By using this site, you agree to our updated Privacy Policy and our Terms of Use. Manage your Cookies Settings.
440,939 Members | 1,600 Online
Bytes IT Community
+ Ask a Question
Need help? Post your question and get tips & solutions from a community of 440,939 IT Pros & Developers. It's quick & easy.

Fast Response in Visual Studio, slow response under IIS 6.0

P: n/a
I'm pulling my hair out over this one. I wrote a simple web app that
requests a comma delimited list of IP addresses from another system's
web service. It then splits the list into a very big array, iterates
the array and sends a 250k UDP packet to each IP address. For my
testing I'm hitting near a thousand IPs, eventually the app could hit
up to 15 thousand IPs.

On my laptop running Visual Studio's web server I complete my test
broadcast in 3 seconds. When I run the app on my Win 2k3 server with
IIS 6.0, it takes a little over 8 minutes.

The system monitor, network monitor, and task manager all show
absolutely no stress on the server during this test.

I'm not sure what to try. I appreciate any ideas you may give me.

Jul 24 '06 #1
Share this Question
Share on Google+
4 Replies


P: n/a
It could be that you could be forcing a security check. IIS could be set up
incorrectly. You might have the proxy set up correctly on your machine and
not on the server. Etc.

Set up trace statements around the loop and output which iteration and where
you are. You will notice one statement (or a set) that takes a long time.

--
Gregory A. Beamer
MVP; MCP: +I, SE, SD, DBA

*************************************************
Think Outside the Box!
*************************************************
<tj*******@hotmail.comwrote in message
news:11*********************@m79g2000cwm.googlegro ups.com...
I'm pulling my hair out over this one. I wrote a simple web app that
requests a comma delimited list of IP addresses from another system's
web service. It then splits the list into a very big array, iterates
the array and sends a 250k UDP packet to each IP address. For my
testing I'm hitting near a thousand IPs, eventually the app could hit
up to 15 thousand IPs.

On my laptop running Visual Studio's web server I complete my test
broadcast in 3 seconds. When I run the app on my Win 2k3 server with
IIS 6.0, it takes a little over 8 minutes.

The system monitor, network monitor, and task manager all show
absolutely no stress on the server during this test.

I'm not sure what to try. I appreciate any ideas you may give me.

Jul 24 '06 #2

P: n/a
msnews.microsoft.com wrote:
It could be that you could be forcing a security check. IIS could be set up
incorrectly. You might have the proxy set up correctly on your machine and
not on the server. Etc.

Set up trace statements around the loop and output which iteration and where
you are. You will notice one statement (or a set) that takes a long time.

--
Gregory A. Beamer
MVP; MCP: +I, SE, SD, DBA

*************************************************
Think Outside the Box!
*************************************************
IIS is set to allow anonymous at this point. I'll check the proxy when
I get to work tomorrow, the intranet is supposed to bypass it. I'll
see if I've got it configured correctly.

Before I left work, I did output to a log file the date/time of each
iteration of the loop and at various positions in the preceding logic.
The pattern I noticed was 5 second pauses between bursts of UDP packet
sends, around 25-30 per second with a little fluctuation. On my laptop
it was a constant flow of sends without the 5 second interruption.

Jul 25 '06 #3

P: n/a
Before I left work, I did output to a log file the date/time of each
iteration of the loop and at various positions in the preceding logic.
The pattern I noticed was 5 second pauses between bursts of UDP packet
sends, around 25-30 per second with a little fluctuation. On my laptop
it was a constant flow of sends without the 5 second interruption.
Just a thought, maybe it could have something to do with the application
pool used? e.g. IIS pings each workerthread every 30 seconds by default.
Changing that number might help?

HTH,

Bastiaan de Rijber
Jul 25 '06 #4

P: n/a
Just a thought, maybe it could have something to do with the application
pool used? e.g. IIS pings each workerthread every 30 seconds by default.
Changing that number might help?

HTH,

Bastiaan de Rijber
I created a new app pool and turned off the health monitoring
yesterday. That made no difference.

However, we finally fixed it today. My partner found a beta version of
..Net 2.0 installed, that neither him nor I installed. We uninstalled
it and installed the correct version of .Net 2.0. It does all 15
thousand workstations in 8 seconds now.

Thanks to both of you for your responses!

Jul 25 '06 #5

This discussion thread is closed

Replies have been disabled for this discussion.