Hello,
I am using the PDB module and server components for number portability. 2 instances of PDB Server runs on (10.12.19.51/10001/10002), Kamailio on (10.12.19.21). With a small amount of traffic (-cmax 150 -cps 10 -callduration 3), where are timeouts: WARNING: pdb [pdb.c:260]: exceeded timeout while waiting for response.
One requested number was 307111094, where the module prints out a timeout.
The funny part is, that I can see the responses at least arriving at the 10.12.19.21 interface on time.
Request send: 0,200855 s Answer received: 0,201027 s
That are 0,172 ms and far away from a timeout.
What could be the reason ?
regards,
Thomas
ps. A small change on the server part was done: handle 4 digit carrier codes.
level3_logs:
Feb 9 17:19:42 node2 /openser/sbin/kamailio[21200]: DEBUG: pdb [pdb.c:201]: querying '307111094'... Feb 9 17:19:42 node2 /openser/sbin/kamailio[21200]: WARNING: pdb [pdb.c:260]: exceeded timeout while waiting for response.
___________________________________________________________ Schon gehört? WEB.DE hat einen genialen Phishing-Filter in die Toolbar eingebaut! http://produkte.web.de/go/toolbar
On Wednesday 09 February 2011, Thomas Baumann wrote:
I am using the PDB module and server components for number portability. 2 instances of PDB Server runs on (10.12.19.51/10001/10002), Kamailio on (10.12.19.21). With a small amount of traffic (-cmax 150 -cps 10 -callduration 3), where are timeouts: WARNING: pdb [pdb.c:260]: exceeded timeout while waiting for response.
One requested number was 307111094, where the module prints out a timeout.
The funny part is, that I can see the responses at least arriving at the 10.12.19.21 interface on time.
Request send: 0,200855 s Answer received: 0,201027 s
That are 0,172 ms and far away from a timeout.
What could be the reason ?
Hi Thomas,
a bit difficult to say on a first sight.. Maybe some scheduling or load issues on the local machine? I've just checked two production server here, i don't see it so far in the available logs.
How much load do you've on the machine? And what kind of timeout value do you specified?
ps. A small change on the server part was done: handle 4 digit carrier codes.
Ah, i see. I guess its not a big change?
Regards,
Henning
Hello Henning,
the load I have on one machine is about 10 cps max. The timeout is configured to the standard value of 50ms. I the logs I can see that the response times are not continuous.
Feb 11 17:28:34 m-lab-prx-as-2 /openser/sbin/kamailio[7609]: INFO: pdb [pdb.c:290]: got an answer in 35.950000 ms Feb 11 17:28:34 m-lab-prx-as-2 /openser/sbin/kamailio[7607]: INFO: pdb [pdb.c:290]: got an answer in 0.201000 ms Feb 11 17:28:38 m-lab-prx-as-2 /openser/sbin/kamailio[7609]: INFO: pdb [pdb.c:290]: got an answer in 0.626000 ms Feb 11 17:28:38 m-lab-prx-as-2 /openser/sbin/kamailio[7605]: INFO: pdb [pdb.c:290]: got an answer in 41.026000 ms Feb 11 17:28:38 m-lab-prx-as-2 /openser/sbin/kamailio[7607]: INFO: pdb [pdb.c:290]: got an answer in 29.807000 ms Feb 11 17:28:38 m-lab-prx-as-2 /openser/sbin/kamailio[7609]: INFO: pdb [pdb.c:290]: got an answer in 0.320000 ms Feb 11 17:28:38 m-lab-prx-as-2 /openser/sbin/kamailio[7607]: INFO: pdb [pdb.c:290]: got an answer in 0.173000 ms
I will do a fallback to the original source and check it again. Maybe there is something from with my modifications.
thanks,
Thomas
-----Ursprüngliche Nachricht----- Von: "Henning Westerholt" henning.westerholt@1und1.de Gesendet: 11.02.2011 09:33:05 An: sr-users@lists.sip-router.org Betreff: Re: [SR-Users] pdb module timeouts
On Wednesday 09 February 2011, Thomas Baumann wrote:
I am using the PDB module and server components for number portability. 2 instances of PDB Server runs on (10.12.19.51/10001/10002), Kamailio on (10.12.19.21). With a small amount of traffic (-cmax 150 -cps 10 -callduration 3), where are timeouts: WARNING: pdb [pdb.c:260]: exceeded timeout while waiting for response.
One requested number was 307111094, where the module prints out a timeout.
The funny part is, that I can see the responses at least arriving at the 10.12.19.21 interface on time.
Request send: 0,200855 s Answer received: 0,201027 s
That are 0,172 ms and far away from a timeout.
What could be the reason ?
Hi Thomas,
a bit difficult to say on a first sight.. Maybe some scheduling or load issues on the local machine? I've just checked two production server here, i don't see it so far in the available logs.
How much load do you've on the machine? And what kind of timeout value do you specified?
ps. A small change on the server part was done: handle 4 digit carrier codes.
Ah, i see. I guess its not a big change?
Regards,
Henning
___________________________________________________________ NEU: FreePhone - kostenlos mobil telefonieren und surfen! Jetzt informieren: http://produkte.web.de/go/webdefreephone
On Friday 11 February 2011, Thomas Baumann wrote:
the load I have on one machine is about 10 cps max. The timeout is configured to the standard value of 50ms. I the logs I can see that the response times are not continuous. Feb 11 17:28:34 m-lab-prx-as-2 /openser/sbin/kamailio[7609]: INFO: pdb [pdb.c:290]: got an answer in 35.950000 ms Feb 11 17:28:34 m-lab-prx-as-2 /openser/sbin/kamailio[7607]: INFO: pdb [pdb.c:290]: got an answer in 0.201000 ms
Hi Thomas,
this is not much load, and a long timeout. Maybe its something special about the used scheduler, or some other process that causes delays? The response times looks indeed somehow strange.
Cheers,
Henning
Henning Westerholt writes:
ps. A small change on the server part was done: handle 4 digit carrier codes.
Ah, i see. I guess its not a big change?
i have done the same change. perhaps it would make sense to make the module a bit more flexible in terms of length of carrier code.
-- juha
-
Henning Westerholt -
Juha Heinanen -
Thomas Baumann