Description of problem: I've got mcastport: 5405 in both corosync.conf (attached) but according to strace corosync is using port 7445: [pid 4977] bind(4, {sa_family=AF_INET, sin_port=htons(7445), sin_addr=inet_addr("226.94.1.1")}, 16) = 0 This is confirmed by changing firewall settings: corosync works if I allow udp 7445 but if I allow udp 5405 it goes into a loop repeating this in the log file: Dec 29 9:59:49.265860 [TOTEM] entering GATHER state from 3. Dec 29 9:59:51.094313 [TOTEM] The consensus timeout expired.
Created attachment 327924 [details] configuration file
verified with trunk of corosync. I'll sort it out in the next revision into fedora. Thanks for the bug report. Regards -steve
also please note it isn't ignoring the setting. Instead it has wrong byte order for the port number (7445 is actually 5405 if you endian swap).
*** Bug 478890 has been marked as a duplicate of this bug. ***
I believe this is fixed in latest revision of tree.