|
Hello,
On Tue, 16 Jul 2002, Peter Mueller wrote:
> I am trying to use testlvs to test throughput and perform a "burn-in" of the
> software. It seems I have succeeded (it is stable) but I can effectively
> DOS the machine during the test.
Yes, this is expected, see my tests in the HOWTO:
http://www.linuxvirtualserver.org/Joseph.Mack/HOWTO/LVS-HOWTO-11.html#ss11.14
> director = 2.4.19-rc1 + [base redhat .config] + ipvs patch + julian's hidden
> patch + heartbeat + ldirectord. 2-700mhz P3 & 512mb of ram, 7200rpm scsi
> disk, etc. (valinux 2230). << I am using 2.4.19-rc1 because I have eepro
> NICs >>.
One of the most meaningful parameters: the PCI bus speed?
> real server = 2.2.x, apache, 2 proc, 512mb, requests set to reject according
> to README file. specifically, srcnet is changed to 10.10.0.1 and then
> make). I issue the command "route add -net 10.10.0.0 netmask 255.255.0.0
> reject" to drop/reject the packets from LVS-director during test.
You can even use ipchains just to drop the flood. The
route reject command wastes CPU in the RS.
> client/test box = 2.2.x, 2 proc, 512mb box. all three boxes are essentially
> the same hardware. the command used is ./testlvs 10.0.1.10:80 -tcp -srcnum
> 62500 -packets 100000000
>
> my actual question -- show_traffic.sh on the real server shows ~20-25k
> packets/sec. this seems small to cause CPU to be 100% utilized. what am I
> doing wrong?
May be you have PCI 33MHz, ipchains rules, etc. The other
problem could be the routing cache performance. You can reduce the
number of clients to 10 just to see what happens.
Regards
--
Julian Anastasov <ja@xxxxxx>
|
