Timeout Expired from Application

Question

Timeout Expired from Application

sswords

SSCrazy Eights

Points: 8207
More actions
December 22, 2004 at 9:32 am

#84398

Greetings All,
We just moved a production DB to Windows/SQL Server 2000 and upgraded the application version in the same weekend. These are spanking new dual 3 Gig processor servers with 4 gigs RAM on the SQL server and 2 on the Metaframe server. Both have latest service packs and security/hot fixes. After moving to production, users began complaining of numerous Timeout Expired errors that seem to be happening randomly thoughout the app from the Citrix box. Sometimes they get kicked out, sometimes they just click again and they're on their way. I've been able to reproduce it once or twice on a fat client so I don't think it's Metaframe.
An example of one of the errors is as follows:
++VB Error (-2147467259) trapped by proc: Microsoft OLE DB Provider for SQL Server - clsDataAccessor:HCDataAccessorInit - TranObj::ReadListData - UsrCurPgmModLstAll::GetKeys - UsrCurPgm::LoadUsrCurPgmMod - UsrCurPgm::UsrCurPgmModCount - UsrCurPgm::AddUsrCurPgmMod: ConcurrencyMonitor_Add() Desc : Timeout expired 12/17/2004 12:04:31 PM
I've checked all the obvious stuff. NIC is fine, configured properly. Ran pingplotter for several hours from both servers and the response time was never greater than 16 ms so it's not any network latency. I've also checked SQL client configuration for any alias references, made sure TCP/IP enabled and preferred. Checked about everything else I can think of and we're still seeing the errors numerous times daily. I'm no developer but know enough to suspect the API. Vendor keeps saying it's our environment and they can't reproduce it. If anyone out there can think of anything else for me to check, I'd appreciate it. Thanks and Happy Holidays to all.

My hovercraft is full of eels.

Viewing 7 posts - 1 through 6 (of 6 total)

You must be logged in to reply to this topic. Login to reply

dtevlin Ten Centuries Points: 1084 More actions · Answer 1

We had slightly similar problems with Citrix timing out or booting folks off the application, or just running slowly. We traced it back to XP SP2 and the changes to the TCP/IP that it included. Microsoft has a hotfix for the problem. Don't know if you already considered that, but thought that I should mention it

sswords SSCrazy Eights Points: 8207 More actions · Answer 2

The clients aren't running XP SP2 yet but the fix would worth taking a look at. Do you have a KB number for this? Thanks a lot.

My hovercraft is full of eels.

dtevlin Ten Centuries Points: 1084 More actions · Answer 3

Sure the KB is KB884020 or just go here.

http://www.microsoft.com/downloads/details.aspx?FamilyID=17D997D2-5034-4BBB-B74D-AD8430A1F7C8&displaylang=en

Hope it helps.

sswords SSCrazy Eights Points: 8207 More actions · Answer 4

Thanks but that doesn't look to be our issue. The clients are not running SP2 yet and we're resolving by server name not IP address. Per one of the vendor's reccommendations, we attempted to have the app resolve by the IP address by modifying the ini file and this did seem to actually make the problem worse so there may be a connection somewhere. I am having little success finding anything much that addresses 'Timeout Expired' messages.

I took a look at another log file that's being generated by the API and am also seeing a lot of this error:

: ++VB Error (-2147024894) trapped by proc: ErrorHandler: clsHHCClient_Remove() Desc : Automation error

The system cannot find the file specified. 12/22/2004 2:48:58 PM

Now this would make me think that there's some kind of delay or problem with the network or possibly with user rights, but I can't find anything that looks abnormal. If it were a question of rights, it wouldn't work at all but instead we're seeing this happen randomly. Sometimes it throws an error, then you can click on the same icon again from the same session and it goes right in. Really wierd. We're on a fully switched network and these boxes are in the same rack. I'm running out of ideas so any insight is appreciated.

My hovercraft is full of eels.

sswords SSCrazy Eights Points: 8207 More actions · Answer 5

Just by means of an update (and to bump my original thread )we're still seeing this daily and users are nearly ready to boil us in oil. The most frustrating part is they're on brand new servers with lots of room to grow and they should be tooling along nicely. In fact, performance monitor shows the CPU utilization stays in the single digits for most of the day. I've run numerous traces basically just filtering on errors and warnings and that's not helping much. Exceptions are being thrown but if I'm translating correctly. they're to be expected.

I know the 'Timeout Expired' is a fairly generic error message and you don't find out a lot just by searching on that. You find out a little more by searching on the (-2147467259) VB error number but it's still pretty generic. This vendor has other clients running the app and has seen the issue before but it's not well documented as to how (or if) it was resolved.

About the only thing left that I can think of is possibly processor Hyperthreading is causing a problem and if anyone can substantiate that, or give me another direction, I'd appreciate it. TIA.

My hovercraft is full of eels.

dtevlin Ten Centuries Points: 1084 More actions · Answer 6

You might want to check out this KB article on scaleability and SQL Server 2k when AWE is enabled. Don't know if this is the fix your looking for but worth a look.

http://support.microsoft.com/kb/332004