Redireted AppData

[Guest ms-dog1]

Hi,

Currentlyt we are redirecting our AppData (via a Group Policy) to users H

drive which is housed on a MS 2003 Cluster. We have seen a few instances of

the cluster node is failing. Failover is also failing to happen, as well.

There is no clues in the event logs. Could this be due to the Appdata

redirection? Too many files open??? Is it time to do some SMB tuning on the

cluster and TS servers involved? I realize I have omitted pertinent info

such as amount of users, etc but just a possible confirmation this

investigation is heading in the right directon as to the possibilites for

this issue.

 

Thanks, Andie

[Guest Jeff Pitsch]

Re: Redireted AppData

 

And how have you determined that any of what you mentioned is causing this?

 

Jeff Pitsch

Microsoft MVP - Terminal Server

Citrix Technology Professional

Provision Networks VIP

 

Forums not enough?

Get support from the experts at your business

http://jeffpitschconsulting.com

 

ms-dog1 wrote:

> Hi,

> Currentlyt we are redirecting our AppData (via a Group Policy) to users H

> drive which is housed on a MS 2003 Cluster. We have seen a few instances of

> the cluster node is failing. Failover is also failing to happen, as well.

> There is no clues in the event logs. Could this be due to the Appdata

> redirection? Too many files open??? Is it time to do some SMB tuning on the

> cluster and TS servers involved? I realize I have omitted pertinent info

> such as amount of users, etc but just a possible confirmation this

> investigation is heading in the right directon as to the possibilites for

> this issue.

>

> Thanks, Andie

[Guest Vera Noest [MVP]]

Re: Redireted AppData

 

Impossible to say with no facts to go on, but it seems unlikely to

me.

Have you installed the recommended hotfixes for clusters?

 

935640 - Recommended hotfixes for Windows Server 2003 Service Pack

2-based server clusters

http://support.microsoft.com/?kbid=935640

 

Does the node really fail, or is it "hanging", i.e. is a fail-over

actually attempted and fails, or is it not attemped at all?

What does the cluster log tells you?

 

For troubleshooting methods, check:

 

815267 - How to enable User Mode Hang Detection on a server cluster

in Windows Server 2003 and in Windows 2000 Server SP4

http://support.microsoft.com/kb/815267/en-us

_________________________________________________________

Vera Noest

MCSE, CCEA, Microsoft MVP - Terminal Server

TS troubleshooting: http://ts.veranoest.net

___ please respond in newsgroup, NOT by private email ___

 

=?Utf-8?B?bXMtZG9nMQ==?= <msdog1@discussions.microsoft.com> wrote

on 21 aug 2007 in microsoft.public.windows.terminal_services:

> Hi,

> Currentlyt we are redirecting our AppData (via a Group Policy)

> to users H drive which is housed on a MS 2003 Cluster. We have

> seen a few instances of the cluster node is failing. Failover

> is also failing to happen, as well. There is no clues in the

> event logs. Could this be due to the Appdata redirection? Too

> many files open??? Is it time to do some SMB tuning on the

> cluster and TS servers involved? I realize I have omitted

> pertinent info such as amount of users, etc but just a possible

> confirmation this investigation is heading in the right directon

> as to the possibilites for this issue.

>

> Thanks, Andie

[Guest ms-dog1]

Re: Redireted AppData

 

All valid points! Just beginning the investigation. The only deduction at

this time is the problem began happening after we deployed Appdata

Redirection but of course may not be related!

 

Thanks!

 

"Jeff Pitsch" wrote:

> And how have you determined that any of what you mentioned is causing this?

>

> Jeff Pitsch

> Microsoft MVP - Terminal Server

> Citrix Technology Professional

> Provision Networks VIP

>

> Forums not enough?

> Get support from the experts at your business

> http://jeffpitschconsulting.com

>

> ms-dog1 wrote:

> > Hi,

> > Currentlyt we are redirecting our AppData (via a Group Policy) to users H

> > drive which is housed on a MS 2003 Cluster. We have seen a few instances of

> > the cluster node is failing. Failover is also failing to happen, as well.

> > There is no clues in the event logs. Could this be due to the Appdata

> > redirection? Too many files open??? Is it time to do some SMB tuning on the

> > cluster and TS servers involved? I realize I have omitted pertinent info

> > such as amount of users, etc but just a possible confirmation this

> > investigation is heading in the right directon as to the possibilites for

> > this issue.

> >

> > Thanks, Andie

>

[Guest Helge Klein]

Re: Redireted AppData

 

The number of open files might negatively affect a file server -

cluster node or not. But the limit for 32 bit Windows is relatively

high - around 100,000 concurrently open files, I think. That numer is

by no means "official" and depends, of course, on installed software

and hardware.

 

I recommend analyzing the cluster log and opening a support call with

MS if necessary.

 

I hope this helps.

 

Helge

 

On 21 Aug., 22:10, ms-dog1 <msd...@discussions.microsoft.com> wrote:

> All valid points! Just beginning the investigation. The only deduction at

> this time is the problem began happening after we deployed Appdata

> Redirection but of course may not be related!

>

> Thanks!

>

> "Jeff Pitsch" wrote:

> > And how have you determined that any of what you mentioned is causing this?

>

> > Jeff Pitsch

> > Microsoft MVP - Terminal Server

> > Citrix Technology Professional

> > Provision Networks VIP

>

> > Forums not enough?

> > Get support from the experts at your business

> >http://jeffpitschconsulting.com

>

> > ms-dog1 wrote:

> > > Hi,

> > > Currentlyt we are redirecting our AppData (via a Group Policy) to users H

> > > drive which is housed on a MS 2003 Cluster. We have seen a few instances of

> > > the cluster node is failing. Failover is also failing to happen, as well.

> > > There is no clues in the event logs. Could this be due to the Appdata

> > > redirection? Too many files open??? Is it time to do some SMB tuning on the

> > > cluster and TS servers involved? I realize I have omitted pertinent info

> > > such as amount of users, etc but just a possible confirmation this

> > > investigation is heading in the right directon as to the possibilites for

> > > this issue.

>

> > > Thanks, Andie