THIS IS AN AUTOMATED MESSAGE, DO NOT REPLY.
A user has added themself to the list of users assigned to this task.
FS#130 - kamailio debian build error
User who did this - Marcus Hunger (fnordian)
http://sip-router.org/tracker/index.php?do=details&task_id=130
You are receiving this message because you have requested it from the Flyspray bugtracking system. If you did not expect this message or don't want to receive mails in future, you can change your notification settings at the URL shown above.
Hello,
I’m asking your help because I’m facing with the following issue:
1. The user A deletes the contact B from its own contact list;
2. The client sends a XCAP PUT event in order to update the user A
resource- list (removing the contact B) and to remove the watcher A from the
user B presence-rules;
3. Then the Presence Server, instead of remove the record “A watcher of
B presentity-uri” in the watcher table, changes the status value of this
record from Active (1) to Pending (2). So that , correctly, the “pending
record” generates a Notify event against the user B: the authorization
pop-up is shown to the user B as if A had added contact B.
The expected behavior should see the Presence Server removing the record “A
watcher and B presentity-uri” from the watcher table or set the status in a
“terminated” status.
Could you please tell me how to solve this issue?
It’s a known bug?
About the env: Kamailio server v. 3.1.3 as registrar and presence server and
Jitsi Sip Communicator as SIP Client.
Thanks a lot and kind Regards.
laura
Module: sip-router
Branch: 3.1
Commit: 521b16ea814d68f9d872e2f6e1b0f081f285254b
URL: http://git.sip-router.org/cgi-bin/gitweb.cgi/sip-router/?a=commit;h=521b16e…
Author: Andrei Pelinescu-Onciul <andrei(a)iptel.org>
Committer: Andrei Pelinescu-Onciul <andrei(a)iptel.org>
Date: Thu May 5 00:13:36 2011 +0200
tcp: fix multiple hash removal attempts
In some corner cases (pending new tcp connection created in async
mode, other processes try to append data to it but fail and the
initial send fails after that) it was possible to attempt removing
the connection from the hash and the local timer multiple times.
(cherry picked from commit 76cb63c6f120c22a134babc8948611819b9081c3)
---
tcp_main.c | 41 +++++++++++++++++++++++------------------
1 files changed, 23 insertions(+), 18 deletions(-)
diff --git a/tcp_main.c b/tcp_main.c
index 31fc0fa..0c9db6b 100644
--- a/tcp_main.c
+++ b/tcp_main.c
@@ -2167,20 +2167,19 @@ conn_wait_close:
su2a(&c->rcv.src_su, sizeof(c->rcv.src_su)),
fd, c->flags, strerror(errno), errno);
}
+ /* here the connection is for sure in the hash (tcp_main will not
+ remove it because it's marked as PENDing) and the refcnt is at least
+ 2
+ */
TCPCONN_LOCK;
- if (c->flags & F_CONN_HASHED){
- /* if some other parallel tcp_send did send CONN_ERROR to
- * tcp_main, the connection might be already detached */
- _tcpconn_detach(c);
- c->flags&=~F_CONN_HASHED;
- TCPCONN_UNLOCK;
- tcpconn_put(c);
- }else
- TCPCONN_UNLOCK;
+ _tcpconn_detach(c);
+ c->flags&=~F_CONN_HASHED;
+ tcpconn_put(c);
+ TCPCONN_UNLOCK;
/* dec refcnt -> mark it for destruction */
tcpconn_chld_put(c);
return n;
-#endif /* TCP_CONNET_WAIT */
+#endif /* TCP_CONNECT_WAIT */
release_c:
tcpconn_chld_put(c); /* release c (dec refcnt & free on 0) */
end_no_deref:
@@ -3594,16 +3593,22 @@ inline static int handle_ser_child(struct process_table* p, int fd_i)
tcpconn->flags);
case CONN_EOF: /* forced EOF after full send, due to send flags */
#ifdef TCP_CONNECT_WAIT
- /* if the connection is pending => it might be on the way of
- * reaching tcp_main (e.g. CONN_NEW_COMPLETE or
- * CONN_NEW_PENDING_WRITE) => it cannot be destroyed here */
- if ( !(tcpconn->flags & F_CONN_PENDING) &&
- tcpconn_try_unhash(tcpconn) )
- tcpconn_put(tcpconn);
-#else /* ! TCP_CONNECT_WAIT */
+ /* if the connection is marked as pending => it might be on
+ * the way of reaching tcp_main (e.g. CONN_NEW_COMPLETE or
+ * CONN_NEW_PENDING_WRITE) => it cannot be destroyed here,
+ * it will be destroyed on CONN_NEW_COMPLETE /
+ * CONN_NEW_PENDING_WRITE or in the send error case by the
+ * sender process */
+ if (unlikely(tcpconn->flags & F_CONN_PENDING)) {
+ if (tcpconn_put(tcpconn))
+ tcpconn_destroy(tcpconn);
+ /* no need for io_watch_del(), if PENDING it should not
+ be watched for anything in tcp_main */
+ break;
+ }
+#endif /* TCP_CONNECT_WAIT */
if ( tcpconn_try_unhash(tcpconn) )
tcpconn_put(tcpconn);
-#endif /* TCP_CONNECT_WAIT */
if ( ((tcpconn->flags & (F_CONN_WRITE_W|F_CONN_READ_W)) ) &&
(tcpconn->s!=-1)){
io_watch_del(&io_h, tcpconn->s, -1, IO_FD_CLOSING);
Module: sip-router
Branch: master
Commit: 76cb63c6f120c22a134babc8948611819b9081c3
URL: http://git.sip-router.org/cgi-bin/gitweb.cgi/sip-router/?a=commit;h=76cb63c…
Author: Andrei Pelinescu-Onciul <andrei(a)iptel.org>
Committer: Andrei Pelinescu-Onciul <andrei(a)iptel.org>
Date: Thu May 5 00:13:36 2011 +0200
tcp: fix multiple hash removal attempts
In some corner cases (pending new tcp connection created in async
mode, other processes try to append data to it but fail and the
initial send fails after that) it was possible to attempt removing
the connection from the hash and the local timer multiple times.
