VMware Cloud Community
mattkenny
Contributor
Contributor
‎10-01-2010 08:39 AM

esxi PSOD "NOT_IMPLEMENTED bora/vmkernel/main/mpage.c:624"

I'm running ESXi 4.1 (build 260247) on a box here, and in the past couple of months it has been crashing a lot (usually between a couple of days and a week or so).

It's always the same error message:

NOT_IMPLEMENTED bora/vmkernel/main/mpage.c:624

Here's a couple of screenshots of the PSOD:

http://img.whirltools.com/i/60953995477546980776.jpg

http://img.whirltools.com/i/11895140979643031838.jpg

And here's the crash log (I think)

http://dl.dropbox.com/u/8877202/Esxi/vmkernel-log.1

I've been running this setup (see below) a couple of years I think and only in the past couple of months have I had any issues with crashing. I've searched but can't find any reference to that particular error.

Any ideas why this might have started recently? Or why it's happening? It's starting to drive me nuts having to reboot the server every day or two, since this box is acting as my router.

(The error sounds like a VM is trying to call something that isn't implemented by the hypervisor, but I would have thought if that was the case, then someone else would have posted this error somewhere on the net. But I get no hits on google for any combination of that error, other than the generic "not_implemented" error type...)

Here's the specs of the box:

  • Mobo: P5K-VM

  • CPU: Q6600

  • RAM: 8GB

  • NIC1: intel gigabit PCI card

  • NIC2: intel dual port gigabit PCI-e card

  • Storage controller: PERC 5/i 256MB cache + BBU

  • HDDs: RAID5 using 3x 750GB WD drives (TLER)

Here's the VMs I'm running:

  • 1x pfsense

  • 1x Win Server 2003 (dns, dhcp, wsus)

  • 2x ubuntu 10.04.1 server (LAMP)

  • 1x Win XP (utorrent)

Tags (2)
0 Kudos
1 Reply
mattkenny
Contributor
Contributor
‎02-20-2011 08:57 PM

I seem to have found the cause of this problem - I had mismatched RAM. 2 sticks of DDR2-6400 and 2 sticks of DDR2-8500. Silly me. After removing the two sticks of 6400 ram, I haven't had any crashes since.  I'm surprised that this caused the exact same error every time though. I would have thought it would have been more likely to be a random error each time...

0 Kudos