I modified t22_timeout() in l4isup.c so that it would pretend as received GRA and reset the circuit to idle. I tried making an outgoing call and it is working(my phone rings), but I cannot hear anything. When I ended the call on my phone, asterisk doesn't get anything from the other side and therefore holding the circuit as busy. If I end the call on the asterisk side, the call is ended on the phone properly and asterisk will get T1 timeout(waiting for RLC).
static int t22_timeout(void *arg) {
struct ss7_chan *pvt = arg;
ast_log(LOG_NOTICE, "T22 timeout (No \"circuit group reset acknowledge\" from peer) CIC=%d.\n", pvt->cic);
//isup_send_grs(pvt, pvt->grs_count, 0);
/* hack start */
stop_timer(pvt->t22);
pvt->t22 = -1;
stop_timer(pvt->t23);
pvt->t23 = -1;
pvt->grs_count = -1;
pvt->reset_done = 1;
/* hack end */
return 1; /* Run us again the next period */
}
Another thing that I noticed, is that it seems so far I have never received any ISUP message from the switch side(therefore getting timeout), although the switch seems to be able to receive message from asterisk(else my phone wont ring). I am suspecting that right now the ISUP traffic only work one-way(from asterisk to switch) but the telco guy told me that they have checked the SS7/ISUP configuration on their side and doesn't find any problem. Does anyone has any idea about what possibly went wrong with the configuration(on the telco side)? I am not familiar with how the telco/switch environment works.
Thanks again.
--- On Sun, 20/7/08, Jakub Klausa <j.klausa at ss7.pl> wrote:
> From: Jakub Klausa <j.klausa at ss7.pl>
> Subject: Re: [asterisk-ss7] chan_ss7 - T22 timeout (No'circuitgroup resetacknowledge' from peer)
> To: asterisk-ss7 at lists.digium.com
> Date: Sunday, 20 July, 2008, 3:55 AM
> On Sat, Jul 19, 2008 at 08:49:20PM +0200, Pawel Ratajewski
> (Forweb) wrote:
>
> => i call 'fine' if I'm able to
> block/unblock one or group of channels :)
>
> Either you're using different chan_ss7, or we differ on
> the definition
> 'working fine'. The fact that the 'ss7 show
> channels' shows you what you
> expects it to doesn't mean the other side of the link
> sees the same. That's
> what interconnect signalling tests are for, and if
> you're using stock
> chan_ss7 they shouldn't have passed. First of all, you
> can't invoke BLOs
> from the chan_ss7 interface, which is what you could and
> should use for a
> single CIC blocking. That's Q.784 tests 1.3.2 - the
> whole group. Second, the CGB/CGUs sent by stock chan_ss7
> are always sent with range=32, no matter what
> you'll give it as an input option. Here you should have
> failed on tests
> 1,3,1 - the whole group again. And it definetly should have
> been tested,
> because it's used in everyday operations.
>
> => It works fine for me :) The range is always shown as
> 32, but it's really
> => different. But the problem is, chan_ss7 sends to many
> octets - the last is
> => empty, but some od DGT does not recognize its as
> empty.
>
> Well, if it works fine, and you insist on the fact that the
> chan_ss7 'ss7
> block/unblock' works allright, then the other end has
> some serious problems
> sending the CGAs with range=32 everytime.
>
> But seriously, belive me, it's chan_ss7's fault. It
> even got mentioned here
> on the list once or twice. It was supposed to get fixed,
> but i'm not
> entirely sure it did.
>
> Anyway - there's an easy way finding out - either put
> your protocol analyzer
> up to the task, or contact the other end for a test run to
> see if they see
> what that you're supposedly sending towards them.
>
> --
> Jakub Klausa | j.klausa at ss7.pl | http://www.ss7.pl/ |
> http://www.ngpbx.pl/
> Dane rejestrowe ->
> http://kontakt.ss7.pl_______________________________________________
> --Bandwidth and Colocation Provided by
> http://www.api-digital.com--
>
> asterisk-ss7 mailing list
> To UNSUBSCRIBE or update options visit:
> http://lists.digium.com/mailman/listinfo/asterisk-ss7
__________________________________________________________
Not happy with your email address?.
Get the one you really want - millions of new email addresses available now at Yahoo! http://uk.docs.yahoo.com/ymail/new.html
|