Description of problem: Similarly to the multipathd problem; ccsd is spamming the syslog every 30 seconds after a default @Everything kickstart. I don't have any cluster stuff set up. May 5 13:23:04 trash ccsd[2102]: Unable to parse /etc/cluster/cluster.conf May 5 13:23:04 trash ccsd[2102]: Searching cluster for valid copy. May 5 13:23:04 trash ccsd[2102]: The appropriate config file could not be loaded. May 5 13:23:04 trash ccsd[2102]: Error while processing connect: No data available May 5 13:23:34 trash ccsd[2102]: Unable to connect to cluster infrastructure after 30 seconds. May 5 13:24:07 trash ccsd[2102]: Unable to connect to cluster infrastructure after 60 seconds. ... etc see bug 155893 for similar discussion w.r.t. multipathd. Version-Release number of selected component (if applicable): ccs-0.25-0.1 Marked as FC4 Blocker.
This "spamming" was actually a requested feature to help customers debug problems when setting up clusters. I think the real problem here is that the init scripts are on by default (bug 156410). In FC, these scripts should be off. In RHEL, where these packages are part of a subchannel, they should probably be on.