|
Mark Weaver wrote:
>
> Software:
> debian woody, stock 2.4.24 kernel + frees/wan.
does freeswan involve a lot of overhead?
> Hardware:
> AMD XP2400+
> VT8377 [KT400] MB
> eth0=via rhine II (VT6102, on board)
> eth1=3c950C
> 1Gb memory
>
> LVS Configuration:
> LVS-NAT + firewall mark setup. Ports 1755/tcp, 554/tcp, 554/udp,
> 1024:5000/udp are grouped together. I realise LVS-NAT is not ideal here
LVS-NAT was rewritten for 2.4.x and doesn't have the overhead of the
earlier implementations, so it's shouldn't be a problem.
> but the director is also the gateway, so it seemed like the simplest way to
> proceed initially. I then have:
>
> ipvsadm -D -f 1
> ipvsadm -A -f 1 -s rr -p 600
> ipvsadm -a -f 1 -r 10.90.90.7:0
>
> i.e. single media server and using persistence to keep the grouping. This
> all works fine (i.e. I can stream media).
>
> I then load test this using a setup like:
>
> eth1
> test box --- director --- media server
> | |
> ------------- eth0
> |
> outside world via some cisco router or other
>
> so the test box and the director are on the same LAN (connected via some
> cisco switch), and the media server and the director are on a different LAN
> (connected via a cheap D-Link switch).
>
> The test box is using the Windows Media Load Simulator. This just makes a
> lot of connections and streams back data. The average stream only gets up
> to about 35Mbit. At this point, CPU usage on the director is ~20% (which
> would seem to indicate that I should be able to get a lot more out of it).
if you're using top, it doesn't show all the kernel activity, but I'm not sure
what the story is there. what is the load average? If you have an X-window on
the
director, does the mouse respond sluggishly? Does it take a real long time
to load and run a trivial script (eg #!/bin/bash; echo "foo")?
> CPU on the test box is at about 25% and on the media server at 4%.
>
> The problematic part is that the director begins dropping about 10% of
> externally originated packets at this level of load. I wouldn't say any
> machine involved is stressed here, but pinging the external IP of the
> director gives that huge loss. This noticeably affects say, SSH, on the
> director or TS to the media server. This is constrasted with pinging the
> external IP of the test box, which gives 0% loss.
:-(
> I would therefore conclude that this is an issue with the director, but I'm
> not sure what. My next guess would be to try swapping the VIA NIC for
> another 3com one, but could it really be that bad?
it's a cheap test.
I'm surprised to hear this. I don't have any suggestions sorry
Joe
--
Joseph Mack PhD, High Performance Computing & Scientific Visualization
SAIC, Supporting the EPA Research Triangle Park, NC 919-541-0007
Federal Contact - John B. Smith 919-541-1087 - smith.johnb@xxxxxxx
|
