A customer is witnessing this issue in RHEL 5 as well. +++ This bug was initially created as a clone of Bug #455540 +++ Description of problem: It looks like this is was likely addressed in bug 426382, however there is no fix incorporated into RH 4 (and presumably 5, though I have not verified). If named is stopped via rndc, it won't exit immediately and as such the 'start' of a 'restart' will be attempted before the prior instance had completed shutdown. Version-Release number of selected component (if applicable): bind-9.2.4-28.0.1.el4 How reproducible: Get a reasonably loaded bind instance going (~200mb of heap in this case) and execute a restart via the stock init script. Most of the time the start will fail due to the daemon still exiting. A fix as trivial as: 128a129,131 > while pidofproc named >/dev/null 2>&1; do > sleep 1 > done 158,159d160 < # wait a couple of seconds for the named to finish closing down < sleep 2 ...resolves the issue. --- Additional comment from atkac on 2008-07-16 04:52:56 EDT --- Right you are. Thanks for your report --- Additional comment from pm-rhel on 2008-10-31 12:38:28 EDT --- This request was evaluated by Red Hat Product Management for inclusion, but this component is not scheduled to be updated in the current Red Hat Enterprise Linux release. If you would like this request to be reviewed for the next minor release, ask your support representative to set the next rhel-x.y flag to "?". --- Additional comment from kgraham on 2009-01-30 16:32:26 EDT --- re comment 2 -- Applying RHSA-2009:0020 resulted in yet another service outage due to this bug. Could Red Hat Product Management perhaps re-consider the implications of this 3 line patch?