My machine just crashed hard with oom killer taking out a lot of needed processes. I reviewed the atop log and noticed that several vmware-vmx processes had grown out of control using ~23G ram each. Obviously there is something wrong with vmware-vmx but I don't know what it could be??
Since I restart the VMs every minute is there some way I can get the vmware-vmx processes to recycle themselves?
PID MINFLT MAJFLT VSTEXT VSIZE RSIZE VGROW RGROW MEM CMD 1/51
3019 28492 0 8112K 24.0G 22.9G 111.1M 111.1M 16% vmware-vmx
3007 28051 0 8112K 24.0G 22.9G 109.3M 109.4M 16% vmware-vmx
3031 27978 0 8112K 23.8G 22.8G 109.1M 109.1M 16% vmware-vmx
2993 28157 0 8112K 23.7G 22.7G 109.8M 109.8M 16% vmware-vmx
2996 28712 0 8112K 23.5G 22.4G 112.0M 112.0M 16% vmware-vmx
2984 28091 0 8112K 23.5G 22.4G 109.6M 109.6M 16% vmware-vmx
14179 11931 0 8112K 2.4G 1.3G 47652K 47552K 1% vmware-vmx
Host
Ubuntu Lucid 2.6.32-38-generic
VMware Workstation 8.0.1 build-528992
VMs
Windows XP 512MB 1cpu
Usage pattern (all using Vix API)
-start VM
-copy small file to VM and run
-stop VM (revert to snapshot)
-repeat 1 minute later