Charles, i had the same error on a fairly busy site as well, although at the time of the error the visits where not that much (around 500 page views per hour).

This is the stacktrace of the error, maybe it tells you something. I would agree that it is a caching issue, hand from the differing errors, it seems to happen in diferent locations. Could we add defensive coding in tab controller to check for this error occuring, and if it does clear the cache (or maybe even do an app restart, as that seems to help)

Stack trace: at System.Collections.Generic.Dictionary`2.Insert(TKey key, TValue value, Boolean add)

Add me to the list of 'infected'.

Our environment is a 8-CPU 4Gb box (64 bit Enterprise w/ SQL 2005). 

I have finally had to add application CPU restarts to the the affected app otherwise it kills the entire box. I've tried various host cache 'settings' all to no avail.

Although it's not a 'cure' but a bandaid, it seems that Pageblaster helps... since many of our audience is 'non-registered' users. 

However the situation is pretty bad, it can peg all 8 CPU's in about 15 seconds.. so I have the app set to recycle if the CPU is > 70% for 1 min. and so it's recycling about every 10-15 mintues. Not good for user usage, since their connections keep getting reset.

This is an upgraded instance from a way's back, however, this issue seemed to have started in the 4.8.1 upgrade.

Any info would be appreciates, and yes I tried the module SQL statement for SupportedFeatures.. that didn't help.

Any suggestions or patch's would be appreciated.

Like the recently edited reply above, I also tried to set the host cache to 'no cache'.  That certainly didn't have the same effect for this environment, as it rapidly pegged the CPU's, I did turn off compression.. I'll be watching and editting one variable at a time to see if I can isolate it to any 'host setttings'. What's odd is the Pages/Sec... I've looked at other 'recommended' values they say about 2-3. Mine is consistently around 25-30 at 10 minute snapshots.

Cheers,

Thought I'd give an update on my experience with this issue.

I have tried the no-cache route, that seems to always just peg the SQL cpu's to 100%.

However, after noticing some event log issues (and the fact SQL always times out trying to get into it ), I truncated the EventLog table, made sure that all the event log 'max' records we're reasonable, and restarted the app.

Within seconds the event log General Exception buffers we're filled (500 entries), so it looks like this is a constant issue. However, i am getting a different but similar error now.

DNN 4.8.0

AssemblyVersion: -1
PortalID: -1
PortalName:
UserID: -1
UserName:
ActiveTabID: -1
ActiveTabName:
RawURL:
AbsoluteURL:
AbsoluteURLReferrer:
UserAgent:
DefaultDataProvider:
ExceptionGUID:
InnerException: An item with the same key has already been added.
FileName:
FileLineNumber: -1
FileColumnNumber: -1
Method:
StackTrace:
Message: System.ArgumentException: An item with the same key has already been added. at System.ThrowHelper.ThrowArgumentException(ExceptionResource resource) at System.Collections.Generic.Dictionary`2.Insert(TKey key, TValue value, Boolean add) at DotNetNuke.Services.FileSystem.FolderController.FillFolderInfoDictionary(IDataReader dr)
Source:

This happens constantly, and it keeps the SQL and App pools churning.. as the limit is always hit, and it's constanlty being deleted and refreshed with new errors.

This is on a production server with fairly light traffic (perhaps 1-2k visitors a day), it's becoming quite a chore to not know how to alieviate this issue, and have the event log run like a ticker tape...

Any help would be appreciated