I don't think this would be a problem for this particular case, However with
older versions of Witango (I think 5.5.006 or so) setting DSLife to 0 under
heavy loads would cause Witango to fail. I have not retested with later
versions of Witango but be aware that setting DSLife to 0 can cause
different problems in my experience.

 
Regards,
Christian
 

-----Original Message-----
From: Robert Garcia [mailto:[EMAIL PROTECTED] 
Sent: Tuesday, January 23, 2007 12:24 PM
To: [email protected]
Subject: Re: Witango-Talk: Reliability of ODBC connections?

Being that this is across the country, I would see if you could set  
DSLife for these connections to 0, it would probably help with  
reliability. Some ODBC drivers can't recognize when a connection is  
dropped for internet connectivity reasons, and so when witango may go  
to reuse the connection, it will cause an error. If dslife is set to  
0, it will initiate a new connection on each hit, which probably  
doesn't help the latency, but will probably be more reliable than  
expecting a connection accross the country to be up for 30-60 mins  
without interruption.

I am not sure if you can set dslife, by connection with scoping, cuz  
I have never had a need, but just a thought, since your original  
problem mention disconnects.

-- 

Robert Garcia
President - BigHead Technology
VP Application Development - eventpix.com
13653 West Park Dr
Magalia, Ca 95954
ph: 530.645.4040 x222 fax: 530.645.4040
[EMAIL PROTECTED] - [EMAIL PROTECTED]
http://bighead.net/ - http://eventpix.com/

On Jan 23, 2007, at 9:12 AM, Kent Swisher wrote:

> In my case, I have very light load.  Rarely more than 2 concurrent  
> users.  Not experiencing hung threads.  Instead, the ODBC  
> connection is being dropped and Witango query times out.  Next hit,  
> Witango re-establishes connection and off we go again until the  
> next time.  Next time could be hours or days.
>
> In several instances, the first thread had not timed out yet (query  
> timeout=5 min), user refreshed at 120 seconds, Witango starts new  
> thread, recognizes connection was dropped, re-establish connection,  
> first thread completes returning data, then second thread returns  
> data.
>
> -> Kent
>
> On 1/23/2007 6:20 AM, Christian Carrier wrote:
>> I agree with Robert. We have quite a few intermittent failures  
>> especially
>> with Windows 2003 Server SP1+. It seems that threads get hung in  
>> Witango and
>> the only way to clear them is by recycling the Application. We see  
>> this with
>> MSSQL. Any long running queries / stored procedures seem to have  
>> the ability
>> to hang a Witango thread.
>>  -----Original Message-----
>> From: Robert Garcia [mailto:[EMAIL PROTECTED] Sent: Tuesday,  
>> January 23, 2007 2:09 AM
>> To: [email protected]
>> Subject: Re: Witango-Talk: Reliability of ODBC connections?
>> ALSO, when witango is is under load with slow connections, it can   
>> hang, and crash, it is a witango killer also. I have shown this  
>> over  much testing with several DB vendors.
> ______________________________________________________________________ 
> __
> TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf
>
________________________________________________________________________
TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf
________________________________________________________________________
TO UNSUBSCRIBE: Go to http://www.witango.com/developer/maillist.taf

Reply via email to