Posted: Fri Mar 20, 2009 8:05 am Post subject: [asterisk-dev] Bug or not - "no reply to our critical packet
Hello everybody !
Last day we experiment a communication drop with this error message :
[Mar 18 12:35:27] WARNING[3854]: chan_sip.c:1958 retrans_pkt: Maximum retries exceeded on transmission MzhjNTAyZjNiOTM4MGNjMGZhOGJmYmQ2ZjFlMWJhNzA. for seqno 2 (Critical Response) -- See doc/sip-retransmit.txt.
[Mar 18 12:35:27] WARNING[3854]: chan_sip.c:1980 retrans_pkt: Hanging up call MzhjNTAyZjNiOTM4MGNjMGZhOGJmYmQ2ZjFlMWJhNzA. - no reply to our critical packet (see doc/sip-retransmit.txt).
I check on the web and see that long time ago, some user experiment had same problem.
But I see also that, since version 1.4, this kind of error “no reply to our critical packet” wasn’t considered as a Fatal error… So is it normal that cause hang up ?
I’m running Asterisk 1.4.22 and communication was established between an Aastra 55i and X-Light
Posted: Fri Mar 20, 2009 9:06 am Post subject: [asterisk-dev] Bug or not - "no reply to our critical packet
I guess that Asterisk sends 200 OK but does not receive ACK. This is the
usual problem if call drops after 30 seconds.
IMO it is the correct behavior to drop the call if ACK is not received,
becuase usually this is introduced by NAT problems and usually fi ACK is
not sent to Asterisk, other important messages (like BYE ...) will also
not make it through to Asterisk
use ngrep to debug your retransmission issues (sniff also on the SIP
phone side)
regards
klaus
Serge Berney schrieb:
Quote:
Hello everybody !
Last day we experiment a communication drop with this error message :
[Mar 18 12:35:27] WARNING[3854]: chan_sip.c:1958 retrans_pkt: Maximum
retries exceeded on transmission
MzhjNTAyZjNiOTM4MGNjMGZhOGJmYmQ2ZjFlMWJhNzA. for seqno 2 (Critical
Response) -- See doc/sip-retransmit.txt.
[Mar 18 12:35:27] WARNING[3854]: chan_sip.c:1980 retrans_pkt: Hanging up
call MzhjNTAyZjNiOTM4MGNjMGZhOGJmYmQ2ZjFlMWJhNzA. - no reply to our
critical packet (see doc/sip-retransmit.txt).
I check on the web and see that long time ago, some user experiment had
same problem.
But I see also that, since version 1.4, this kind of error “no reply to
our critical packet” wasn’t considered as a Fatal error… So is it normal
that cause hang up ?
I’m running Asterisk 1.4.22 and communication was established between an
Aastra 55i and X-Light
Posted: Fri Mar 20, 2009 9:06 am Post subject: [asterisk-dev] Bug or not - "no reply to our critical packet
First of all check all network nodes (switches, routers,...) and check whether the SIP signalling is OK in terms of NAT handling. Your problem is that an ACK is lost and the session is torn down, this behaviour is OK.
If everything is OK, I would recommend that you use 1.4.18 because the NAT handling is better. I've seen that from 1.4.18 there's a problem but I don't know enough about the * internals to know where it is (i tried 1.4.20,1.4.22, 1.4.23)
The problem I've seen is the following:
How to detect with keep alive mechanism (easily to spot):
SIP endpoints behind a router without any ALG enabled and * in a public IP.
When SIP endpoints sends the keep alive (NOTIFY for example) the Via header is a private one (192.168.1.1) and the source IP is the public interface of the router.
Even though you have nat=yes in those endpoints, * returns the reply to the private address (as stated by the RFCs) but it does not reach the endpoint because the destination address is a private onte, instead of the IP where the packet came from.
I know it's not RFC compliant but when I see this behaviour, some calls are dropped with the "critical response" message because the ACK handling is also not well done.
Hope anyone with enough knowledge can shed some light on this issue.
Last day we experiment a communication drop with this error message :
[Mar 18 12:35:27] WARNING[3854]: chan_sip.c:1958 retrans_pkt: Maximum retries exceeded on transmission MzhjNTAyZjNiOTM4MGNjMGZhOGJmYmQ2ZjFlMWJhNzA. for seqno 2 (Critical Response) -- See doc/sip-retransmit.txt.
[Mar 18 12:35:27] WARNING[3854]: chan_sip.c:1980 retrans_pkt: Hanging up call MzhjNTAyZjNiOTM4MGNjMGZhOGJmYmQ2ZjFlMWJhNzA. - no reply to our critical packet (see doc/sip-retransmit.txt).
I check on the web and see that long time ago, some user experiment had same problem.
But I see also that, since version 1.4, this kind of error “no reply to our critical packet” wasn’t considered as a Fatal error… So is it normal that cause hang up ?
I’m running Asterisk 1.4.22 and communication was established between an Aastra 55i and X-Light
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum