Hi,
We are running some Kamailio proxies with a lot of TLS clients. And they log a lot of TCP "noise" during the day, like those messages:
ERROR: <core> [core/tcp_read.c:267]: tcp_read_data(): error reading: Connection timed out (110) ([]:29265 -> []:5061) ERROR: <core> [core/tcp_read.c:267]: tcp_read_data(): error reading: Connection reset by peer (104) ([]:12013 -> []:5061)
That's when connections close, apparently. (BTW, those are original messages, I did not remove IP addresses.)
But sometimes I see this line:
ERROR: <core> [core/tcp_read.c:267]: tcp_read_data(): error reading: No space left on device (28) ([]:62892 -> []:5061)
To me, this does not make sense. Can somebody explain what this means? Do I have to tweak something somewhere?
Thanks, Regards, Sebastian
Hi Sebastian,
we're observing the same Log lines since upgrading to 5.8.x We did some research and compared the port number of those log lines with some pcap traces.
In all cases it was a connection timeout to the tls client.
Maybe it's the same problem like the one in this issue: https://github.com/kamailio/kamailio/issues/3695
Regards,
Hendrik Wedhorn sipgate