Description of problem: lock_gulmd ignores integer values for entries that are suppose to be floats (e.g. heartbeat_rate). Version-Release number of selected component (if applicable): GFS-6.0.2.25-0 How reproducible: Always Steps to Reproduce: 1. ccs_read file cluster.ccs cluster { name = "floater" lock_gulm { servers = [ "trin-01.lab.msp.redhat.com" ] heartbeat_rate = 10 allowed_misses = 1 } } 2. lock_gulmd 3. killall -10 lock_gulmd 4. cat /tmp/Gulm_config.wTCXvI #=============================================================================== # hashed: 0xee0e8746 cluster { name = "floater" lock_gulm { heartbeat_rate = 15.000 allowed_misses = 1 coreport = 40040 new_connection_timeout = 15.000 # server cnt: 1 # servers = ["trin-01.lab.msp.redhat.com"] servers = ["10.15.84.171"] lt_partitions = 1 lt_base_port = 41040 lt_high_locks = 1048576 lt_drop_req_rate = 10 prealloc_locks = 90000 prealloc_holders = 130000 prealloc_lkrqs = 60 ltpx_port = 40042 } } Actual results: defaults to 15.0 seconds for heartbeat rate Expected results: integer values should work Additional info:
so far, a work around for this is to use floating point numbers instead of integer values when setting heartbeat_rate.
fixed in CVS Checking in config_main.c; /cvs/gulm/daemon/config_main.c,v <-- config_main.c new revision: 1.1.2.7; previous revision: 1.1.2.6 done
Cleanup. This bugzilla should be resolved with the current packages from RHN for Red Hat Global File System for Red Hat Enterprise Linux 3. GFS-6.0.2.36-9