Scenario:
- VMWare ESXi 4.1 (just one server)
- SCSI storage;
During last night host lost comunication with Storage, for a few seconds, and then auto recovered it.
All VMs that were running got locked, and they appered like Unknow 1, Unknow 2, etc, on Inventory list.
What I tried:
1 - run vmkfstools -D namefile.vmx to identify the problem - result:
Lock [type 10c00001 offset 24461312 v 227, hb offset 3567616
gen 33, mode 1, owner 522ca88a-f172d435-4a87-00e081742652 mtime 46205]
Addr <4, 31, 64>, gen 80, links 1, type reg, flags 0, uid 0, gid 0, mode 100755
len 3303, nb 1 tbz 0, cow 0, zla 2, bs 65536
2 - Removed the "Unknow" VM from inventory;
3 - try to reboot host several times, using and not using Maintenance Mode
4 - copy/backup files - did not work;
5 - look for process on the ESXi that could be using files of the VMs - found none;
6 - tried to use vmkfstools -B and -L release, on locked files - no success;
7 - tried to read files and got "invalid argument" message. Actually, all tries I made to access .vmx, vmdk, etc, resulted in "invalid argument". Other files on VMs folder's are OK.
8 - create a new VM and attach an existing virtual disk (VMDK) - no success.
9 - try to add vm to inventory from datastore browsing. Option is greyed out when try .vmx right click.
The virtual machines that were not running when the incident ocurred are completly fine... They exist in the same datastore.
Hope you guys can give me some good tips about it.
Regards.
Rafael
