[THIN] Re: Multiple servers load set to 10000 at same time
- From: "Jeremy Saunders" <jeremy@xxxxxxxxxxxxxxxxxxxx>
- To: <thin@xxxxxxxxxxxxx>
- Date: Thu, 18 Oct 2012 17:56:49 +0800
Hi Ang,
Two things:
1) Yes, you can run Perfmon from a remote machine.
2) Do you have the Health Monitoring & Recovery (HMR) tests in place?
The default recovery action for the XML Service test is set to Remove Server
from Load Balancing. This can cause confusion, so I change it to Alert Only,
allowing administrators to monitor it instead. If a server is removed from
Load Balancing due to a Health Monitoring Test failure, it will remain out
of Load Balancing until one of the following actions occurs: The enablelb
command is executed from a command prompt or the server is rebooted.
But I?d say there is some underlying problem if you can?t run any of your
tools from the servers themselves.
Cheers,
Jeremy.
From: thin-bounce@xxxxxxxxxxxxx [mailto:thin-bounce@xxxxxxxxxxxxx] On Behalf
Of Angela Smith
Sent: Thursday, 18 October 2012 5:16 PM
To: thin@xxxxxxxxxxxxx
Subject: [THIN] Re: Multiple servers load set to 10000 at same time
Hi
Yes, when the server reboots, all is working OK. Server isn't set to 10000
and there are no down sessions. However, issue re-occurs after a few hours.
Issue is happening on random servers so I don't think its a WMI Issue (very
unlikely that WMI is broken on 6 different servers). Nothing has changed on
our servers either.
We have a custom Load Evaluator which we have been using for several years
without issue. It basically checks CPU, Memory and User Load. Load is
definitely not an issue as the servers only have 15 users on them when the
issue occurs whereas they normally have 40. I cannot easily check resources
as all GUI tools fail to load. Task Manager, Perfmon, Sysinternals Process
Explorer etc don't launch. Whats the best way to see what is happening with
CPU and Memory? Perhaps I can run the tool remotely? As a test should I
change all servers to the Adv evaluator?
Ive never had any issues with memory. Each server has 16GB RAM. In
relation to CPU, Ive had the occasional process go crazy but its never used
all cores. Servers have 2 x Quad core CPU's. Print Spooler and Antivirus
all appears OK in that the service is started but I'm working blind as I
cannot see the processes due to the tools not launching
Perfmon wont load so I cannot rell is perfmon counters are corrupt
We do have SCOM. How do I reload the MOF?. Would the MOF generate a ICA
down session? Every server that goes to 10000 has a down session so its
definitely causing the issue or is related..
Thanks
Ang
_____
CC: thin@xxxxxxxxxxxxx
From: magnus@xxxxxxxx
Subject: [THIN] Re: Multiple servers load set to 10000 at same time
Date: Wed, 17 Oct 2012 22:17:55 -0400
To: thin@xxxxxxxxxxxxx
We notice wmi issues as well, this was do to MOM/SCOM. We had to reload the
mof's to fix it.
Sent from my iPhone
On Oct 17, 2012, at 20:15, Alan Tropper <Alan.Tropper@xxxxxxxxxxxxx> wrote:
Hi,
I have had WMI issues also in the past and had to re-build WMI components as
below:
http://musumeci.blogspot.com.au/2009/01/ms-rebuild-wmi.html
Cheers
Al
From: thin-bounce@xxxxxxxxxxxxx [mailto:thin-bounce@xxxxxxxxxxxxx] On Behalf
Of julien, Sybille
Sent: Wednesday, 17 October 2012 6:56 PM
To: thin@xxxxxxxxxxxxx
Subject: [THIN] Re: Multiple servers load set to 10000 at same time
did you check if your perfmon counter is corrupted ?
http://support.citrix.com/article/CTX129350
<~WRD000.jpg> Julien
De : Jeremy Saunders <jeremy@xxxxxxxxxxxxxxxxxxxx>
À : thin@xxxxxxxxxxxxx
Envoyé le : Mercredi 17 octobre 2012 10h38
Objet : [THIN] Re: Multiple servers load set to 10000 at same time
The server thinks it?s at full load. What?s the load evaluator assigned to
it?
If Task Manager is not loading, it sounds like perhaps a CPU/memory
utilisation issue, which is effecting the load evaluator, which is placing
the server at full load.
So what?s the CPU and memory utilisation on the server? Could you have a
rouge service, such as antivirus causing problems? What about the print
spooler and Citrix print manager services?
Cheers,
Jeremy.
From: thin-bounce@xxxxxxxxxxxxx [mailto:thin-bounce@xxxxxxxxxxxxx] On Behalf
Of Hamilton, Ronnie
Sent: Wednesday, 17 October 2012 4:00 PM
To: thin@xxxxxxxxxxxxx
Subject: [THIN] Re: Multiple servers load set to 10000 at same time
Hi,
You say that even a reboot doesn?t help ? Is that correct ?
I had an RDP session in a down state that was doing the same thing I had to
reset it via the Microsoft Terminal Services Console and this resolved that
issue.
It did take a few times and I think I deleted my admin user profile as well
as they don?t get deleted like all our user ones.
Regards,
Ronnie
From: thin-bounce@xxxxxxxxxxxxx [mailto:thin-bounce@xxxxxxxxxxxxx] On Behalf
Of Angela Smith
Sent: 17 October 2012 05:10
To: thin@xxxxxxxxxxxxx
Subject: [THIN] Re: Multiple servers load set to 10000 at same time
Hi
I jumped the gun.. PSE450R05W2K3026 is in Rollup 6 which I already have
installed.. Back to the drawing board!
