Hi,
I am experiencing two problem:
First:
I am experiencing that tls read buffer overflow, if i use default value
of tcp_rd_buf_size
If i doubled the size of read buffer and i use
tcp_rd_buf_size=8192
then the problem seems to be solved.
Second:
I am experiencing that TLS connection is dropped by sip proxy very
frequently!
And i think the tcp connection shouldn't dropped! So my guess is that
TLS communication is not restarting the timeout counters!
So in every 2 minutes the sip-router is restarting the TCP connection
with tcp RST.
like this:
Sep 5 11:28:48 hal /usr/sbin/kamailio[8129]: DEBUG: <core>
[tcp_main.c:4304]: tcp_main: entering timer for 0xac537c74
(ticks=446299809, ti
meout=446299809 (0 s), wr_timeout=0 (240541717 s)), write queue: 0 bytes
Sep 5 11:28:48 hal /usr/sbin/kamailio[8129]: DEBUG: <core>
[tcp_main.c:4346]: tcp_main: timeout for 0xac537c74
Sep 5 11:28:48 hal /usr/sbin/kamailio[8129]: DEBUG: <core>
[io_wait.h:617]: DBG: io_watch_del (0x824cf80, 48, -1, 0x10) fd_no=41 called
Sep 5 11:28:48 hal /usr/sbin/kamailio[8129]: DEBUG: tls
[tls_server.c:597]: Closing SSL connection 0xac4a0374
Sep 5 11:28:48 hal /usr/sbin/kamailio[8129]: DEBUG: <core>
[ip_addr.c:247]: tcpconn_new: new tcp connection: 10.10.10.10
Sep 5 11:28:48 hal /usr/sbin/kamailio[8129]: DEBUG: <core>
[tcp_main.c:1081]: tcpconn_new: on port 36142, type 3
Sep 5 11:28:48 hal /usr/sbin/kamailio[8129]: DEBUG: <core>
[tcp_main.c:1382]: tcpconn_add: hashes: 2473:3781:3367, 5
This second problem is not solved and very urgent for me.
Any help suggestion is highly appreciated!
Thanks,
Misi