I'm going to retract what I said about IAX2 also failing - i don't think it does.
I think this bug only affects SIP. After the SIP channel has frozen in this way, I can still make calls with IAX2. Interestingly, I can still dial my SIP extension and my SIP client rings (despite not having been able to register). However trying to answer does not work, the SIP client just says "Status: Awaiting Acknowledgement", so I think the SIP client has received the call message but OpenPBX is unable to recieve the response from the client. I've added this update to the bug I've created for this on trac, and edited the title, at: http://trac.openpbx.org/cgi-bin/trac.cgi/ticket/245 Jesse On 23/01/2007, at 8:00 AM, Jesse Reynolds wrote: > OK, you're right Marc because it's still happening! I now have IPv6 > disabled on the en0 interface, and I've allocated a static IP for the > machine now. I've also moved SIP to 5071 so it doesn't conflict with > other SIP equipment behind my NAT router. > > I've found that IAX2 is also affected, as I can use an IAX2 softphone > before the freeze up and not afterwards. > > It seems that the CLI is using a UNIX socket to connect to the > running OpenPBX, is that correct? So really it could be that all > network communication from OpenPBX is failing after this freeze up. > > Here are the only network file descriptors open in OpenPBX while in > this network-frozen state: > > openpbx 197 openpbx 9u IPv4 0x1f489c0 0t0 UDP *:4569 > openpbx 197 openpbx 10u IPv4 0x1f48d00 0t0 UDP *:2727 > openpbx 197 openpbx 11u IPv4 0x1f48ea0 0t0 UDP *:5071 > openpbx 197 openpbx 12u IPv4 0x1f48b60 0t0 UDP *:4520 > > I've tried turning up debug level with the CLI but nothing is > generated in the logs. > > doing a "reload" also does not fix the issue. The only thing that > fixes it is stopping and starting OpenPBX. > > I am keen to carry out more debugging but I'm not sure what to do > next. Any suggestions? ... Perhaps I should trace the system calls of > the running binary and try and see what it's doing when it fails? ... > Is there any SIP or IAX2 load testing programs so that I can try and > make it freeze up on demand? > > Thank you > Jesse > > On 22/01/2007, at 7:42 AM, Marc Blanchet wrote: > >> - openpbx and asterisk are not IPv6-enabled. so turning off or on >> IPv6 will not change anything regarding openpbx or asterisk. >> >> Marc. >> >> Le 07-01-21 à 14:28, Jesse Reynolds a écrit : >> >>> That is interesting. I do have IPv6 enabled on the ethernet >>> interface, however there is only one active interface, 'en0': >>> >>> phoney:/usr/local/openpbx.org/etc root# ifconfig -a >>> lo0: flags=8049<UP,LOOPBACK,RUNNING,MULTICAST> mtu 16384 >>> inet6 ::1 prefixlen 128 >>> inet6 fe80::1%lo0 prefixlen 64 scopeid 0x1 >>> inet 127.0.0.1 netmask 0xff000000 >>> gif0: flags=8010<POINTOPOINT,MULTICAST> mtu 1280 >>> stf0: flags=0<> mtu 1280 >>> en0: flags=8863<UP,BROADCAST,SMART,RUNNING,SIMPLEX,MULTICAST> mtu >>> 1500 >>> inet6 fe80::216:cbff:fea5:b549%en0 prefixlen 64 scopeid 0x4 >>> inet 192.168.1.8 netmask 0xffffff00 broadcast 192.168.1.255 >>> ether 00:16:cb:a5:b5:49 >>> media: autoselect (100baseTX <full-duplex,flow-control>) >>> status: