[OpenSIPS-Users] fr_timer not working properly

[Muhammad Shahzad Shafi]

This is pretty strange, since it never happened to me on an AMD
machine, though i have different model, AMD Athlon II X2 250e. 

Can you
try using fr_timer_avp? e.g. 

modparam("tm", "fr_timer_avp",
"$avp(10)") 
modparam("tm", "fr_inv_timer_avp", "$avp(20)") 

See more
info,


http://www.opensips.org/html/docs/modules/1.10.x/tm.html#id293753 [3]


Set these vars and print xlog when INVITE comes in. 

branch_route {

... 

 $avp(10) = 5; 
 $avp(20) = 60; 
 xlog("L_NOTICE",
"[$pr:$fU@$si:$sp]: Setting TRY timer to '$avp(10)' and RING timer to
'$avp(20)' for call from '$fu' to '$ru' n"); 
... 
} 

Then check in
failure route, 

failure_route { 
... 

 if (t_check_status("408")) { 

if ( t_local_replied("all") ) { 
 # local timeout with no reply received
-> fr_timer 
 xlog("L_NOTICE", "[$pr:$fU@$si:$sp]: Timeout on TRY timer
for '$rm' from '$fu' to '$ru' after '$avp(10)' seconds n"); 
 } else if
( t_local_replied("last") ) { 
 # timeout with replies received ->
fr_inv_timer 
 xlog("L_NOTICE", "[$pr:$fU@$si:$sp]: Timeout on RING
timer for '$rm' from '$fu' to '$ru' after '$avp(20)' seconds n"); 
 }
else { 
 # received timeout 
 xlog("L_NOTICE", "[$pr:$fU@$si:$sp]:
Timeout on T2 timer for '$rm' from '$fu' to '$ru' after '$avp(20)'
seconds n"); 
 }; 
 }; 
... 
} 

This will make things clearer if there
is any problem. 

Thank you. 

On 2013-10-25 09:35, mayamatakeshi wrote:


> On Tue, Oct 15, 2013 at 10:09 AM, mayamatakeshi
<mayamatakeshi at gmail.com [2]> wrote:
> 
>> Hello,
>> I have some servers
running opensips 1.10 commit bfd86ab25554082053167a82655e669ca1c0ea7c
>>

>> I am seeing an issue in them that i cannot reproduce in other
servers: the fr_timer is not respected.
>> I set tm fr_timer=5 and use
sipp to make a call and make opensips to send the call to an address
that is not being listened by any app like 127.0.0.1:5010 [1]. In these
servers, sometimes timeout happens after 6, 9, 12 seconds but never (so
far after) 5 seconds. But in other machines, fr_timer is consistently
respected. So I am trying to figure out what would be causing this. 
>>
Does anyone have any idea of configuration of the server that could
cause this? 
>> The only significant difference I see is that on the
problematic servers corosync was installed for HA (to control opensips
switchover) but I don't see how this could have such impact.
> 
> Just
in case someone else gets into this. 
> We verified the problem only
happens if the VM runs on a VM host using this processor:
> 
> model
name : AMD Opteron(tm) Processor 6344 
> 
> On the server where we have
this one: 
> 
> model name : Intel(R) Xeon(R) CPU X5550 @ 2.67GHz there
is no problem.
> 
> Regards,
> Takeshi

-- 
Mit freundlichen
Grüßen
Muhammad Shahzad
-----------------------------------
CISCO Rich
Media Communication Specialist (CRMCS)
CISCO Certified Network Associate
(CCNA)
Cell: +49 176 99 83 10 85
MSN: shari_786pk at hotmail.com
Email:
shaheryarkh at googlemail.com
 

Links:
------
[1]
http://127.0.0.1:5010
[2] mailto:mayamatakeshi at gmail.com
[3]
http://www.opensips.org/html/docs/modules/1.10.x/tm.html#id293753
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opensips.org/pipermail/users/attachments/20131025/12f39fc5/attachment.htm>