concerning sticky cluster host connection failure

Read 10886 times
It states:
If you receive the following message:

Code: [Select]
Cluster host connection failure for Local server: Connection refused (111)
...then cc-control has exited for some reason.
---------------------------------------------------------

My v3 Beta was functioning with no error until I had an issue with my original license.

My cc-control seems to be running but I went ahead and shut down and re-started with debug.

Debug output so far:

Reading symbols from /usr/local/centovacast/sbin/cc-control...done.
Licensed to George L... spin-stream.....com <all correct
Centova Cast RPC server build 20121026.62 listening on 127.0.0.1:2198

After restarting I still get the error.

note:
My v3 Beta was functioning with no error until I had an issue with my original license.


any idea?
my direct email is cage@linuxcage.org
TO fix it I had to change the listen host to the external ip of the server. cc-control.conf. Line MASTER_ADDR=127.0.0.1 change to MASTER_ADDR=externalip .. This might not be the correct way to do this but It worked for me.
Then I changed the "nodes" ip in the cpanel to the external.
Debug output so far:
Per the sticky, if it's crashing, you need to leave it running under gdb until it crashes again.  The startup information doesn't provide any information about why it crashed.

That said, if it's not coming back up, then you have bigger problems on your server... namely, your cron jobs likely aren't running properly.  We added a watchdog system to Centova Cast awhile back so even if cc-control goes down, it should be restarted within 60 seconds so you'll rarely even notice.
TO fix it I had to change the listen host to the external ip of the server. cc-control.conf. Line MASTER_ADDR=127.0.0.1 change to MASTER_ADDR=externalip .. This might not be the correct way to do this but It worked for me.
Unless you're running cc-control on a different server than cc-web, this will not make any difference.