Ang
From: angela_smith9@xxxxxxxxxxx
To: thin@xxxxxxxxxxxxx
Subject: [THIN] Re: Multiple servers load set to 10000 at same time
Date: Wed, 17 Oct 2012 15:07:23 +1100
Hi
This patch looks promising.. Only issue is I cannot find it.
http://support.citrix.com/article/CTX123956
This has been superseded by R07 patch which I dont want to install right
now. Im currently running R06.
Does anyone have hotfix PSE450R05W2K3026 that I can download for x86
systems?
Thanks
Ang
CC: thin@xxxxxxxxxxxxx
From: magnus@xxxxxxxx
Subject: [THIN] Re: Multiple servers load set to 10000 at same time
Date: Tue, 16 Oct 2012 22:13:30 -0400
To: thin@xxxxxxxxxxxxx
Check the nonpagepool memory usage in taskmgr. I have seen this usage spike
which was due to an ms bug and XTE.exe. Good news there is a fix for it.
936655 I beleive the number is
Sent from my iPhone
On Oct 16, 2012, at 18:10, Angela Smith <angela_smith9@xxxxxxxxxxx> wrote:
Hi
I think my issue is different. All user sessions on these servers lock up
and they end up calling the helpdesk to reset them.. I'm getting down
sessions on the same servers each day (even after a reboot).. I applied
CTX112103 but it didn't make any difference..
Any other ideas apart from rebuilding?
Thanks
Ang
Subject: [THIN] Re: Multiple servers load set to 10000 at same time
Date: Mon, 15 Oct 2012 12:48:11 +0100
From: ronnie.hamilton@xxxxxxx
To: thin@xxxxxxxxxxxxx
Hi Angela,
This does ring a bell and I?m sure I have one server with this patch and
made no difference to me, but the csrss.exe process wasn?t showing a high
CPU usage for me.
It more just creating error in console Session Down state every few mins. It
only stopped users connecting on the very odd ocasion
Regards,
Ronnie
From: thin-bounce@xxxxxxxxxxxxx [mailto:thin-bounce@xxxxxxxxxxxxx] On Behalf
Of Angela Smith
Sent: 15 October 2012 12:36
To: thin@xxxxxxxxxxxxx
Subject: [THIN] Re: Multiple servers load set to 10000 at same time
Hi Ronnie
Unfortunately rebuilding wont be an option for me.. Did you try:
http://support.microsoft.com/kb/934330
My entire farm will reboot tonight. If issue re-occurs then I might try
this patch..
Thanks
Ang
Subject: [THIN] Re: Multiple servers load set to 10000 at same time
Date: Mon, 15 Oct 2012 12:25:28 +0100
From: ronnie.hamilton@xxxxxxx
To: thin@xxxxxxxxxxxxx
Hi Angela,
I get this from time to time, server doesn?t always report full load but
there?s always a session down state, and a reboot usually clears.
I had a look at the following thread but never actually got to the bottom of
it. I am currently rebuilding all my server to upgrade to Office 2010 and
hoping that the rebuild does the trick.
http://forums.citrix.com/message.jspa?messageID=1386285#1386285
It?s seems to be going well as all the new builds have not reports the
issue, but one of the pre Office 2010 builds did yesterday.
Regards,
Ronnie
From: thin-bounce@xxxxxxxxxxxxx [mailto:thin-bounce@xxxxxxxxxxxxx] On Behalf
Of Angela Smith
Sent: 15 October 2012 11:54
To: thin@xxxxxxxxxxxxx
Subject: [THIN] Multiple servers load set to 10000 at same time
Hi
I have an issue on my XenApp 4.5 farm where 4 servers are reporting Server
load 10000. I tried to RDP to the servers and launch Task Manager.
Unfortunately this doesn't load on any of the 4 servers. I tried to use
Process Explorer from Sysinternals tools but that also didn't load. What I
did notice though was all 4 servers had a session in a down state which I
couldn't reset. It simply came back each time. Error.jpg attached showing
issue. I don't believe Im having a resource issue but I cannot be 100% sure.
I tried to change my Load Evaluator from Advanced to Default as a test but
this made no difference. Also tried to clear some sessions but load didn't
change, still set to 10000.
Can anyone suggest how I can troubleshoot this 10000 Load issue without
rebooting server? FYI I do reboot my servers nightly but this issue has
happened a few times now. Also is there a way/tool to check system
resources when Task Manager doesn't work?
Thanks
Ang
Visit our website : http://www.ltai.ie/
__________________________________________
Lufthansa Technik Airmotive Ireland Limited. Registered in Ireland. Reg. No.
45999. Registered Office: Naas Road, Rathcoole, Co.Dublin.
Lufthansa Technik Airmotive Ireland Leasing Limited. Registered in Ireland.
Reg. No. 140891. Registered Office: Naas Road, Rathcoole, Co.Dublin.
__________________________________________
The information in this email and in any attachments is confidential and may
be privileged. If you are not the intended recipient, please destroy this
message, delete any copies held on your systems and notify the sender by
return email. You should not read, retain, copy, disseminate, distribute,
disclose or use this email or its contents in any way. Any such action is
strictly prohibited. Thank you.
This message may contain privileged and confidential information and is
intended for the exclusive use of the addressee(s). You must not disclose
this communication to anyone without the prior consent of the Department for
Child Protection (DCP). If you have received this email in error, please
notify us by return mail, delete it from your system and destroy all copies.
DCP has exercised care to avoid errors in the information contained in this
email but does not warrant that it is error or omission free.
Other related posts:
