>>> Graeme Fowler <graeme@xxxxxxxxxxx> schrieb am 26.02.2007 um 15:55 in
> On 26/02/2007 13:31, Stephan Hendl wrote:
>>> Are you using persistence?
>> the default lvs parameters only
> There aren't any LVS defaults with regards to virtual/realserver
> settings. I suppose this would be a good time to ask for the output from
> "ipvsadm -L -n".
[root ~]# ipvsadm -L -n
IP Virtual Server version 1.2.0 (size=4096)
Prot LocalAddress:Port Scheduler Flags
-> RemoteAddress:Port Forward Weight ActiveConn InActConn
TCP <VIP>:80 wlc
-> 192.168.232.212:80 Masq 5 0 0
-> 192.168.232.211:80 Masq 2 0 0
-> 192.168.232.214:80 Masq 10 0 0
-> 192.168.232.213:80 Masq 10 0 0
>>> What tool are you using to run the test?
>> http_load (http://www.acme.com/jef/)
>> root# ./http_load -parallel 200 -seconds 60 ../ism_baseline.txt
> OK, so you're simulating 200 simultaneous connections.
>>> Do you get any errors in your Apache logs (assuming Apache server)?
>> no error messages at all
> But then you say...
sorry, after the first question I tested a bit more with two servers instead of
one in order to see how the system works with more than one test-client
>> no, with two server at the same time I got the apache error:
>> [Mon Feb 26 12:47:27 2007] [error] server reached MaxClients setting,
> consider raising the MaxClients setting
> Not entirely surprising. Apache defaults (if I recall) to a MaxClients
> setting of either 100 or 200, which you are going to exceed with your test.
> Make MaxClients something like 500 and run the tests again. It looks
> like your problems aren't LVS related, but application tuning is needed
I'll try it.
> LinuxVirtualServer.org mailing list - lvs-users@xxxxxxxxxxxxxxxxxxxxxx
> Send requests to lvs-users-request@xxxxxxxxxxxxxxxxxxxxxx
> or go to http://www.in-addr.de/mailman/listinfo/lvs-users