ok guys - the calendar completely breaks and dern near shuts down the entire dotnetnuke portal application (and even the server if allowed to progress long enough and the server isn't configured properly secured against this)...

When logging in as the administrator, it works fine.  I presume this is because it is not loading from the cache and is instead generating the calendar control from scratch each time...

I can't find any GOOD errors, other than an out of memory error (Cause yep the process hits about 800mb if I let it run after this starts happening)... 

this is the closest thing I can find in the logs to a decent error...

ModuleId: 377
ModuleDefId: 96
FriendlyName: Upcoming Events
ModuleControlSource: DesktopModules/Events/Events.ascx
AssemblyVersion: 04.00.03
Method: System.String.Concat
FileName:
FileLineNumber: 0
FileColumnNumber: 0
PortalID: 0
PortalName: South Carolina State Constables Association
UserID: -1
UserName:
ActiveTabID: 59
ActiveTabName: Home
AbsoluteURL: /Default.aspx
AbsoluteURLReferrer:
ExceptionGUID: c0d036bf-ee5a-4f85-89f9-c7e92bf7b6ce
DefaultDataProvider: DotNetNuke.Data.SqlDataProvider, DotNetNuke.SqlDataProvider
InnerException: Exception of type 'System.OutOfMemoryException' was thrown.
Message: DotNetNuke.Services.Exceptions.ModuleLoadException: Exception of type 'System.OutOfMemoryException' was thrown. ---> System.OutOfMemoryException: Exception of type 'System.OutOfMemoryException' was thrown. at System.String.Concat(String str0, String str1, String str2) at DotNetNuke.Data.SqlDataProvider.GetModuleSettings(Int32 ModuleId) at DotNetNuke.Entities.Portals.PortalSettings.GetModuleSettings(Int32 ModuleId) at DotNetNuke.Modules.Events.EventMonth.BindDataGrid() at DotNetNuke.Modules.Events.EventMonth.Page_Load(Object sender, EventArgs e) --- End of inner exception stack trace ---
StackTrace:
Source:
Server Name: SCTWMHWEB01

We get the same error without using the Events module about every 30 days or so.  We have since simply started rebooting the server every 3 weeks.  We have been unable to pinpoint a specific module or even if DNN is actually causing the problem.  The server has 4Gb ram, RAID5, DNN 3.2.2, Veritas Backup Exec (which is a pig), SSL/PKI enabled, logging and anti-virus software.

Pick any of the above.  They are all contributing to eating ram and not throwing it back up when they are done.  Performance is great until its not.

mikez

Ok to answer the questions asked...

The two "list mode" ones are pointed at two calendars that have maybe 10-20 events TOTAL in the next 365 days...

Further, I can _PROMISE_ its directly related to the events module and is reproducable.  I turn caching on for any instance of the calendar module, logout, try to hit that page and BOOM - w3wp.exe goes crazy - 30-100%cpu usage, and starts eating up RAM.  I watched it get up to 800mb before I finally killed the process.

I can sit and watch at the taskmgr and w3wp.exe... if i let it sit long enough, I'll get a timeout error... if I check the log, I'll get a outofmemory exception, a timeout exception, hell I even saw a date format error...

This isn't a question of "bad performance with lots of events."  this is a matter of "control completely breaks when under cache mode."

I didn't notice it until I logged out of admin, logged in as a test user, and tried to access one of the calendars...

Chavous