asked on

Vmware hard disks activity

Hi there

A server in the company has a disk issue and here are a couple of events that are from 2 months ago and a number of NTFS events leading to the current date.

I have performed a read only CHKDSK disk and this has taken over 70 minute to complete on the c: drive when its less than half the size of the D: drive which only took 2 minutes.

You can also see from the AVG Disk Bytes/Write column in perfmon the bytes per second is really low and have seen triple this speed on a LUN even when installing an OS, there are also periods where the average drops to 000 bytes written so the hardware certainly needs investigating as badly performing hard drive or RAID array seems to be the culprit.

Now the server is a Virtual machine.

Any help please

Note: I know very little about virtual machines (not my area) but this incident was thrown my way therefore the simpler your answer is the more appreciated.

Thanks
Event2.png
Event-1.png
AVG-Read--Write.png

Andrew Hancock (VMware vExpert PRO / EE Fellow/British Beekeeper)

Error in July is too long ago to consider, however delay write fail is interesting, in Sept?

Do you have any logs for this event? e.g. it could have been a network down, path failure etc

How large are the disks you are using chkdsk on, this could be normal disk activity for your datastores.

What is the SAN, do you use Jumbo Frames, what is the networking?

Do you have a dedicated storage LAN?

What type of datastores are you using, RAID type, Disk type, SAS, SATA, SSD, disk speed, 7k, 10k, 15k?

Do you know how to use ESXTOP, you can check the latency, to see if the datastore is too slow for your VM ?

see my EE Article

HOW TO: Performance Monitor vSphere 4.x or 5.0

M SOS

ASKER

So if I get you all these answers will you be able to help?

Thanks

M SOS

ASKER

I have never ever heard of Esxtop ever in my whole life.

M SOS

ASKER

and the event viewer events and the performance monitor do not help?

Andrew Hancock (VMware vExpert PRO / EE Fellow/British Beekeeper)

I can certainly try and help you, that's what I try to do as an Expert!

Windows VM, does not know it's virtual! So they do not really help, we need to look at the Host!

Have a read of my EE Article, and I can show you how to use esxtop.

M SOS

ASKER

So are you saying that windows monitoring tools on a VM are not actually descriptive of anything ?

The thing is I won't know all these answers until mid morning UK time...Will post them then and will be very grateful of you could have a look.

now the above questions are the one needed?

M SOS

ASKER

Are you saying that Windows monitoring tools when it is a VM machine are not descriptive of anything?

The thing is I won;t be able to get the answers to your questions until tomorrow mid morning UK time? Will have a look?

Are the above question needed? Any thing else?

Andrew Hancock (VMware vExpert PRO / EE Fellow/British Beekeeper)

Correct, the Windows OS does not know it's a VM, so an event triggered in the log, about disk, as the disk is virtual and hosted on a hypervisor, you need to look at the whole picture to discover the issue.

If it was a physical disk, troubleshooting would be different.

Take as long as you like, gathering information. Any further Windows event logs could be handy, if they are disk related!

Those questions are good to start, I'm on GMT (UK Time!), I'm based in the UK!

M SOS

ASKER

Great, Thank you very much but just one last question for tonight... Why are these Windows logs then happening

why the scan disk is taking longer and why the up-normal activity on the performance monitor?

Thanks

M SOS

ASKER

What is the SAN, do you use Jumbo Frames, what is the networking?

can you clarify this question a bit clearer please?

ASKER CERTIFIED SOLUTION

Andrew Hancock (VMware vExpert PRO / EE Fellow/British Beekeeper)

membership

This solution is only available to members.

To access this solution, you must be a member of Experts Exchange.

Start Free Trial