VMware Cloud Community
Jimmy_J97
Contributor
Contributor
Jump to solution

PSoD Help

Hi guys,

I'm currently in need of assistance.

Fault: Hosts randomly Purple Screen of Death’ing

OS: VMware ESXI 6.0.0 build 3620759

Recent changes: Installed Trend Micro Deep Security v10.0

Installed NSX v6.3.4

Deployed Guest Introspection VD’s

Delopyed Trend Micro Virtual Appliances

_________________________________________________________________________

I have uploaded a screenshot of the VMKernal-zdump file and an image with the PSoD error.

I am wondering if anyone can help find the cause of this and if possible, a resolution.

pastedImage_0.png

pastedImage_1.png

Tags (2)
Reply
0 Kudos
1 Solution

Accepted Solutions
TheBobkin
Champion
Champion
Jump to solution

Hello Jimmy,

Still no real info provided there -  if there is no messages in the screenshot then unlikely they are in here either, you should be looking for messages after 'ESXi' of the format:

0xhhhhhhhhhhhh:[0xhhhhhhhhhhhh]MESSAGES_HERE@vmkernel#nover+0x37e

0xhhhhhhhhhhhh:[0xhhhhhhhhhhhh]MESSAGES_HERE@vmkernel#nover+0x4d

0xhhhhhhhhhhhh:[0xhhhhhhhhhhhh]MESSAGES_HERE@vmkernel#nover+0x12c

The messages between ']' and '@' specifically - if it is not capturing these then configure an appropriate core-dump partition so it does (and of course consider upgrading as there are multiple potential causes that have been fixed).

Bob

View solution in original post

Reply
0 Kudos
4 Replies
daphnissov
Immortal
Immortal
Jump to solution

Because of the numerous factors and variables involved in troublshooting PSODs, I'd recommend you open a support case with GSS and submit your dumps and logs.

Reply
0 Kudos
TheBobkin
Champion
Champion
Jump to solution

Hello Jimmy,

Welcome to VMTN Communities.

How many times has this occurred and any correlation with certain workloads or activities?

(e.g. after X amount of days uptime, during back-ups, when vMotioning multiple or specific VMs etc.)

Looks like your dump partition is not adequate in size to capture complete dump so the data stored might not be sufficient to provide the necessary information for troubleshooting, consider configuring a larger coredump partition so that the necessary data is saved:

https://kb.vmware.com/s/article/2004299

Can you search for 'ESXi' in that zdump in Notepad++ and copy out any backtraces messages following this?

You may benefit from formatting that to log format properly:

https://kb.vmware.com/s/article/1006796

Any particular reason (e.g. company-policy, security, interoperatibility) why running on such an old build of ESXI/vSAN?

I would advise updating to a minimum of 6.0 U2 P04 (4600944) for vSAN and ESXi fixes and later if possible to go to U3.

Bob

Reply
0 Kudos
Jimmy_J97
Contributor
Contributor
Jump to solution

Hi Bob,

Thanks for you reply, as its helped me out greatly.

I have screenshotted the output from search for "ESXi"

pastedImage_0.png

Thanks,

Jimmy

Reply
0 Kudos
TheBobkin
Champion
Champion
Jump to solution

Hello Jimmy,

Still no real info provided there -  if there is no messages in the screenshot then unlikely they are in here either, you should be looking for messages after 'ESXi' of the format:

0xhhhhhhhhhhhh:[0xhhhhhhhhhhhh]MESSAGES_HERE@vmkernel#nover+0x37e

0xhhhhhhhhhhhh:[0xhhhhhhhhhhhh]MESSAGES_HERE@vmkernel#nover+0x4d

0xhhhhhhhhhhhh:[0xhhhhhhhhhhhh]MESSAGES_HERE@vmkernel#nover+0x12c

The messages between ']' and '@' specifically - if it is not capturing these then configure an appropriate core-dump partition so it does (and of course consider upgrading as there are multiple potential causes that have been fixed).

Bob

Reply
0 Kudos