vMotion Error at 78% with NFS storage

When I add a new host (ESX4.1) in my VMware Cluster and i wanted to do a vMotion to the new host  a got a error at 78%. The popup box says :

A general system error occurred sources detected that destination failed.

when i ask my friend Google a got  a hit at vmfaq. It says that:

“This error message indicates that the shared storage is not available equally to both host involved in the VMotion event.”.

I went  in to vCenter and a did not see any strange thinks en i could browse the NFS datastore form every host.When i went tot the datastore through the inventory  datastores i say that the ESX host (destination) had a red marker and the other hosts did not but could not find any other information in vCenter. Even at the configuration tab of the esx host i compared the old host and the newly add host. When you look at the datastore details you’ll find no discrepancy when comparing these details.

so i went to the  did a ls –l so this is what i found on one of the hosts where it worked correctly.

figure 1

When i look ad the new add host if found the result of ls –l /vmfs/volumes deference from the first host (look ad the green shaded texts).

 

figure 2

In the same article is says that “Remove the volumes involved from the ESX hosts and re-add them by using the exact same host names to solve this problem” I did a un-mount of the NFS datastore on the new host. and mounted the datastore again and when i looked i gave the same resulted as figure 2. So i did it again and it work? why the first time was little different then the first time

The fist time i typed /mnt/volume group1/vol1

The second time a typed /mnt/volume group1/vol1/

So the remount NFS datastore with the /  at the end works. it was like a needle in a haystack. Here is the result of the ls –l /vmfs/volume of the new host.

It does not matter whether you use the / at the end or not but it must be done the same way every where. So mount you’re datastore with a script and document this very good.