Cluster disk chkdsk

Hi,

I have a 2 node W2K SP4 cluster with EMC CX500 SAN fc storage. One of the
cluster disk resources is saying it is corrupt and needs chkdsk to be run.
This appears to be initiated automatically when the resource/group is brough
online.
I have read http://support.microsoft.com/kb/176970/en-us and this seems all
you need to do is allow it to run to completion.
However a colleague has said tghis does not work (or at leat did not for him
on a prior occasion) and that you need to shut down one node, disable the
cluster driver on the other, then reboot and run chkdsk in non-cluster mode.
Is this necessary as the MS article says this is only reqd for NT4 cluster
disks.

TIA,

Geejay

>
> I have a 2 node W2K SP4 cluster with EMC CX500 SAN fc storage. One of the
> cluster disk resources is saying it is corrupt and needs chkdsk to be run.
> This appears to be initiated automatically when the resource/group is
brough
> online.

correct, and you will find the output file of this chkdsk in
c:\winnt\cluster

> I have read http://support.microsoft.com/kb/176970/en-us and this seems
all
> you need to do is allow it to run to completion.

yes, leave it run until finished, the physical disk resource will display
"online pending" in cluster administrator

> However a colleague has said tghis does not work (or at leat did not for
him
> on a prior occasion) and that you need to shut down one node, disable the
> cluster driver on the other, then reboot and run chkdsk in non-cluster
mode.

only if the "cluster initiated chkdsk upon an online" is not fixing your
problem


> Is this necessary as the MS article says this is only reqd for NT4 cluster
> disks.

in early days it was the only way, but now just let the disk on "online
pending" until finished and examine the output.

HTH,
Edwin.

Thank you for the response. When attempting to bring the disk online, chkdsk
is started, but it fails after a short time and the log file says the system
does not have exclusive access to the disk (even though it not in online
state?).
All the shares for that drive have it as a dependency so I thought would
wait for it to come online and not be able to open handle to it.
Is there a way to determine what has the disk ??

Regards,

Geejay

"Edwin vMierlo [MVP]" <> wrote in
message news:ucFc%...
>
>>
>> I have a 2 node W2K SP4 cluster with EMC CX500 SAN fc storage. One of the
>> cluster disk resources is saying it is corrupt and needs chkdsk to be
>> run.
>> This appears to be initiated automatically when the resource/group is
> brough
>> online.
>
> correct, and you will find the output file of this chkdsk in
> c:\winnt\cluster
>
>> I have read http://support.microsoft.com/kb/176970/en-us and this seems
> all
>> you need to do is allow it to run to completion.
>
> yes, leave it run until finished, the physical disk resource will display
> "online pending" in cluster administrator
>
>> However a colleague has said tghis does not work (or at leat did not for
> him
>> on a prior occasion) and that you need to shut down one node, disable the
>> cluster driver on the other, then reboot and run chkdsk in non-cluster
> mode.
>
> only if the "cluster initiated chkdsk upon an online" is not fixing your
> problem
>
>
>> Is this necessary as the MS article says this is only reqd for NT4
>> cluster
>> disks.
>
> in early days it was the only way, but now just let the disk on "online
> pending" until finished and examine the output.
>
> HTH,
> Edwin.
>
>

any virus scanners ?
disable them, or exclude this disk from scanning for now


"gjb" <> wrote in message
news:%...
> Thank you for the response. When attempting to bring the disk online,
chkdsk
> is started, but it fails after a short time and the log file says the
system
> does not have exclusive access to the disk (even though it not in online
> state?).
> All the shares for that drive have it as a dependency so I thought would
> wait for it to come online and not be able to open handle to it.
> Is there a way to determine what has the disk ??
>
> Regards,
>
> Geejay
>
> "Edwin vMierlo [MVP]" <> wrote in
> message news:ucFc%...
> >
> >>
> >> I have a 2 node W2K SP4 cluster with EMC CX500 SAN fc storage. One of
the
> >> cluster disk resources is saying it is corrupt and needs chkdsk to be
> >> run.
> >> This appears to be initiated automatically when the resource/group is
> > brough
> >> online.
> >
> > correct, and you will find the output file of this chkdsk in
> > c:\winnt\cluster
> >
> >> I have read http://support.microsoft.com/kb/176970/en-us and this
seems
> > all
> >> you need to do is allow it to run to completion.
> >
> > yes, leave it run until finished, the physical disk resource will
display
> > "online pending" in cluster administrator
> >
> >> However a colleague has said tghis does not work (or at leat did not
for
> > him
> >> on a prior occasion) and that you need to shut down one node, disable
the
> >> cluster driver on the other, then reboot and run chkdsk in non-cluster
> > mode.
> >
> > only if the "cluster initiated chkdsk upon an online" is not fixing your
> > problem
> >
> >
> >> Is this necessary as the MS article says this is only reqd for NT4
> >> cluster
> >> disks.
> >
> > in early days it was the only way, but now just let the disk on "online
> > pending" until finished and examine the output.
> >
> > HTH,
> > Edwin.
> >
> >
>
>