ESXi hangs

[Deadmorozka]

I have VMware ESXi 3.5 Installable Update 3 Build 123629 with Windows 2008 Ent VM. Periodically (1-2-3 times a day) this VM hangs for about 5-10 minutes.

VI Client shows no CPU and Memory load at that moment. It is impossible to connect to ESXi using VI Client or SSH when the VM stopped because of timeout and I can ping the management interface.

Then everything becomes OK. And there is no any record in VI Client or vmware.log file after that...

Help !!!

[Formatter]

Have you considered updateing using the update manager that you can install when you install the vi client? Lots of updating to do maybe.

[kooltechies]

Hi,

What application is that VM running , have you considered giving some more resources to the VM.

Thanks,

Samir

P.S : If you think that the answer is helpful please consider rewarding points.

[Lightbulb]

Are there sanpshots on the VM?

Is Shadow Copy services enabled on the 2003 ENT VM? I ask because by default this runs a couple of times a day when enabled.

This sort of fells like an I/O subsystem issue. Check ESXi logs for signs of SCSI errors.

[Deadmorozka]

This VM is running some kind of ERP-system.

About resource allocation: At the present moment the upper limit for this VM is 15000Mhz of 21336Mhz. (and when it was "Unlimited" it didn't change the situation).

The Windows VSS service is disabled. SQL Server VSS Writer is running. But there is no record in Event logs concerning its activity.

And another strange thing: i've enabled debug-level logging for this VM. But it seems to me there are too few records. Log-file is in attachment.

[Lightbulb]

I may be interpreting this totally wrong but did you guys Hot expand the system/boot volume on this VM twice in the last few days?

Feb 12 20:15:21.382: vmx| HotExtend: Extending scsi0:0 to 136314880

<SNIP>

Feb 12 20:15:21.432: vmx| DISKLIB-LIB : Growing disk '/vmfs/volumes/4979d8d3-609d6d5c-c2fc-0015175bab44/vm-server-3/vm-server-3.vmdk' : capacity = 104857600 sectors - 50.0 Gb

Feb 12 20:15:21.432: vmx| DISKLIB-LIB : Growing disk '/vmfs/volumes/4979d8d3-609d6d5c-c2fc-0015175bab44/vm-server-3/vm-server-3.vmdk' : new capacity = 136314880 sectors - 65.0 Gb

Feb 15 13:05:31.087: vmx| HotExtend: Extending scsi0:0 to 146800640

<SNIP>

Feb 15 13:05:31.142: vmx| DISKLIB-LIB : Growing disk '/vmfs/volumes/4979d8d3-609d6d5c-c2fc-0015175bab44/vm-server-3/vm-server-3.vmdk' : capacity = 136314880 sectors - 65.0 Gb

Feb 15 13:05:31.142: vmx| DISKLIB-LIB : Growing disk '/vmfs/volumes/4979d8d3-609d6d5c-c2fc-0015175bab44/vm-server-3/vm-server-3.vmdk' : new capacity = 146800640 sectors - 70.0 Gb

[Deadmorozka]

Yes, i have rebooted VM to expand its hard drive. There is no problem with it.

Some thoughts:

I have 2 hard drives in this VM. One for system files and another for database files. There are 2 datastores in ESXi host for that drives. Every store is situated on its own RAID1 physical volume.

May be that is the problem ?

[Lightbulb]

Have you checked the ESXi logs for signs of SCSI errors?

Are there any errors in the App or System log of the VM at the time these "freezes" happen?

Having the vmdks on 2 different datastores off the same local controller should be ok, although I don't see what you gain from having it setup this way.

[Deadmorozka]

Here is the part of "/var/log/messages" file, when the last accident happened:

Can't see anything strange...

The VM was hanging since 13:06 till 13:09...