On Fri, 2009-02-06 at 14:46 -0500, Jason Pyeron wrote:
> > -----Original Message-----
> > From: centos-bounces@xxxxxxxxxx
> > [mailto:centos-bounces@xxxxxxxxxx] On Behalf Of JohnS
> > Sent: Friday, February 06, 2009 12:52
> > To: CentOS mailing list
> > Subject: Re: VM Ware guest shutdown hangs withhigh
> > loadin uninterruptible sleep was RE: Deciphering top's data
> >
> >
> > On Fri, 2009-02-06 at 18:17 +0100, Tru Huynh wrote:
> > > On Fri, Feb 06, 2009 at 11:57:15AM -0500, Jason Pyeron wrote:
> > > >
> > > > $ ps -f 8112
> > > > UID PID PPID C STIME TTY STAT TIME CMD
> > > > root 8112 1 17 11:49 ? Ds 0:51
> > > > /usr/lib/vmware/bin/vmware-vmx -C
> > /var/lib/vmware/VMs/proxy/RHEL.vmx
> > > have you applied the recommended vmware boot flags for your vm?
> > >
> > > other ideas but this is becoming off topic for CentOS since
> > it's seems
> > > to be a vmware server issue for a RHEL guest..
> > > - moved vmware guest and a vmware console question left unanswered.
> > > - damaged vmdk
> > > - anything on the vmware logs?
> > > ...
> > >
> > > Tru
> > ------
> > And the only other reason I can think of why it is happening
> > is your possibly running Multipath but that Issue is supposed
> > to be fixed in version 4.5? Rh released an update to
> > Multipathd to fix this. Should have trickled down to centos as well.
>
> How do I check this?
--------
rpm -qa | grep mapper
Will show you your current version.
The fix started in 4.5 and for version 5 release 1 of RHEL. That being
said does not mean it is in CentOS.
Centos 4.7 should have it as below.
device-mapper-multipath-0.4.5-31.el4.i386.rpm
device-mapper-1.02.25-2.el4.i386.rpm
They are up to date with ftp.redhat!
What you are describing only was supposed to happen on ESX and not
VMWare Server so you may be totally into a different problem. LIKE,
running out ram? A 64 bit system takes more ram to actually allocate ram
than a 32 bit system. Check /var/log/vmwarev/mware-serverd.log ? Check
your disks for bad sectors (local disks). Memory Errors. Bad firmware
on the Raid Controller or controller going bad under heavy loads.
Loading images to boot over the network? Network Latency? You need some
kind of log output from the raid controller. I take it it is a pretty
high end machine with all those images so you should have some kind of
monitoring process like Dell OMSA on it to get that info. Top and vmstat
is not going to show you what the problem is. That's fantasy world. All
that shows is memory allocation, cpu usage etc etc. You need some log
files to ponder over and look at possible hardware. I am just assuming
it is a hardware prob since you updated and because the multipath
problem is not supposed to affect vmw-server.
Ohh how you check that? It is called Red Hat Bugzilla and keeping up
with the technology you support like you should be doing.
JohnStanley
_______________________________________________
CentOS mailing list
CentOS@xxxxxxxxxx
http://lists.centos.org/mailman/listinfo/centos
