Memory Leak in Bind 9?

Mark Andrews Mark_Andrews at isc.org
Fri Dec 8 23:57:55 UTC 2006


> Hi Mark,
> 
> We did do that on one of our name servers .. I even tried 9.3.3rc3. I  
> don't quite understand why this would start suddenly.
> 
> Thanks,
> 
> --Wes

	Because your usage pattern changed.  You can be running at
	just under the per-process memory limit for years then a
	slight change of usage will cause you to hit the per-process
	memory limit.

	Named has controls for raising the per-process memory limit.
	It also has controls for releasing memory earlier than it
	otherwise would.  You should look at these and set them
	appropiately in named.conf.

	Mark
 
> On Dec 8, 2006, at 7:39 AM, Mark Watts wrote:
> 
> >> After using Bind for years with the same relative configuration we
> >> started to see named stop responding.
> >>
> >> Looking at the logs I see this:
> >>
> >> Dec 08 01:28:53.584 general: error: timer.c:603: unexpected error:
> >> Dec 08 01:28:53.585 general: error: couldn't allocate event
> >>
> >> this is bind 9.2.2 on FreeBSD 4.4. I do see some other posts for the
> >> same issue on this list without and answer for FreeBSD 5.5.
> >>
> >> As I said this just suddenly started occurring. We have a cronjob to
> >> restart named twice a day to keep it living.
> >>
> >> Anyone have any ideas?
> >>
> >> Thanks,
> >>
> >> --Wes
> >
> > Upgrade to BIND 9.3.2-P2 and try again.
> >
> > Mark.
> >
> > -- 
> > Mark Watts BSc RHCE MBCS
> > Senior Systems Engineer
> > QinetiQ Trusted Information Management
> > Trusted Solutions and Services Group
> > GPG Public Key ID: 455420ED
> >
> >
> >
> >
> 
> 
-- 
Mark Andrews, ISC
1 Seymour St., Dundas Valley, NSW 2117, Australia
PHONE: +61 2 9871 4742                 INTERNET: Mark_Andrews at isc.org



More information about the bind-users mailing list