Hi,
I have an ASP.NET application that runs on its own dedicated server.
About 4 weeks ago it started to respond to quite a few requests with request timed out errors. It has steadily degraded to the point where it is basically unavailable 90% of the time. As you may guess this has become a real issue as it is a business critical application.
I have ran a SQLProfiler trace and found that the database (which is on a different server) is performing great.
I have ran PerfMon on the web server and have seen some rather alarming (or so my uneducated developer eye tells me![[dazed]](/data/assets/smilies/dazed.gif "[dazed] [dazed]")
) results. Here is a synopsis:
Available MBytes: Avg. 2075 Max. 2092
Page Faults/sec: Avg. 33.712 Max. 3639.389
Page Reads/sec: Avg. 0.208 Max. 126.534
Request Execution Time: Avg. 2144 Max. 165058
Request Wait Time: Avg. 1 Max. 31
Current Connections: Avg. 0 Max. 12
Connection Attempts/sec: Avg. 59652.429 Max. 286331740
Bytes Received/sec: Avg. 2597.381 Max. 246678.701
Bytes Sent/sec: Avg. 52138.766 Max. 12314916
Worker Process Restarts: Avg. 0 Max. 0
I have the App Pool to recycle every 60 minutes (it used to be after 35000 requests but I changed it incase this was the problem)
When I look at the System event log I can see one enntry with the following warning:
A process serving application pool 'xxxx' exceeded time limits during shut down. The process id was '3868'.
It is preceded by the standard message format as follows (which also occurs every hour as expected with the recycle setting for the application pool):
A worker process with process id of '3868' serving application pool 'xxxx' has requested a recycle because the worker process reached its allowed processing time limit.
It should be noted that the web application is designed to auto-save every 15 minutes so if a user leaves the browser open the session never ends so I am reasonably certain that this is what is causing the worker processes to fail to recycle but just want clarification.
The server is an IBM blade on an IBM blade server chassis. We seem to have some issues with the network cards on the chassis as when we reboot the blades sometimes fail to detect the network cards.
The last time it happned the following appeared in the event log:
Broadcom NetXtreme Gigab: The network link is down. Check to make sure the network cable is properly connected.
and
Network adapter \DEVICE\{B8F84C2E-0773-4106-8D14-2B68EA6ADAA6} is de-activated and is no longer participating in network traffic.
Thanks in advance,
James![[sadeyes]](/data/assets/smilies/sadeyes.gif "[sadeyes] [sadeyes]")
James Culshaw
I have an ASP.NET application that runs on its own dedicated server.
About 4 weeks ago it started to respond to quite a few requests with request timed out errors. It has steadily degraded to the point where it is basically unavailable 90% of the time. As you may guess this has become a real issue as it is a business critical application.
I have ran a SQLProfiler trace and found that the database (which is on a different server) is performing great.
I have ran PerfMon on the web server and have seen some rather alarming (or so my uneducated developer eye tells me
) results. Here is a synopsis:Available MBytes: Avg. 2075 Max. 2092
Page Faults/sec: Avg. 33.712 Max. 3639.389
Page Reads/sec: Avg. 0.208 Max. 126.534
Request Execution Time: Avg. 2144 Max. 165058
Request Wait Time: Avg. 1 Max. 31
Current Connections: Avg. 0 Max. 12
Connection Attempts/sec: Avg. 59652.429 Max. 286331740
Bytes Received/sec: Avg. 2597.381 Max. 246678.701
Bytes Sent/sec: Avg. 52138.766 Max. 12314916
Worker Process Restarts: Avg. 0 Max. 0
I have the App Pool to recycle every 60 minutes (it used to be after 35000 requests but I changed it incase this was the problem)
When I look at the System event log I can see one enntry with the following warning:
A process serving application pool 'xxxx' exceeded time limits during shut down. The process id was '3868'.
It is preceded by the standard message format as follows (which also occurs every hour as expected with the recycle setting for the application pool):
A worker process with process id of '3868' serving application pool 'xxxx' has requested a recycle because the worker process reached its allowed processing time limit.
It should be noted that the web application is designed to auto-save every 15 minutes so if a user leaves the browser open the session never ends so I am reasonably certain that this is what is causing the worker processes to fail to recycle but just want clarification.
The server is an IBM blade on an IBM blade server chassis. We seem to have some issues with the network cards on the chassis as when we reboot the blades sometimes fail to detect the network cards.
The last time it happned the following appeared in the event log:
Broadcom NetXtreme Gigab: The network link is down. Check to make sure the network cable is properly connected.
and
Network adapter \DEVICE\{B8F84C2E-0773-4106-8D14-2B68EA6ADAA6} is de-activated and is no longer participating in network traffic.
Thanks in advance,
James
James Culshaw
