asked on
SVCHOST.exe consuming all memory in Windows 2008 R2
As the title states, I have an issue where SVCHOST.exe is slowly consuming all the memory on a server running Windows 2008 R2. The services being used by the svchost.exe PID are: nsi, netprofm, FontCache and EventSystem. Within two days it will consume all memory causing IIS to throw out errors.
I've tried various hotfixes, but none are applicable to this system. The issue appeared within the past 2 months. I've tried updating the drivers for the nic as well.
About the server: 2008 R2 Server, using IIS (Primary role), anti-virus is System Center 2012 Endpoint Protection (and SCCM 2012 agent), Landesk Agent, Broadcom BCM5716, 12GB of memory and it is the head node of a clustering search application where it displays results via a webpage (IIS) and communicates to four other nodes. The system has all MS security updates installed.
I've tried various hotfixes, but none are applicable to this system. The issue appeared within the past 2 months. I've tried updating the drivers for the nic as well.
About the server: 2008 R2 Server, using IIS (Primary role), anti-virus is System Center 2012 Endpoint Protection (and SCCM 2012 agent), Landesk Agent, Broadcom BCM5716, 12GB of memory and it is the head node of a clustering search application where it displays results via a webpage (IIS) and communicates to four other nodes. The system has all MS security updates installed.
Windows Server 2008Microsoft Server OSMicrosoft IIS Web Server
Last Comment
futureman0
ASKER
NSI, Network Store Interface Service, appears to be the culprit. I stopped the other three services, but they didn't have any affect till NSI was stopped.
SVCHOST.exe usage dropped back to normal once NSI was stopped (or attempted to) and the system became unstable and had to be reboot.
I'm just using the task manager, going to the services tab and the PID number is a column.
SVCHOST.exe usage dropped back to normal once NSI was stopped (or attempted to) and the system became unstable and had to be reboot.
I'm just using the task manager, going to the services tab and the PID number is a column.
Well it seems that is could be caused by several things like AV, malware, excessive pings, bad patches, and/or driver and firmware issues.
I would:
- Check to see if there are any updates for the NIC drivers and firmware.
- Make sure you server is free from infections
- Check your AV configuration
- Update your AV software
Also, are there any network related tasks that the server does on a regular basis that would use ping a lot?
I would:
- Check to see if there are any updates for the NIC drivers and firmware.
- Make sure you server is free from infections
- Check your AV configuration
- Update your AV software
Also, are there any network related tasks that the server does on a regular basis that would use ping a lot?
ASKER
I've already tried updating the NIC drivers and it didn't help. The anti-virus, SCEP 2012, has not reported any infections. Additionally looking at the network firewall I'm not see any sort of suspicious incoming/outgoing traffic.
This is the primary system in a cluster, and ping is used to see if the other nodes are online. This configuration has been running for over 3 years without a problem - any windows updates that might have caused this?
This is the primary system in a cluster, and ping is used to see if the other nodes are online. This configuration has been running for over 3 years without a problem - any windows updates that might have caused this?
It's definitely possible an update caused this but I'm seeing people having this issue dating back several years.
I would run a manual full scan with the AV and possibly use a few different products. Also check the processes running to see if there's any suspicious activity. In my experience, just because the AV doesn't report anything doesn't mean the system isn't infected.
I would run a manual full scan with the AV and possibly use a few different products. Also check the processes running to see if there's any suspicious activity. In my experience, just because the AV doesn't report anything doesn't mean the system isn't infected.
Use following command to without quote to seperate all the services in individual svchost.exe "sc <service_name> type= own" , here you should use all the services under the same PID . Thereafter restart those services . In order to confirm whether its working or not ,you can use "tasklist /svc >tasks.txt" and then open tasks.txt file to validate all services are running in seperate svchost container with seperate PID.
This will give you clear picture of the culprit, once you have real culprit it will be easy to chalk-out solution for the issue.
Updating these files to the latest may fix the issue Nsi.dll, siproxy.sys, Nsisvc.dll, Winnsi.dll
Would also request to block ICMP i.e ping ack packets , if possible you can check whats the amount of ICMP packets in netmon trace.
This will give you clear picture of the culprit, once you have real culprit it will be easy to chalk-out solution for the issue.
Updating these files to the latest may fix the issue Nsi.dll, siproxy.sys, Nsisvc.dll, Winnsi.dll
Would also request to block ICMP i.e ping ack packets , if possible you can check whats the amount of ICMP packets in netmon trace.
SOLUTION
THIS SOLUTION ONLY AVAILABLE TO MEMBERS.
View this solution by signing up for a free trial.
Members can start a 7-Day free trial and enjoy unlimited access to the platform.
ASKER CERTIFIED SOLUTION
THIS SOLUTION ONLY AVAILABLE TO MEMBERS.
View this solution by signing up for a free trial.
Members can start a 7-Day free trial and enjoy unlimited access to the platform.
ASKER
I found the reason what was causing the issue
BTW, what are you using to find the services running in that instance of svchost?