---
tcp_main.c | 41 +++++++++++++++++++++++------------------
1 files changed, 23 insertions(+), 18 deletions(-)
diff --git a/tcp_main.c b/tcp_main.c
index 31fc0fa..0c9db6b 100644
--- a/tcp_main.c
+++ b/tcp_main.c
@@ -2167,20 +2167,19 @@ conn_wait_close:
su2a(&c->rcv.src_su, sizeof(c->rcv.src_su)),
fd, c->flags, strerror(errno), errno);
}
+ /* here the connection is for sure in the hash (tcp_main will not
+ remove it because it's marked as PENDing) and the refcnt is at least
+ 2
+ */
TCPCONN_LOCK;
- if (c->flags & F_CONN_HASHED){
- /* if some other parallel tcp_send did send CONN_ERROR to
- * tcp_main, the connection might be already detached */
- _tcpconn_detach(c);
- c->flags&=~F_CONN_HASHED;
- TCPCONN_UNLOCK;
- tcpconn_put(c);
- }else
- TCPCONN_UNLOCK;
+ _tcpconn_detach(c);
+ c->flags&=~F_CONN_HASHED;
+ tcpconn_put(c);
+ TCPCONN_UNLOCK;
/* dec refcnt -> mark it for destruction */
tcpconn_chld_put(c);
return n;
-#endif /* TCP_CONNET_WAIT */
+#endif /* TCP_CONNECT_WAIT */
release_c:
tcpconn_chld_put(c); /* release c (dec refcnt & free on 0) */
end_no_deref:
@@ -3594,16 +3593,22 @@ inline static int handle_ser_child(struct process_table* p, int fd_i)
tcpconn->flags);
case CONN_EOF: /* forced EOF after full send, due to send flags */
#ifdef TCP_CONNECT_WAIT
- /* if the connection is pending => it might be on the way of
- * reaching tcp_main (e.g. CONN_NEW_COMPLETE or
- * CONN_NEW_PENDING_WRITE) => it cannot be destroyed here */
- if ( !(tcpconn->flags & F_CONN_PENDING) &&
- tcpconn_try_unhash(tcpconn) )
- tcpconn_put(tcpconn);
-#else /* ! TCP_CONNECT_WAIT */
+ /* if the connection is marked as pending => it might be on
+ * the way of reaching tcp_main (e.g. CONN_NEW_COMPLETE or
+ * CONN_NEW_PENDING_WRITE) => it cannot be destroyed here,
+ * it will be destroyed on CONN_NEW_COMPLETE /
+ * CONN_NEW_PENDING_WRITE or in the send error case by the
+ * sender process */
+ if (unlikely(tcpconn->flags & F_CONN_PENDING)) {
+ if (tcpconn_put(tcpconn))
+ tcpconn_destroy(tcpconn);
+ /* no need for io_watch_del(), if PENDING it should not
+ be watched for anything in tcp_main */
+ break;
+ }
+#endif /* TCP_CONNECT_WAIT */
if ( tcpconn_try_unhash(tcpconn) )
tcpconn_put(tcpconn);
-#endif /* TCP_CONNECT_WAIT */
if ( ((tcpconn->flags & (F_CONN_WRITE_W|F_CONN_READ_W)) ) &&
(tcpconn->s!=-1)){
io_watch_del(&io_h, tcpconn->s, -1, IO_FD_CLOSING);
