Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
On Wed, Feb 16, 2005 at 03:13:01PM +1100, Mark Andrews wrote:
> 1808.	[bug]		zone.c:notify_zone() contained a race condition,
> 			zone->db could change underneath it.  [RT #13511]
As i have it this change would not be something that could trigger high user
cpu usage when running threaded, right?
On my resolver-Cluster i am mainly running current bind8 versions but i
thought i would give bind9 a try again as you are clearly planning on ending
security bugfix support for it in the long term.
Please see attached image for cpu usage. The query-load was stable and
underlying a natural (dsl) usage curve throughout the whole time.
Sustained peak query rate was 600-700 qps, minimum 130 qps. Graphs on reque=
st.
I only switched bind versions; kernel version, uptime, swapping behaviour
was/has not changed in any way.
The image shows cpu usage change when i switched the machine from _two_!
bind 8.4.6-REL processes to _one_ bind 9.2.5beta2-threaded.
The machine is running Linux 2.6.11-rc2 debian/unstable with libc6-i686 on a
Compaq Proliant DL360 G1 (i.e. 2xPIII 800MHz 1GB RAM).

Btw.: No Problems so far with 8.4.6-REL.

Dear List, what are you using for/on your recursive Nameservers?

best regards,
		Stefan Schmidt
--=20
"As we know, There are known knowns. There are things we know we know. We a=
lso
know there are known unknowns.  That is to say we know there are some thing=
s we
do not know. But there are also unknown unknowns, the ones we don't know we
don't know."
- Donald Rumsfeld


-- Binary/unsupported file stripped by Ecartis --
-- Type: image/png
-- File: was-zum-frell.png



-- Attached file included as plaintext by Ecartis --
-- File: signature.asc
-- Desc: Digital signature

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.0 (GNU/Linux)

iD8DBQFCFLg4wCTKSDyz3swRAkpKAKDGn0XDDR9DlFItJLYqvJe533dm4ACeLAql
uryO2BifzKdGMt9aPLcbpTs=
=OMJg
-----END PGP SIGNATURE-----