotFred and multi-GPU running.

Yes, Core i7 has 8 threads and 4 cores. But if you turn Hyperthreading off, you'll lose about 47% of your PPD. So a hardware thread is about as good as a core to me for folding.

Thanks again, Guys. I'll be back.

 

[T4rd]

On my rig I found that Prime95 stressed my CPU out more than folding because my temps were about 5°C higher running Prime95. Also, Prime95 would throw me an error within about 5 mins and my linux SMP would error out after a couple days into a -bigadv WU, that's why it vexed me that my OC was unstable because it took so long for it to crash. But when I ran Prime, it almost immediately let me know I wasn't as stable as I thought I was. Trust me, it took me almost 2 weeks to troubleshoot this and to figure out the right voltages, multiplier, and BLCK speeds to keep my rig stable.

Yeah, I wasn't insinuating you turn off Hyperthreading, I just noticed you said "(8 cores)" next to your CPU in your desc and wanted to make sure you knew about HT/threads. But you obviously do .

Anyways, glad you fixed it and try just running one VM using all 8 threads/cores and see how that works.

GL!