I'm running Centos 6.4 and was running WS 8.0.4. I "yum upgraded" and now Windows VMs don't start anymore. The kernel went from 2.6.32-279 to 2.6.32-358 ... I rolled back the kernel and removed the .ko files, and still get the same issue. I upgraded to WS 8.0.6 and get the same issue. I upgraded to 9.0.2 and get the same issue. Linux VMs start fine.
When I click on "Power on the virtual machine" the GUI freezes; no buttons can be clicked. If I try to switch to a different tab (like Home), it eventually after several minutes lets me navigate there. When I navigate back to the VM tab, I get a popup: "Unable to change virtual machine power state: Cannot find a valid peer process to connect to" ... but the vmware-vmx instance continues to run. I cannot connect to the VNC instance (yet; could before the upgrade).
No error in the vmware.log file, and see "MKS PowerOn" and "MKS thread is alive" ...
I get some errors in /var/log/hostd.log
2013-04-02T11:21:55.703-07:00 [7F62150BD700 verbose 'Default'] Timed out reading between HTTP requests. : Read timeout after approximately 1205000ms. Closing stream TCP(local=127.0.0.1:8307, peer=127.0.0.1:43402)
I can telnet successfully to 127.0.0.1:8307 ... I get the same symptom even if I turn off iptables.
"service vmware start" generates all OK
vmware-vdiskmanager -R *.vmdk generates no errors.
I'm stuck. Any ideas of where else to look? Thanks!