active >>> supported media: autoselect 10baseT/UTP <half-duplex> >>> 10baseT/UTP <full-duplex> 10baseT/UTP <full-duplex,hw-loopback> >>> 10baseT/UTP <full-duplex,flow-control> 100baseTX <half-duplex> >>> 100baseTX <full-duplex> 100baseTX <full-duplex,hw-loopback> >>> 100baseTX >>> <full-duplex,flow-control> 1000baseT <full-duplex> 1000baseT <full- >>> duplex,hw-loopback> 1000baseT <full-duplex,flow-control> none >>> en1: flags=8863<UP,BROADCAST,SMART,RUNNING,SIMPLEX,MULTICAST> mtu >>> 1500 >>> ether 00:17:f2:51:ff:07 >>> media: autoselect (<unknown type>) status: inactive >>> supported media: autoselect >>> wlt1: flags=41<UP,RUNNING> mtu 1500 >>> fw0: flags=8863<UP,BROADCAST,SMART,RUNNING,SIMPLEX,MULTICAST> mtu >>> 2030 >>> lladdr 00:16:cb:ff:fe:e8:b2:d8 >>> media: autoselect <full-duplex> status: inactive >>> supported media: autoselect <full-duplex> >>> >>> And I do have DHCP elso enabled, perhaps this in combination with >>> IPv6 is causing oddness. I'll first try turning off IPv6 and see if >>> this helps. If not, I'll move to a static IP and try that. >>> >>> Jesse >>> >>> >>> On 22/01/2007, at 2:37 AM, Marc Olivier Chouinard wrote: >>> >>>> I know that when a machine has multiple NIC interface, you have to >>>> force >>>> it to 1 IP. I had these problems before but since I did force to >>>> one of >>>> my IP, it fixed the problem. >>>> >>>> I spoked with someone before chrismast that work on IPv6 update of >>>> chan_sip and actually did rewrite the binding support for >>>> chan_sip. He >>>> said he might beable to port his asterisk chan_sip changes to opbx >>>> but >>>> he would have to get back to me. >>>> >>>> I'll try to contact him but in the mean time if it not set static, >>>> try >>>> it out. >>>> >>>> Marc O. >>>> >>>> Jesse Reynolds wrote: >>>>> I think I'm having the same problem. After some hours of being up >>>>> (less than 12) my SJPhone can no longer register with SIP, and the >>>>> command "sip show registry" on the CLI shows my two remote sip >>>>> proxies as "Request Sent": >>>>> >>>>> phoney*CLI> sip show registry >>>>> Host Username Refresh State >>>>> fwd.pulver.com:5060 627598 105 Request >>>>> Sent >>>>> sip.sipme.com.au:5060 1777101186 105 Request >>>>> Sent >>>>> >>>>> netstat shows port 5060 udp listening OK, but it seems that the >>>>> SIP >>>>> stuff has hung in OpenPBX: >>>>> >>>>> phoney:~ root# netstat -an | grep 5060 >>>>> udp4 41864 0 *.5060 *.* >>>>> >>>>> looking at tcpdump on my mac mini openpbx server I can see many >>>>> packets coming from my laptop (192.168.1.30) on port 5060 but >>>>> never >>>>> anything sent in reply from OpenPBX: >>>>> >>>>> phoney:~ root# tcpdump -n host lester-young-iii.local and not >>>>> port 22 >>>>> tcpdump: verbose output suppressed, use -v or -vv for full >>>>> protocol >>>>> decode >>>>> listening on en0, link-type EN10MB (Ethernet), capture size 96 >>>>> bytes >>>>> 20:26:04.113960 IP 192.168.1.30.5003 > 192.168.1.255.5003: UDP, >>>>> length 306 >>>>> 20:26:04.130943 IP 192.168.1.30.5060 > 192.168.1.8.5060: SIP, >>>>> length: >>>>> 427 >>>>> 20:26:04.282297 IP 192.168.1.30.5003 > 192.168.1.255.5003: UDP, >>>>> length 24 >>>>> 20:26:04.630762 IP 192.168.1.30.5060 > 192.168.1.8.5060: SIP, >>>>> length: >>>>> 427 >>>>> 20:26:05.088671 IP 192.168.1.30.5003 > 192.168.1.255.5003: UDP, >>>>> length 312 >>>>> 20:26:05.093987 IP 192.168.1.30.5003 > 192.168.1.255.5003: UDP, >>>>> length 312 >>>>> 20:26:05.631358 IP 192.168.1.30.5060 > 192.168.1.8.5060: SIP, >>>>> length: >>>>> 427 >>>>> 20:26:07.632156 IP 192.168.1.30.5060 > 192.168.1.8.5060: SIP, >>>>> length: >>>>> 427 >>>>> 20:26:11.634999 IP 192.168.1.30.5060 > 192.168.1.8.5060: SIP, >>>>> length: >>>>> 427 >>>>> 20:26:15.635441 IP 192.168.1.30.5060 > 192.168.1.8.5060: SIP, >>>>> length: >>>>> 427 >>>>> 20:26:19.637148 IP 192.168.1.30.5060 > 192.168.1.8.5060: SIP, >>>>> length: >>>>> 427 >>>>> >>>>> Also, IAXPing mostly returns 'No data in one second', although >>>>> there >>>>> were two replies (See below), what does that mean? >>>>> >>>>> I've just compiled RC3 on a Mac Mini (intel) running 10.4.8 >>>>> according >>>>> to the howto on voip-info.org. >>>>> >>>>> Note, I haven't added any "qualify=yes" remarks in sip.conf. >>>>> >>>>> Has anyone logged a bug for this on openpbx's trac ? (Speaking of >>>>> which, I am disheartened to find that I can't find any way of >>>>> adding >>>>> a new ticket in trac! How do you do it?!) >>>>> >>>>> Jan/21/2007 20:22:31: Sending IAX ping to 192.168.1.8 >>>>> Jan/21/2007 20:22:31: Reply from: 192.168.1.8 >>>>> Jan/21/2007 20:22:32: No data in one second >>>>> Jan/21/2007 20:22:33: Reply from: 192.168.1.8 >>>>> Jan/21/2007 20:22:34: No data in one second >>>>> Jan/21/2007 20:22:35: No data in one second >>>>> Jan/21/2007 20:22:36: No data in one second >>>>> Jan/21/2007 20:22:37: No data in one second >>>>> Jan/21/2007 20:22:38: No data in one second >>>>> Jan/21/2007 20:22:39: No data in one second >>>>> Jan/21/2007 20:22:40: No data in one second >>>>> Jan/21/2007 20:22:41: No data in one second >>>>> Jan/21/2007 20:22:42: No data in one second >>>>> Jan/21/2007 20:22:43: No data in one second >>>>> Jan/21/2007 20:22:44: No data in one second >>>>> Jan/21/2007 20:22:45: No data in one second >>>>> Jan/21/2007 20:22:46: No data in one second >>>>> Jan/21/2007 20:22:47: No data in one second >>>>> Jan/21/2007 20:22:48: No data in one second >>>>> Jan/21/2007 20:22:49: No data in one second >>>>> Jan/21/2007 20:22:50: No data in one second >>>>> Jan/21/2007 20:22:51: No data in one second >>>>> Jan/21/2007 20:22:52: No data in one second >>>>> Jan/21/2007 20:22:53: No data in one second >>>>> Jan/21/2007 20:22:54: No data in one second >>>>> Jan/21/2007 20:22:55: No data in one second >>>>> Jan/21/2007 20:22:56: No data in one second >>>>> Jan/21/2007 20:22:57: No data in one second >>>>> Jan/21/2007 20:22:58: No data in one second >>>>> Jan/21/2007 20:23:00: No data in one second >>>>> >>>>> Regards >>>>> Jesse >>>>> >>>>> On 10/01/2007, at 5:03 PM, Cristian Draghici via List wrote: >>>>> >>>>> >>>>>> If openpbx works like Asterisk and the IAX module is loaded, you >>>>>> should get an answer when iax pinging it. IAX ping sends a UDP >>>>>> IAX >>>>>> ping packet and expects an answer back from the server. >>>>>> >>>>>> Cocoa version here: >>>>>> >>>>>> http://freshmeat.net/projects/cocoaiaxping/ >>>>>> >>>>>> Also note that IAX ping works on UDP port 4569 (I think 5060 is >>>>>> SIP). >>>>>> >>>>>> Hope this helps, >>>>>> Cristi >>>>>> >>>>>> On 1/9/07, Mark Engelhardt via List <mac-telephony- >>>>>> [EMAIL PROTECTED]> wrote: >>>>>> >>>>>>> On Jan 9, 2007, at 2:39 AM, Jason Garland via List wrote: >>>>>>> >>>>>>> >>>>>>>> telnet uses TCP and openpbx uses UDP. You would not be able to >>>>>>>> telnet to it. >>>>>>>> >>>>>>>> Type this command to see if openpbx is listening on port 5060: >>>>>>>> netstat -an|grep 5060 >>>>>>>> >>>>>>> Hum, so I have to amend my bug report. openpbx does seem to be >>>>>>> listening on 5060, but there is no indication that it is >>>>>>> responding >>>>>>> to traffic. >>>>>>> >>>>>>> I have this happening on my laptop right now if anyone can >>>>>>> suggest >>>>>>> some diagnostic tests I might run. >>>>>>> >>>>>>> Mark >>>>>>> >>>>>>>> Are these phones behind NAT? >>>>>>>> >>>>>>>> - Jason >>>>>>>> >>>>>>>> >>>>>>>> On 1/9/07, Mark Engelhardt via List <mac-telephony- >>>>>>>> [EMAIL PROTECTED]> wrote: >>>>>>>> >>>>>>>>> Hello Everyone, >>>>>>>>> >>>>>>>>> I have found what I think is a new bug in OpenPBX PR3 >>>>>>>>> >>>>>>>>> Tonight, On a mac G4 Xserver running 10.4.8: >>>>>>>>> >>>>>>>>> I shutdown asterisk, installed OpenPBX, Copied over my >>>>>>>>> iax.conf, >>>>>>>>> extensions.conf and sip.conf from my asterisk config, >>>>>>>>> >>>>>>>>> then started up openPBX... >>>>>>>>> >>>>>>>>> OpenPBX launches fine and them after about 3-4 seconds it >>>>>>>>> stops >>>>>>>>> registering phones and while the CLI commands work, nothing >>>>>>>>> changes >>>>>>>>> after that. I did a telnet to port 5060 and it was not open. >>>>>>>>> >>>>>>>>> I have seen this same sort of behavior on mac os x with only >>>>>>>>> one >>>>>>>>> or 2 >>>>>>>>> phones, but it was not reproducible, it could take from a few >>>>>>>>> hours >>>>>>>>> to a few days to fail, but with tonight's setup (about 30 >>>>>>>>> phones, it >>>>>>>>> fails within a few seconds of launch. >>>>>>>>> >>>>>>>>> Is this a new bug? How can I help to isolate it? What should I >>>>>>>>> try? >>>>>>>>> Shall I post this to Trak? Can anyone reproduce this? >>>>>>>>> >>>>>>>>> Here are the messages I get: >>>>>>>>> >>>>>>>>> OpenPBX Ready. >>>>>>>>> *CLI> -- Registered SIP 'matt' at 192.168.21.162 port >>>>>>>>> 52281 >>>>>>>>> expires 60 >>>>>>>>> -- Saved useragent "Cisco-CP7960G/8.0" for peer matt >>>>>>>>> -- Saved useragent "Cisco-CP7960G/8.0" for peer marke >>>>>>>>> -- Registered SIP 'curt' at 192.168.21.159 port 50302 >>>>>>>>> expires 60 >>>>>>>>> -- Saved useragent "Cisco-CP7960G/8.0" for peer curt >>>>>>>>> -- Registered SIP 'hapy' at 192.168.21.156 port 51999 >>>>>>>>> expires 60 >>>>>>>>> -- Saved useragent "Cisco-CP7960G/8.0" for peer hapy >>>>>>>>> -- Registered SIP 'joyce' at 192.168.21.230 port 52846 >>>>>>>>> expires 60 >>>>>>>>> -- Saved useragent "Cisco-CP7960G/8.0" for peer joyce >>>>>>>>> -- Registered SIP 'jaime' at 192.168.21.241 port 50163 >>>>>>>>> expires 60 >>>>>>>>> -- Saved useragent "Cisco-CP7960G/8.0" for peer jaime >>>>>>>>> -- Registered SIP 'daniel' at 192.168.21.236 port 52782 >>>>>>>>> expires 60 >>>>>>>>> -- Saved useragent "Cisco-CP7960G/8.0" for peer daniel >>>>>>>>> -- Registered SIP 'Veronica' at 192.168.21.203 port 52015 >>>>>>>>> expires 60 >>>>>>>>> -- Saved useragent "Cisco-CP7960G/8.0" for peer Veronica >>>>>>>>> -- Registered SIP 'morgan' at 192.168.21.152 port 51506 >>>>>>>>> expires 60 >>>>>>>>> -- Saved useragent "Cisco-CP7960G/8.0" for peer morgan >>>>>>>>> -- Registered SIP 'rob' at 192.168.21.228 port 51725 >>>>>>>>> expires 60 >>>>>>>>> -- Saved useragent "Cisco-CP7960G/8.0" for peer rob >>>>>>>>> -- Registered SIP 'jaime' at 192.168.21.241 port 50164 >>>>>>>>> expires 60 >>>>>>>>> -- Registered SIP 'jon' at 192.168.21.244 port 52861 >>>>>>>>> expires 60 >>>>>>>>> -- Saved useragent "Cisco-CP7960G/8.0" for peer jon >>>>>>>>> -- Saved useragent "Cisco-CP7960G/8.0" for peer 576 >>>>>>>>> -- Registered SIP 'tony' at 192.168.21.248 port 51913 >>>>>>>>> expires 60 >>>>>>>>> -- Saved useragent "Cisco-CP7960G/8.0" for peer tony >>>>>>>>> -- Saved useragent "Cisco-CP7960G/8.0" for peer 575 >>>>>>>>> -- Registered SIP 'jaime' at 192.168.21.241 port 50165 >>>>>>>>> expires 60 >>>>>>>>> -- Registered SIP 'WaitsStore2' at 192.168.21.232 port >>>>>>>>> 52648 >>>>>>>>> expires 60 >>>>>>>>> -- Saved useragent "Cisco-CP7960G/8.0" for peer >>>>>>>>> WaitsStore2 >>>>>>>>> -- Saved useragent "Cisco-CP7905/8.0.1-060412A" for peer >>>>>>>>> ConsultRm2 >>>>>>>>> -- Registered SIP 'edward' at 192.168.21.240 port 50778 >>>>>>>>> expires 60 >>>>>>>>> -- Saved useragent "Cisco-CP7960G/8.0" for peer edward >>>>>>>>> Jan 9 04:10:47 NOTICE[6938]: chan_sip.c:14427 >>>>>>>>> sip_poke_noanswer: >>>>>>>>> Peer 'marke' is now UNREACHABLE! Last qualify: 0 >>>>>>>>> Jan 9 04:10:47 NOTICE[6938]: chan_sip.c:14427 >>>>>>>>> sip_poke_noanswer: >>>>>>>>> Peer '576' is now UNREACHABLE! Last qualify: 0 >>>>>>>>> Jan 9 04:10:47 NOTICE[6938]: chan_sip.c:14427 >>>>>>>>> sip_poke_noanswer: >>>>>>>>> Peer '575' is now UNREACHABLE! Last qualify: 0 >>>>>>>>> Jan 9 04:10:47 NOTICE[6938]: chan_sip.c:14427 >>>>>>>>> sip_poke_noanswer: >>>>>>>>> Peer 'greg' is now UNREACHABLE! Last qualify: 0 >>>>>>>>> >>>>>>>>> After that message, the openpbx cli still works, but something >>>>>>>>> internal is gone, (like sip show peers never updates) >>>>>>>>> >>>>>>>>> >>>>>>>>> >>> _______________________________________________ >>> Openpbx-dev mailing list >>> [email protected] >>> http://lists.openpbx.org/mailman/listinfo/openpbx-dev >> >> ----- >> IPv6 book: Migrating to IPv6, Wiley, 2006, http://www.ipv6book.ca >> >> >> _______________________________________________ >> Openpbx-dev mailing list >> [email protected] >> http://lists.openpbx.org/mailman/listinfo/openpbx-dev > > Jesse Reynolds > Virtual Artists Pty Ltd - http://www.va.com.au/ > Phone: 08 8223 2288 Mobile: 0414 669 790 > > _______________________________________________ > Openpbx-dev mailing list > [email protected] > http://lists.openpbx.org/mailman/listinfo/openpbx-dev Jesse Reynolds Virtual Artists Pty Ltd - http://www.va.com.au/ Phone: 08 8223 2288 Mobile: 0414 669 790 _______________________________________________ Openpbx-dev mailing list [email protected] http://lists.openpbx.org/mailman/listinfo/openpbx-dev
