2008-12-05 06:45 AM
We've got a big nfs-pool for our esx-servers containg about 100 virtual machines. Most of the machines have been migrated to this pool.
My problem is, that the snapshots are getting to big (25% in 1 Week). I think that there are some machines doing database-dumps, but I don't know which.
Is there a way to find the files which are mostly overwritten?
2008-12-05 02:40 PM
You have not said now your storage is configured on Vmware Block or Nas access?
NFS has some tools you could use but they may not be helpful.
Have you looked at the problem from the ESX side? Have a look at this
VMware ESX Server 2
Using esxtop to Troubleshoot Performance Problems
If you are still struggling please describe your filers, connectivity and ESX disc configuration.
Welcome to the forum.
2008-12-10 03:21 AM
Thank you for the information.
We have a metrocluster based on two FAS6030, the storage is configured on vmware nas-access. Each esxhost has a separate 1Gbit storage interface. That means i'm interessted in your mentioned nfs-tools.
It would be nice to log the disk writes for each vmdk.