[mythtv-users] Mythbackend restarts at top of hour when remote backend is asked to record

[John Nissley]

I am running the latest trunk of mythtv and for the last couple of 
months I have had a problem with the master backend cpu shooting through 
the roof which appears to cause the master backend to stop. I have 
systemctl restart on failure and after restart things are fine.

I am running three hdhr devices connected to an antenna. Two hdhr 
devices are connected to the master backend and one is connected to the 
slave backend mostly for disk and network i/o reasons. I have found that 
most times a recording starts on the slave backend and a recording also 
starts on the master backend the master backend CPU will shoot through 
the roof and the backend process will die.  I turned on debug logging 
and did not really see anything interesting.

The only error that was interesting is "GLib-ERROR **: Creating pipes 
for GWakeup: Too many open files"

I guess my question is should I just go to one master backend but if I 
do that then the commercial flagging becomes limited and I need to make 
sure I have enough hard drives to spread out the disk IO or does someone 
know how to fix this issue I am seeing.