This means that you took a snapshot of the VM (or multiple snapshots of the VM), and while the VM was running from the snapshots you consumed all of the available disk space in the VMFS volume.
To "fix" this, you will likely have to remove the snapshot from the VM (unless you can allocate additional disk space - in which case you should do that so you don't lose data - follow the 'add extent' section in the ESX 3.5 Admin Guide). To 'remove' one of the snapshots, and hopefully free enough disk space to be able to boot your virtual machine, you can try these steps:
power off the virtual machine if it isn't already
right-click on the VM and select "snapshot -> snapshot manager", and verify that you have a valid snapshot chain:
it should be pointed to one or more snapshots that were created, with a 'you are here' at the bottom
you will need to remove at least the last snapshot in the chain (note: this will cause dataloss in the VM)
select one of the snapshots above 'you are here', and click 'goto'
click 'delete all'
wait for task to complete before attempting to power on the VM
With any luck, between having the VM powered off and snapshot files small enough that they can be moved back into each other, this will work for you. If it does not, you will have to actually remove the files from disk and point the VMX file back to older versions of the snapshots manually.
If this is a critical VM or you cannot afford data loss in it, I would recommend contacting VMware support to assist you with this process so they can review your exact situation and provide the best guidance on how to resolve it.
