Red Hat Bugzilla – Bug 113747
/var/lock/subsys/irqbalance (possibly) improperly handled
Last modified: 2007-11-30 17:07:00 EST
From Bugzilla Helper:
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.4) Gecko/20030922
Description of problem:
/var/lock/subsys/irqbalance is not created by the init script, so it
is never stopped but is always spawned, resulting in multiple
instances. eg: 'telinit 1; telinit 5'. you now have two.
the script doesn't create the lockfile when *not* in 'oneshot' mode,
which i believe is incorrect. irregardless, the handling is such that
multiple instances are created...
Version-Release number of selected component (if applicable):
Steps to Reproduce:
1. telinit 1
2. telinit 5
Actual Results: multiple instances of irqbalance
Expected Results: /var/lock/subsys/irqbalance should be created when
not in 'oneshot' mode so that /etc/rc.d/rc will actually do a
'/etc/init.d/irqbalance stop' (rather than ignoring it and continuing)
Yea, this is annoying
--- irqbalance.orig 2004-07-18 12:05:55.000000000 -0700
+++ irqbalance 2004-07-18 12:06:02.000000000 -0700
@@ -47,7 +47,7 @@
daemon irqbalance $ONESHOT
- [ $RETVAL -eq 0 -a -n "$ONESHOT" ] && touch
+ [ $RETVAL -eq 0 -a -z "$ONESHOT" ] && touch
/var/lock/subsys/irqbalance return $RETVAL
True if the length of string is zero.
string True if the length of string is non-zero.
aim for U3 ?
With RHEL3 in critical-bugfix-only maintenance mode, there will be no further
major updates, so this bug is a WONTFIX.
(In reply to comment #2)
> With RHEL3 in critical-bugfix-only maintenance mode, there will be no further
> major updates, so this bug is a WONTFIX.
? ? ?
after three and a half years and seven updates, no-one can make a one character
? ? ?
Unfortunately, sometimes things slip through the cracks. For three and a half years... :\
All I can say is we're working on doing better going forward, and we did manage to fix it for RHEL4 and 5.