VoIP Mailing List Archives

[markmorreny at gmail.com]

Hi,
 
Thank you for your help. 
 
I get that too, but the callback does not execute the second time. 
 
When I do task->runtime = switch_time_now() + 10;, what does +10 mean?  Does it mean 10 s or 10 mins?
 
Thanks,
Mark


On Wed, Aug 12, 2009 at 11:09 PM, Mathieu Rene <mrene_lists@avgs.ca (mrene_lists@avgs.ca)> wrote:

[mike at jerris.com]

thats in seconds.

Mike

On Aug 14, 2009, at 8:32 AM, mark morreny wrote:

[markmorreny at gmail.com]

Hi Michael,
 
The following code was executed once, but not after the next 10 s. 
 
SWITCH_STANDARD_SCHED_FUNC(data_flush_callback) {
switch_log_printf(SWITCH_CHANNEL_LOG, SWITCH_LOG_ERROR, "starting to flush data  buffer...\n");
        
    task->runtime = switch_time_now() + 10;
}

Any suggestion why?
 
 
Thanks,
Mark

On Sat, Aug 15, 2009 at 2:13 AM, Michael Jerris <mike@jerris.com (mike@jerris.com)> wrote:

[mike at jerris.com]

task->runtime = switch_epoch_time_now(NULL) + 10;

On Aug 14, 2009, at 10:19 PM, mark morreny wrote:

[mrene_lists at avgs.ca]

Because switch_time_now() is in microseconds.

Mathieu Rene
Avant-Garde Solutions Inc
Office: + 1 (514) 664-1044 x100
Cell: +1 (514) 664-1044 x200
mrene@avgs.ca (mrene@avgs.ca)








On 14-Aug-09, at 11:16 PM, Michael Jerris wrote:

[markmorreny at gmail.com]

Hi,
 
Thank you very much.  I am able to get my sched api to work now.
 
I still have one problem.  I am getting core dump if freeswitch is shut down by typing in "shutdown" in the CLI.  How can I control the shutdown process so that it will wait until all the existing sched jobs are done before starting to free the memory?
 
Mark


On Sat, Aug 15, 2009 at 11:26 AM, Mathieu Rene <mrene_lists@avgs.ca (mrene_lists@avgs.ca)> wrote:

[brian at freeswitch.org]

Sounds like you need to put a mutex on your scheduler routine.

/b

On Aug 18, 2009, at 10:18 AM, mark morreny wrote: