actually that is the 31 bit max, not the 32 bit... but thats a different story.  There is now a patch in tree that fixes this issue.

On Sep 8, 2014, at 3:21 PM, Spencer Thomason <spencer@5ninesolutions.com> wrote:

Yes, you are correct, all devices freaked out until they we rebooted.  The issue appears to be that the Cseq exceeded 32bit max.  Before this change I was seeing Cseq values as 2147483647 which is 32bit max.  As soon as the dust clears I’ll file a jira.  I don’t really know what a good upgrade path might be but it seems the calculation method has to change.

I guess we do have to limit the value to 32bits:
8.1.1.5 CSeq

   The CSeq header field serves as a way to identify and order
   transactions.  It consists of a sequence number and a method.  The
   method MUST match that of the request.  For non-REGISTER requests
   outside of a dialog, the sequence number value is arbitrary.  The
   sequence number value MUST be expressible as a 32-bit unsigned
   integer and MUST be less than 2**31.  As long as it follows the above
   guidelines, a client may use any mechanism it would like to select
   CSeq header field values.

Thanks,
Spencer



On Sep 8, 2014, at 9:31 AM, Michael Jerris <mike@jerris.com> wrote:

changing this on a busy switch will almost for sure make presence not work correctly.

On Sep 8, 2014, at 12:07 PM, Spencer Thomason <spencer@5ninesolutions.com> wrote:

Hi Mike,
I know this is an insanely dirty hack but can I change the following line in sofia_presence.c to a smaller multiple?
2115 callsequence = (now - mod_sofia_globals.presence_epoch) * 100;
i.e.:
callsequence = (now - mod_sofia_globals.presence_epoch) * 10;

Thanks in advance!
Spencer



On Sep 8, 2014, at 8:29 AM, Yehavi Bourvine <yehavi.bourvine@gmail.com> wrote:

I tried locating how the Caeq is generated. Didn't manage yet to get all the details, but the initial value is dependant on the current time. Seems like we have Bug-2k on Polycom :-(

                __Yehavi:


2014-09-08 17:41 GMT+03:00 Spencer Thomason <spencer@5ninesolutions.com>:
I just ran into the same issue across several different sites.  Thanks Polycom for the exciting Monday morning.  I’ll try to collect some traces once the fires die down.

BR,
Spencer



On Sep 8, 2014, at 6:48 AM, Michael Jerris <mike@jerris.com> wrote:

> cseq for notify and invite have nothing at all to do with each other.  A sip trace would probably help what is going on make more sense.
>
> On Sep 7, 2014, at 11:14 AM, Yehavi Bourvine <yehavi.bourvine@gmail.com> wrote:
>
>> Hi,
>>
>>  My Polycom phones stopped showing presence today. They reply with "loop detected" after they receive the NOTIFY message. In their Syslog they complain that Cseq is outside boundaries...
>> The Cseq for INVITE is much smaller.
>>
>> How can I reset the Cseq to lower values?
>>
>>                  Thanks! __Yehavi:
>
>
_________________________________________________________________________
Professional FreeSWITCH Consulting Services:
consulting@freeswitch.org
http://www.freeswitchsolutions.com

Official FreeSWITCH Sites
http://www.freeswitch.org
http://confluence.freeswitch.org
http://www.cluecon.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server
http://www.cudatel.com

FreeSWITCH-users mailing list
FreeSWITCH-users@lists.freeswitch.org
http://lists.freeswitch.org/mailman/listinfo/freeswitch-users
UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users
http://www.freeswitch.org

_________________________________________________________________________
Professional FreeSWITCH Consulting Services:
consulting@freeswitch.org
http://www.freeswitchsolutions.com

Official FreeSWITCH Sites
http://www.freeswitch.org
http://confluence.freeswitch.org
http://www.cluecon.com

FreeSWITCH-powered IP PBX: The CudaTel Communication Server
http://www.cudatel.com

FreeSWITCH-users mailing list
FreeSWITCH-users@lists.freeswitch.org
http://lists.freeswitch.org/mailman/listinfo/freeswitch-users
UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users
http://www.freeswitch.org