In the past thread termination (in SeeFusion) seems to take about 5 minutes to recover from.

 

I guess I could do a stack trace if someone told me what and how to do it.

 

Robert P. Reil

Managing Director,

Motorcyclecarbs.com, Inc.

4292 Country Garden Walk NW

Kennesaw, Ga. 30152

Office 770-974-8851

Fax 770-974-8852

www.motorcyclecarbs.com


From: Steve Drucker [mailto:[EMAIL PROTECTED]
Sent: Tuesday, August 29, 2006 11:21 AM
To: discussion@acfug.org
Subject: RE: [ACFUG Discuss] RE: JRUN Hangs REDEUX

 

More importantly, what pages does SeeFusion note as running during your "lockups"?

 

If you terminate the execution of those threads (using SeeFusion), is the server able to handle requests normally?

 

Can you provide a stack trace of the threads running at the time of lockup?

 

Regards,
Steve Drucker
CEO
Fig Leaf Software
http://www.figleaf.com
http://training.figleaf.com
Adobe / Macromedia / Google / Paperthin Premier Consulting and Training Partner

 

 


From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Robert Reil
Sent: Tuesday, August 29, 2006 11:00 AM
To: discussion@acfug.org
Subject: RE: [ACFUG Discuss] RE: JRUN Hangs REDEUX

Well using SeeFusion showed me that if I decrease thread volume I increased UP TIME!

What was more important?????????

 

4 hours between 1-2 minute lock ups, or

2 hours between 40 minute lock ups?

 

I think I did the only logical thing I could do considering my pathetic lack of knowledge.

I am sure that one day I will comment to myself how IGNORANT I was at this time.

 

Robert P. Reil

Managing Director,

Motorcyclecarbs.com, Inc.

4292 Country Garden Walk NW

Kennesaw, Ga. 30152

Office 770-974-8851

Fax 770-974-8852

www.motorcyclecarbs.com


From: Teddy Payne [mailto:[EMAIL PROTECTED]
Sent: Tuesday, August 29, 2006 10:41 AM
To: discussion@acfug.org
Subject: Re: [ACFUG Discuss] RE: JRUN Hangs REDEUX

 

Umm, why would you reduce the thread count? Threads are like available channels that can execute code in tandom.  You have a queue of requests and the threads are the ones that handle all of the requests.   If you have 100 requests and 10 threads handle them, that is faster than 100 requests and 2 threads handling them.  The only thing that has to observed is how much memory is best suited for the threads.  Each threads takes up resources as individual and asynchronous processes.   I typically increase my threads over time and watch the performance of a box.  I try to keep the CPU to around 30% if possible.  60-70% works just fine if you have over 1000 queues per minute.

Teddy

-------------------------------------------------------------
To unsubscribe from this list, manage your profile @
http://www.acfug.org?fa=login.edituserform

For more info, see http://www.acfug.org/mailinglists
Archive @ http://www.mail-archive.com/discussion%40acfug.org/
List hosted by FusionLink
-------------------------------------------------------------

Reply via email to