0

currently we have our IIS7 website defualt logging data to a file. Works great. Now, we've got some system monitoring stuff which 'pings' a file in the site. This file is called HttpMonitor.html If the file exists and it's content is '1' then the ping is considered 'successful'. anything else or the file doesn't exist, the ping is considered 'failed'.

after a certain amount of failed attempts for a period of time, warnings go off, etc.

That's working great.

BUT, it pings the server every ... 3 seconds or 10 seconds or whatever. therefore, the log files are getting filled up with lots of crap entries.

Intially, i created a DIFFERENT website (on the same machine) and let the monitor hit that. Now, that worked great .. but that was only now telling me if the server was up. not if the server AND the website is up.

So, is there a way I can get the logging to NOT log any requests that come from.. say .. and ip address or when it tries to hit a certain resource?

Pure.Krome
  • 6,508
  • 18
  • 73
  • 87

2 Answers2

1

The short answer is that IIS allows you to configure exclusions to what gets logged in IIS. The documentation is pretty easy to follow and should meet your requirements.

Check out my article titled Limit what gets logged in IIS 7...

Dscoduc
  • 1,095
  • 2
  • 8
  • 15
  • That's awesome :) I've also been 'burnt' by locked areas and had to use the command line. it sucks how so many area's are not available via the GUI, so 'finding' and learning about them is complex :( – Pure.Krome Sep 11 '09 at 07:47
0

How about creating a new site for the test and not enabling logging for that site?

joeqwerty
  • 109,901
  • 6
  • 81
  • 172
  • That's what i said i did. But that's only good enough for seeing that the SERVER is up. not the actual *live* website and server. a good example of the server being up but the site being down could be a DDOS on the site, the site app pool failing or the site is running really really slow (eg. endless loop, long running query that's chewed up all the connections, etc. etc). – Pure.Krome Sep 11 '09 at 00:03
  • Sorry. I missed that. – joeqwerty Sep 11 '09 at 00:17
  • What about a new site to check the server, say every minute and have it check the "live" site every 10 minutes? How often do you really need to check the site itself? – joeqwerty Sep 11 '09 at 00:19
  • well, not too often .. but we have $$$ SLA's in place so we do need to check it 'regularly'. no more than 30 secs. Sounds extreamly hardcore, but when there's lots of $$$ in place for downtime, we need to be on our toes immediately. even if it's false reports. better safe than sorry. – Pure.Krome Sep 11 '09 at 00:43
  • I've poked around in my IIS 7 server (I just started working with IIS 7) and can't see any way to adjust the log to suit your needs. – joeqwerty Sep 11 '09 at 01:41
  • Glad to see you found a solution. Based on the info in the article it looks like you'll be able to accomplish what you need to. – joeqwerty Sep 11 '09 at 12:13