I have a rather disturbing problem with our company's Umbraco setup (We're running Umbraco 4.7.1).
We have some code that runs in our ApplicationStart event (in global.asax) - whenever this code runs it's logged out to a physical file.
The premise for putting the code in the applicationstart event is that it would only run directly after an app pool recycle and such.
However - when our marketing team publishes pure content from within Umbraco - we get a hit in our log - meaning that Application_Start has been called.
I can see in our logs that the app pool has NOT been recycled - so I'm at a loss.
Does any of you guys have any experience with this kind of thing - and could perhaps give me some kind of explanation (or maybe even a fix?)? :)
The problem with Application_Start being triggered is that it rebuilds an index - which requires a lot of resources for a short amount of time (I know, I know... not the ideal place to put this kind of code...) - which isn't all that great when there's a lot of activity on our servers (we're running a 3-server cluster).
Publishing content triggers application_start
Hi guys,
I have a rather disturbing problem with our company's Umbraco setup (We're running Umbraco 4.7.1). We have some code that runs in our ApplicationStart event (in global.asax) - whenever this code runs it's logged out to a physical file. The premise for putting the code in the applicationstart event is that it would only run directly after an app pool recycle and such.
However - when our marketing team publishes pure content from within Umbraco - we get a hit in our log - meaning that Application_Start has been called. I can see in our logs that the app pool has NOT been recycled - so I'm at a loss.
Does any of you guys have any experience with this kind of thing - and could perhaps give me some kind of explanation (or maybe even a fix?)? :)
The problem with Application_Start being triggered is that it rebuilds an index - which requires a lot of resources for a short amount of time (I know, I know... not the ideal place to put this kind of code...) - which isn't all that great when there's a lot of activity on our servers (we're running a 3-server cluster).
Thanks :)
is working on a reply...