[Webinar] Learn how to a build a cloud-first strategyRegister Now

x
  • Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 927
  • Last Modified:

Terminal Server 2003 hangs after a number of hours

I have a very annoying problem. We have a Windows 2003 Terminal server Xeon 3Ghz with 3GB ram. Ten users
Ram and CPU never appeared to be over taxed bothered.

AV is Trend Micro with real time scanning disabled

Every other day, maybe consecutive days or work for a few days and then basically stop working . Users complain they can no longer access files on the main file server and if they log off , when they try and log back on the system says that it cannot load their user profile. (Local no roaming) Plenty of space on the C: Drive
We get different messages such as not being able to Windows cannot access the file gpt.ini for GPO,
An I/O operation initiated byt the Registry failed unrecoverably and a number of others.

I suspected system non-paged pool memory. I adjusted the registry as per some tech articles.

The really strange thing is that it only ever happens after 3:00 in the afternoon on an irregular basis.

Any ideas.

 
0
flynnpaul
Asked:
flynnpaul
1 Solution
 
ocon827679Commented:
Anything in the system event log?  What is your pagefile set to?  It should be 1.5x RAM at a minimum in a TS environment.
0
 
Malli BoppeCommented:
I think its the problem with the Trend.Stop the trend service to see if that makes any difference.
Also check the open files for each users.I
0
 
http:// thevpn.guruCommented:
What about resources....are u monitoring your CPU..RAM resources..you could create alerts..to warn you when low limits are reached.
0
VIDEO: THE CONCERTO CLOUD FOR HEALTHCARE

Modern healthcare requires a modern cloud. View this brief video to understand how the Concerto Cloud for Healthcare can help your organization.

 
jmerullaCommented:
Not sure if this helps but we're having the same or similar issue on our exchange server.  Wanted to put this online for anyone else having a similar problem since google isn't yielding much on my particular issue.  

Every day, for the past 4 days, my exchange server (has some network files), has done this non paged memory nonsense.

Event Type:     Error
Event Source:   Srv
Event Category: None
Event ID:       2000
Date:           11/4/2007
Time:           4:46:59 PM
User:           N/A
Computer:       <<servername>>
Description:
The server's call to a system service failed unexpectedly.

For more information, see Help and Support Center at  <http://go.microsoft.com/fwlink/events.asp> http://go.microsoft.com/fwlink/events.asp.
Data:
0000: 00 00 04 00 01 00 54 00   ......T.
0008: 00 00 00 00 d0 07 00 c0   ....Ð..À
0010: 00 00 00 00 0a 01 00 c0   .......À
0018: 00 00 00 00 00 00 00 00   ........
0020: 00 00 00 00 00 00 00 00   ........
0028: 34 03 bd 00               4.½.


Event Type:     Error
Event Source:   Srv
Event Category: None
Event ID:       2020
Date:           11/6/2007
Time:           8:50:37 AM
User:           N/A
Computer:       <<servername>>
Description:
The server was unable to allocate from the system paged pool because the pool was empty.

For more information, see Help and Support Center at  <http://go.microsoft.com/fwlink/events.asp> http://go.microsoft.com/fwlink/events.asp.
Data:
0000: 00 00 04 00 01 00 54 00   ......T.
0008: 00 00 00 00 e4 07 00 c0   ....ä..À
0010: 00 00 00 00 9a 00 00 c0   ....?..À
0018: 00 00 00 00 00 00 00 00   ........
0020: 00 00 00 00 00 00 00 00   ........
0028: 04 00 00 00  

After checking we found that  the biggest Paged pool memory hog is Tag Vmem with Bytes = 109887488 (related to Trend Micro).

After looking I found this article related to the issue (posted yesterday).
http://esupport.trendmicro.com/support/viewxml.do?ContentID=EN-1035794 however it relates to office scan 8 and we're running office scan 7.6 but it looks very familar.  

So we called TrendMicro.  They have had a few calls related to office scan 7.6 and engine version 8.550 (which is what we have).  However, they haven't gotten enough calls to release a hotfix for it so he gave us the following work arounds.
===============================
Workaround #1.

Enlarge paged pool - reboot required

Please open up the registry editor and edit the values specified below:

HKEY_LOCAL_MACNE\SYSTEM\CurrentControlSet\Control\Session Manager\Memory Management
Value: PagedPoolSize
Data: 0xFFFFFFFF


Workarpound #2.

Enable SystemMapView - reboot required

HKEY_LOCAL_MACNE\SYSTEM\CurrentControlSet\Services\VSApiNt\Parameters
Value: EnableSystemView
Data: 1  (DWORD)

Reboot the system.
======================

I think that is nonsense, and am uninstalling Trend and pushing Symantec on to the server from another office site.

Good luck, hope no one else is getting this stupid issue.
0
 
flynnpaulAuthor Commented:
I eventially found the problem to be a third party application which was casuing a memory leak.

Though in itself it was displaying any great number of errors; the problem was visible if I selected Handle Count under the Process screen in task manger. Gebneraly most apps will max out around 1-2K handles. This app was consuming upwards of 1990K handles which basically sucked all thh resources from the system. I obtained an update from the vendor and the problem was solved. (We had asked them earlier if there were any known problems with their application but they denied there were.. gees who would have thought !!)
0
 
zreismanCommented:
What third party app was it? I'm having the exact same issue with Trend Micro, the memory leak, and those same events in the event viewer.
0
 
flynnpaulAuthor Commented:
The appliocations was a locally developed App so probaly not your cause. The key here was to use task manager and select additional comlumns and select Handles. The maximum handles should be on any system is probaly around 8000. On our system we had processes using handles in execess of  a 190000. If you have such a process, then that would be the culprit
0

Featured Post

Important Lessons on Recovering from Petya

In their most recent webinar, Skyport Systems explores ways to isolate and protect critical databases to keep the core of your company safe from harm.

Tackle projects and never again get stuck behind a technical roadblock.
Join Now