[Varnish] #501: Stats variable to monitor how many threads are actually doing something (n_wrk_busy): busy threads monitoring

[Varnish]

#501: Stats variable to monitor how many threads are actually doing something
(n_wrk_busy): busy threads monitoring
-------------------------+--------------------------------------------------
 Reporter:  stockrt      |       Owner:  phk                    
     Type:  enhancement  |      Status:  new                    
 Priority:  normal       |   Milestone:  Varnish 3.0 dev        
Component:  varnishd     |     Version:  trunk                  
 Severity:  normal       |    Keywords:  busy threads monitoring
-------------------------+--------------------------------------------------
Changes (by kristian):

  * version:  2.0 => trunk
  * milestone:  => Varnish 3.0 dev


Old description:

> Hi!
>

> I made this patch for implementing a new varnishstat variable, called
> 'n_wrk_busy'.
>

> With this new variable we can now know for sure how many of our active
> threads are actually doing something, so the system can be better
> dimensioned in order to know if we are getting near the max threads
> utilization: if we are, we can increase the number of max threads in
> order to instatiate more, if the system can handle more threads, if not,
> go for more hardware.
>

> This is good in the cases when we configure thread_pool_min ==
> thread_pool_max and do not have to know how many of the threads are
> actually used for something.
> I believe the same applies when we have thread_pool_min == 1 and
> thread_pool_max == MAX_WISHED. In this case we can know how many threads
> are idle (waiting for thread_pool_timeout to expire) and how many are
> actually working.
>

> This is good for monitoring and I hope this can help someone else to know
> what is happening under the hood a little better.
>

> Best regards,
>
> Rogério Schneider

New description:

 Hi!


 I made this patch for implementing a new varnishstat variable, called
 'n_wrk_busy'.


 With this new variable we can now know for sure how many of our active
 threads are actually doing something, so the system can be better
 dimensioned in order to know if we are getting near the max threads
 utilization: if we are, we can increase the number of max threads in order
 to instatiate more, if the system can handle more threads, if not, go for
 more hardware.


 This is good in the cases when we configure thread_pool_min ==
 thread_pool_max and do not have to know how many of the threads are
 actually used for something.
 I believe the same applies when we have thread_pool_min == 1 and
 thread_pool_max == MAX_WISHED. In this case we can know how many threads
 are idle (waiting for thread_pool_timeout to expire) and how many are
 actually working.


 This is good for monitoring and I hope this can help someone else to know
 what is happening under the hood a little better.


 Best regards,

 Rogério Schneider

--

Comment:

 This ties in with proposed scoreboard-functionality, but is not going to
 make it for 3.0.

 Sorry for the horrendous response-time.... Ugh.

-- 
Ticket URL: <http://www.varnish-cache.org/trac/ticket/501#comment:5>
Varnish <http://varnish-cache.org/>
The Varnish HTTP Accelerator