Freebsd 8 Release /usr Die After host VMWARE Crash

[Diego Arias]

On Thu, Jul 8, 2010 at 6:18 AM, Simun Mikecin <numisemis at yahoo.com> wrote:

>
>
>
>
> ----- Original Message ----
> > From: Andrew Snow <andrew at modulus.org>
> > To: Simun Mikecin <numisemis at yahoo.com>; freebsd-fs at freebsd.org
> > Sent: Thu, July 8, 2010 9:59:48 AM
> > Subject: Re: Freebsd 8 Release /usr Die After host VMWARE Crash
> >
> > On 08/07/10 17:45, Simun Mikecin wrote:
> > > AFAIK virtual environments  ignore disk sync requests by default. For
> >example, in
> > > VirtualBox they  are ignored by default, by you could enable it if you
> want
> >(with
> > > a  performance penalty). Haven't used VMWare, so not 100% sure about
> it,
> >maybe
> > > someone more knowledgable with VMWare knows what it's defaults  are.
> >
> > VMware vSphere/ESX/ESXi makes all writes synchronous.
> >
> > (It is  terribly slow on RAID cards that lack battery-backed cache!)
>
>
> Thanx for the info.
> But, there is still a chance that the host environment on which (original
> poster's) VMware is running uses a lying ATA drive or some other storage
> that
> behaves the same as a lying ATA drive.
>
>
>
>
>
Hi there:

- I discard a CPU/memory corruption because 8 more VM were running there and
start without problem (Windows, not Unix ,Linux or BSD)
- Actually i dont know if it is FreeBSD Fault but its curious and dangerous.
- Its stock system so, no Async
- Any info you need that might be usefull just ask, im here to help and can
give you all the info you need
- The FSTAB is in the first mail and this report mount

/dev/ad0s1a on / (ufs, local)
devfs on /dev (devfs, local, multilabel)
/dev/ad0s1e on /tmp (ufs, local, soft-updates)
/dev/ad0s1f on /usr (ufs, local, soft-updates)
/dev/ad0s1d on /var (ufs, local, soft-updates)

- the machine its actually running from its backup, i have not touched the
FSTAB or mount stuff since installation
- the machine was installed on VirtualBox then migrated to VMWARE with the
convert utility from Virtualbox
- SAN Hard drives are all Fibre Channel On Raid 5, fully redundant path, the
SAN Switches on the chassis (Enclosure) didn't die when the enclosure start
malfunctioning, other SAN applications works without trouble after restart
on the same chasis as Virtual machines on the same LUN. the Blades Hard
drives are SAS 10000RPM

Diego Arias
-- 
mmm, interesante.....