There appears to be a potentially serious breaking issue when the dnn scheduler is set to REQUEST_MODE.
Basically on application pool startup it is possible for the system to attempt to start multiple schedulerprovider threads.

Of particular issue is that this seems to be a very possible occurrence on systems that dont have and keepalive system - and are targeted by systems like search crawlers which may fire off a number of page requests to the server at the some time - while the application pool is still trying to get itself started up.

This thread talks about the issue and a possible code patch to rectify it.

http://www.dnnsoftware.com/forums/forumid/216/postid/497703/scope/posts#497703

Would be interested in feedback from the core developers on the ramifications of this.

Is it advisable to switch the scheduler to TIMER_MODE in the interim to avoid the issue.

Westa

PS - have raised a ticket for this in support

https://dnntracker.atlassian.net/browse/DNN-4923

While TIMER mode is default, a lot of people use Request mode as TIMER mode does not work well on sites that are not very busy.

Wes - the Platform team at DNN Corp are planning some scheduler enhancements and fixes in 7.3.0 - so we will make sure this is resolved - I know you logged a JIRA ticket already and we will make sure that it gets fixed when we do that work.

---

Charles Nurse Chief Architect Evoq Content Team Lead, DNN Corp. Want to contribute to the Platform project? - See here

MVP (ASP.NET) and ASPInsiders Member