[THIN] Re: XenApp 5 - 2008 x64 Slow
- From: Andy Friar <Andy.Friar@xxxxxxxxxxx>
- To: "thin@xxxxxxxxxxxxx" <thin@xxxxxxxxxxxxx>
- Date: Wed, 12 Jan 2011 11:26:16 +0000
Sorry for the delay getting back, man flu has taken its toll on me.
Bearing in mind this is on a server previously mentioned but with PM disabled @
around 60users (Desktops)
With Process\Page Faults/sec the worst offenders are
Iexplore max @ 225k average 2k
System, max @ 120k, average 5k
EMAgentAssist max @ 40k average of 400
Spoolsv max @ 22k average 2.5k
These are very selective and over a small time window though.
Redirector\Current Commands is still at 0 though,
Going back to Appsense, PM was disabled on Thursday.
Monday and Tuesday showed the biggest load with 80 concurrent desktops with no
lag, or delays. Logon's are still quick and the server feeling responsive.
The call has been logged with Appsense, hopefully they can find the cause of
this.
Appsense Agents were upto date as of Jan4th update as this had numerous EM
fixes.
Rgds
Andy
-----Original Message-----
From: thin-bounce@xxxxxxxxxxxxx [mailto:thin-bounce@xxxxxxxxxxxxx] On Behalf Of
jeremy@xxxxxxxxxxxxxxxxxxxx
Sent: 07 January 2011 00:19
To: thin@xxxxxxxxxxxxx
Subject: [THIN] Re: XenApp 5 - 2008 x64 Slow
Hi Andy,
Forgot about that AppSense issue James is having.
I would only ever expect to see Redirector\Current Commands at 0 in a Novell
environment, or in very quiet times. If the systems are using SMB for file
services, that value should be much higher.
With the page faults per second, add the Process\Page Faults/sec counter for
ALL processes NOT the _Total. There will be a lot here on a busy server with
lots of users. Once added, patiently scroll through them and check the values,
don't focus on the graph. I had one the other day where RTVSCAN.exe (the
realtime scan engine from Symantec) was doing tens of thousands of page faults
per second, due to the fact that the customer had not setup any exclusions as
per best practice. This was causing a performance hit as more users were
logging in. Maybe checkout the AppSense processes. There is also a registy
value you can add that will change the instance # to a PID, so you can match
the process back to a user, and then you can go and interview users to find out
what they are doing. I've seen Excel and Outlook plugins causing issues too.
Do you use HP print drivers? Search the
"%SystemRoot%\System32\spool\drivers" folder for hpzbdi64.msi on 64-bit
systems. This can cause major pauses. I've got a script that will clean this
up. Let me know and I'll send it to you.
Can you let us know if AppSense is the issue, as I'm doing a Performance
Manager rollout next week, and I don't want any surprises?
Cheers,
Jeremy.
> At 60 you'll see lag in excel or outlook, differing amount of lag but
> it's very hard to pinpoint at what user level lag starts.
>
> Redirector\Current Commands is 0, not sure if it's reporting properly
> as it's never moved.
> %Page File Usage is 0
> Pages/second range from 0->2500 averaging 175 over default time window
>
> Disabling Trend Serverprotect made no appreciable difference in
> performance.
>
> Print Drivers are up for discussion though, all past a few runs of
> stressprint but real word some are crashing the citrix print manager
> service. - needs identifying.
>
> I've got one of the hotfixes from that list which updates the smbv1
> driver applying tonight KB981964 (superb list of hotfixes btw)
>
> Thanks for the input
>
> Andy
>
>
>
>
> From: thin-bounce@xxxxxxxxxxxxx [mailto:thin-bounce@xxxxxxxxxxxxx] On
> Behalf Of Jeremy (J House Consulting)
> Sent: 06 January 2011 15:56
> To: thin@xxxxxxxxxxxxx
> Subject: [THIN] Re: XenApp 5 - 2008 x64 Slow
>
> Hi Andy,
>
> Is the lag/pause random or consistent. Ie. At 60 users logged in, will
> you always see a lag selecting a cell in Excel?
>
> What are your Redirector\Current Commands sitting at, as the value
> below is missing?
>
> I would also be checking %Page File Usage and Memory:Pages/Second counter.
> There could be a process paging big time.
>
> Known issues could be AV realtime scanning, Print Drivers, SMB V1.0
> tuning.
>
> Carl posted a heap of relevant hotfixes a couple of months ago.
>
> Cheers,
> Jeremy.
>
> From: thin-bounce@xxxxxxxxxxxxx [mailto:thin-bounce@xxxxxxxxxxxxx] On
> Behalf Of Andy Friar
> Sent: Thursday, 6 January 2011 11:10 PM
> To: thin@xxxxxxxxxxxxx
> Subject: [THIN] XenApp 5 - 2008 x64 Slow
>
> I'm have a problem with the following Windows 2008 SP2 Enterprise x64
> XenApp 5 Enterprise R01
>
> Boxes are Dell blade 8 core + HT 49Gb RAM Local manadatory profiles
> with full Appsense suite utilising EM + Personalisation service.
> Desktop / My Documents / Favourites redirected to a Netapp Filer
> running CIFS.
>
> The more users that come onto the box the slower the machine gets.
>
> @ 0-45 users the box is responsive, logon times are good but it still
> has this apparent lag when accessing local apps @ 50 the lag is
> getting worse sometimes freezes but always comes back.
> @ 60 logon times are substantial, just selecting a cell in excel can
> take
> 30+ seconds.
>
> Performance counters seem fine
>
> 14% processor time
> 25k context switches / sec no massive jumps
> 0 Processor queue length
> 99% disk idle
> 0 disk queue length
> 15-20k threads
> 0 Redirector\Current Commands
>
> The only large one is system calls / sec which can be anywhere from
> 80k to 1.5m, I've cut down the apps but system calls are still high.
>
> To me the server feels as if it has no cpu left, but given the
> resources free it looks more like an i/o problem, I've tuned the smb
> with the following
>
> [HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Lanmanserver\Par
> ameters]
> "MaxWorkItems"=dword:00002000
> "MaxMpxCt"=dword:00000800 (subsequently
> has been changed to match the Netapp)
> "MaxRawWorkItems"=dword:00000200
> "MaxFreeConnections"=dword:00000064
> "MinFreeConnections"=dword:00000020
>
> [HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Lanmanworkstatio
> n\Parameters]
> "MaxCmds"=dword:00000800
>
> [HKEY_LOCAL_MACHINE\System\CurrentControlSet\Control\Session
> Manager\Configuration Manager]
> "RegistryLazyFlushInterval"=dword:0000003C
>
>
> The NetApp filer has had cifs.max_mpx set to 1124 so I've subsequently
> changed the citrix servers to match this number rather than the 2048
> previously.
> It's running around 30%cpu 2000 io/sec 98% cache hit.
>
> Any suggestions on where else I might look into what is causing the
> bottleneck?
>
> Thanks
>
> Andy
>
>
>
************************************************
For Archives, RSS, to Unsubscribe, Subscribe or set Digest or Vacation mode use
the below link:
//www.freelists.org/list/thin
Follow ThinList on Twitter
http://twitter.com/thinlist
************************************************
************************************************
For Archives, RSS, to Unsubscribe, Subscribe or
set Digest or Vacation mode use the below link:
//www.freelists.org/list/thin
Follow ThinList on Twitter
http://twitter.com/thinlist
************************************************
Other related posts:
