Red Hat Bugzilla – Bug 78855
/etc/init.d/nscd reload actually kills nscd totally
Last modified: 2016-11-24 10:01:08 EST
From Bugzilla Helper: User-Agent: Mozilla/4.0 (compatible; MSIE 5.5; Windows NT 5.0; T312461) Description of problem: When /etc/init.d/nscd reload is given, nscd is killed totally and abnormally, instead of reloading the config? This happens on both RH7.3 and RH7.1 using nscd-2.2.5-42 and nscd-2.2.4-31 respectively. Version-Release number of selected component (if applicable): How reproducible: Always Steps to Reproduce: 1. ps -auxw | grep nscd to confirm nscd is running 2. /etc/init.d/nscd reload 3. ps -auxw | grep nscd to find no nscd running 4. /etc/init.d/nscd status Actual Results: 3. No nscd process/thread present 4. nscd dead but pid file exists Expected Results: 2. /etc/init.d/nscd reload sends a -HUP to the NSCD process which should alert it to close the log file, re read the /etc/nscd.conf and go on running. 3. nscd should still be running. 4. should say something like: nscd (pid 7620 7619 7618 7617 7616 7615 7614 7613 7610) is running... Additional info: System armed with latest Errata RPMs. This maybe related to the other nscd bugs where nscd crashes or dies abnormally, leaving behind /var/run/nscd.pid nd /var/run/.nscd_socket.
This might have been the case in that old release. Nobody checked since this bug went unnoticed since the component was nscd not glibc (nscd was never looked at). The current nscd is much more robust. For best results use the code in the upcoming FC3 release of from rawhide today (2.3.3-65 and up). I'm closing this bug since it's outdated.