Bug 170985 - RHEL 4 Update 2 Incompatibility with VMware ESX 2.5.2
RHEL 4 Update 2 Incompatibility with VMware ESX 2.5.2
Product: Red Hat Enterprise Linux 4
Classification: Red Hat
Component: kernel (Show other bugs)
i686 Linux
medium Severity medium
: ---
: ---
Assigned To: Tom Coughlan
Brian Brock
: Regression
Depends On:
Blocks: 168429 175120
  Show dependency treegraph
Reported: 2005-10-16 22:07 EDT by Bruno Clermont
Modified: 2007-11-30 17:07 EST (History)
8 users (show)

See Also:
Fixed In Version: RHSA-2006-0132
Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of:
Last Closed: 2006-03-07 15:27:15 EST
Type: ---
Regression: ---
Mount Type: ---
Documentation: ---
Verified Versions:
Category: ---
oVirt Team: ---
RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: ---

Attachments (Terms of Use)

  None (edit)
Description Bruno Clermont 2005-10-16 22:07:33 EDT
- VMware ESX build 16390 (latest version on 2005-10-16)
- running on a HP Proliant Server P3 host with single CPU w/ SCSI disk
- RHEL 4 Update 2 iso files.

When booting a guest machine with the Update 2 iso file the kernel crash while
loading one of the SCSI module.

If a guest machine is installed with Update 1 it work fine until it's up2date'd
to match "Update 2" package versions and kernel upgraded to 2.6.9-22.EL.
It crashed at next reboot and require a fall back to 2.6.9-11.EL kernel.

I found this message on nahant-beta-list that look like my problem:
Someone else seem to had reproduce this problem before with Update 2-beta.
Comment 2 Tom Coughlan 2005-10-17 14:26:57 EDT
I am investigating the nahant-beta-list posting to see what became of it. 

In the meantime, please post the console messages that print when the system
Comment 3 dgrace 2005-11-14 16:08:59 EST
I found out about this bug today the hard way. Last section of console screen
before VM poweroff:

"Uncompressing Linux... Ok, booting the kernel.
PCI: Cannot allocate resource region 4 of device 0000:00:07.1
Red Hat nash version starting
sda: assuming drive cache: write through"

System boots fine going back to 2.6.9-11. 
Comment 17 Tim Morley 2006-01-11 07:59:03 EST
There is more detail of this problem on the vmware forums. It seems that the
latest mptscsi driver does a target reset on unused targets. The real hardware
doesn't support this, and the vmware emulation dies when this happens.

There is also a patch on the vmware forums that fixes this, it would be nice for
that to get into the distribution kernels.

See http://www.vmware.com/community/thread.jspa?messageID=306923 for more details!
Comment 18 Mike Christie 2006-01-11 13:30:33 EST
We should have something like that already. Could you try this kernel?
http://people.redhat.com/~jbaron/rhel4/RPMS.kernel/ These are experimental kernels.
Comment 19 Tim Morley 2006-01-12 09:05:09 EST
Thanks, I've tried the smp version of that kernel and it seems to work fine.

I take it kernels like these will end up in the next update?
Comment 20 Mike Christie 2006-01-12 12:53:02 EST
Thanks for testing. Yeah, U3.
Comment 23 Red Hat Bugzilla 2006-03-07 15:27:15 EST
An advisory has been issued which should help the problem
described in this bug report. This report is therefore being
closed with a resolution of ERRATA. For more information
on the solution and/or where to find the updated files,
please follow the link below. You may reopen this bug report
if the solution does not work for you.


Note You need to log in before you can comment on or make changes to this bug.