Looking at your site - I'm not convinced that it has much to do with indexes being fragmented.  Yes, it could to a certain extent as they can cause some impact, as can a large event log or site log.  But this often has more to do with memory management and paging overflows or application restarts than just indexes that need reorganizing.

The most basic call you can make to a dnn website is to KeepAlive.aspx as in http://www.northstarcamaroclub.com/ke... 

Two back to back calls to this page resulted in an all most 1 minute turn around.
2/20/2011 6:55:10 PM
2/20/2011 6:56:06 PM

The first place, I would be looking is for any indications that your application is being restarted by your application pool being flushed or recycled.
Check your site event log for evidence of restarts - and also talk to your host.

A page turn around on keepalive.aspx should be almost instant on back to back calls - since it does NOT access the full dnn page framework  as such -  yes it does cause all the http handlers to fire - but a lot of other stuff is ignored since it is not needing to render a dnn page context or load skins or render the menu or containers or modules - the whole idea of keepalive.aspx is to be a minimal processing call - that only tickles the application to start up in its lowest state.

The first time you call it - yes it will take a little extra time as the application pool loads and the core dnn code is compiled and the dnn context initialized -  the second call to it however should be all but instant since everything in the core should be compiled and the dnn context in cache.

The fact that the second back to back call is taking up to a minute would suggest that for some reason the application pool is being shut down or recycled.

The most common reason for this is that the memory allocation assigned to the application pool is lower than the amount of memory needed to run dnn.

There are steps that can be taken - see the previous post re cleaning site event logs etc - since these can consume memory resources.
Also a very large transaction log in ms sql can also contribute in some ways - but thats often more about memory management at a server level that the application poll < if the server has iis and ms sql on the same box - and not enough memory then the server can go into a memory paging cycle where the application pool and the ms sql server effectively start fighting each other for memory - but as a rule I would not expect that to directly impact so severely on back to back calls to keepalive.

Just me 2 cents

Westa

I always am deleting these: Date Log Type Username Portal Summary 2/22/2011 10:34:52 AM Scheduler Started ... Server Name: WEBD19 2/22/2011 10:34:52 AM Application Started ... Server Name: WEBD19 2/22/2011 10:34:52 AM Event Schedule Started EVENT: APPLICATION_START; ... EVENT: APPLICATION_START Server Name: WEBD19 2/22/2011 10:34:26 AM Application Shutting Down Shutdown Details: The AppDomain shut down because of the hosting env ... Shutdown Details: The AppDomain shut down because of the hosting environment. Server Name: WEBD19 2/22/2011 10:34:25 AM Scheduler Stopped ... Server Name: WEBD19 2/22/2011 10:34:17 AM Scheduler Shutting Down Initiator: Stopped by Application_End; ... Initiator: Stopped by Application_End Server Name: WEBD19 2/22/2011 10:33:35 AM Application Started ... Server Name: WEBD19 2/22/2011 10:33:35 AM Scheduler Started ... Server Name: WEBD19 2/22/2011 10:33:35 AM Event Schedule Started EVENT: APPLICATION_START; ... EVENT: APPLICATION_START Server Name: WEBD19 2/22/2011 10:33:26 AM Application Shutting Down Shutdown Details: The AppDomain shut down because of the hosting env ... Shutdown Details: The AppDomain shut down because of the hosting environment. Server Name: WEBD19 Page 1 of 22 Is this what you mean about the application pool? If so, our site has always had this problem, so I had no idea it was a problem.

---

Visit our Website at www.northstarcamaroclub.com

Wow - are you serious  - your application pool is being shut down and restated every minute according to these logs.

2/22/2011 10:34:26 AM Application Shutting Down Shutdown Details:
2/22/2011 10:33:26 AM Application Shutting Down Shutdown Details:

That basically means EVERY time anyone requests a page on your site - the web server has to completely restart you site's application pool.
This would be giving your site about the worst possible performance ever imagined.

The way asp.net is designed - is that a whole lot of extra work is done when the application pool is started - which among other things dynamically compiles much of your website and validates components, dnn also loads core management elements like host and site settings, permissions and navigation elements.  All this information once loaded is cached in memory and all by instantly available to anyone visiting your site while the application pool is alive.

Given the amount of work that happens during this initial application pool startup it is not unusual for the first page request to take a little time.  Its a fact of life for all asp.net systems ...  and the reason why most asp.net sites use a keepalive site monitoring service too ensure that their application pool is kept active.

OK, what this is saying then is that firstly - for some reason the web server is shutting down your application pool.
This can be happening for one of a number or reasons.

1. the host has set the application pool IDLE TIMEOUT to a stupidly low value.
2. the host has set the amount of MEMORY allocated to your application pool to a value less than the application needs
3. the host has set the cpu load limit to a value less than your application needs to complete the load process
4. the host has set the startup time limit to a value less than your application needs to complete the load process. 
5. the host has set the recycle time out to a stupidly low vale.

If this has ALWAYS been happening then your host needs an aggressive kick in the butt - since this would be causing serious performance issues for anyone using that web server - with your site putting huge resource loads on its performance.

Your host should have = in their system event logs = a more detailed log of what exactly is causing the application pool to recycle. And I would be asking them immediately to provide you with the reasons why it is happening.

But to be honest - any host that allowed this sort of thing to occur is NOT a host that I would be wanting managing my web hosting needs.


Westa