|
From: | Aymeric Moizard |
Subject: | [osip-dev] eXosip, fix for eXosip_call_build_prack |
Date: | Thu, 20 Nov 2014 16:29:53 +0100 |
hi aymeric,
sorry, that was my mistake; the initialization of rseq is not necessary.
but in my opinion there is a failure in function eXosip_call_build_prack:
instead of
osip_message_header_get_byname (tr->last_response, "RSeq", 0, &rseq);
if (rseq == NULL && rseq->hvalue == NULL) {
return OSIP_WRONG_FORMAT;
}
it should be:
osip_message_header_get_byname (tr->last_response, "RSeq", 0, &rseq);
if (rseq == NULL || rseq->hvalue == NULL) {
return OSIP_WRONG_FORMAT;
}
if there is no RSeq header present (rseq=NULL) in the last response,
then rseq is nevertheless accessed to evaluate rseq->hvalue !!
br,
christoph
From: Aymeric Moizard [mailto:address@hidden]
Sent: Donnerstag, 20. November 2014 14:57
To: FEICHTER Christoph
Cc: address@hidden
Subject: Re: Auto-discard notification
Hi,
osip_message_header_get_byname contains this code:
"*dest=NULL"
This should set rseq = NULL upon error?
Can you check which version you are using? and update to latest osip and eXosip?
Latest version doesn't seems affected?
Also note that you have to subscribe to the mailing list in order to send email to address@hidden
You can subscribe to the mailing list with by sending a mail to <address@hidden> with the subject set to "subscribe"
Regards
Aymeric
2014-11-20 14:49 GMT+01:00 <address@hidden>:
The attached message has been automatically discarded.
---------- Message transféré ----------
From: FEICHTER Christoph <address@hidden>
To: "address@hidden" <address@hidden>
Cc:
Date: Thu, 20 Nov 2014 13:49:09 +0000
Subject: eXosip, fix for eXosip_call_build_prack
hi aymeric,
I eventually found a problem, when generating a PRACK.
in eXcall_api.c, function eXosip_call_build_prack
the following variable is un-initialized:
osip_header_t *rseq;
I suppose we should replace this line by
osip_header_t *rseq = NULL;
even if this case is quite unlikely,
but we just had this scenario – 18x with missing RSeq header -
and it caused a crash in eXosip J
br,
Christoph
--
Antisip - http://www.antisip.com
[Prev in Thread] | Current Thread | [Next in Thread] |