|
Hello,
> I'm using piranha which provides ipvsadm 0.8.1 (hum, quite old,
> might be
> the problem).
What is the kernel version? Show us the 'ipvsadm -Ln' output please.
> Using the director with a high weight, and 2 realservers with low
> weights does not lead to what I was expecting. The director is not
> used as a realserver as often as it should.
>
> In my setup, I have the following weights: director 8 realserver1 1
> realserver2 1
>
> Out of 100 http requests, no more than one gets directed to the
> director, which is less than 1%, whereas it should get 80% of the
> traffic.
Can you provide me with an output of 'ipvsadm -Ln' after 100 http
requests? So I can see if you use persistency, what scheduler you use
and if there is really a bug in the LVS scheduler (which I don't think)
> Was there something wrong in this release than has since been fixed ?
Well, you have to provide us with the version of LVS and kernel. The
rest is pretty irrelevant to your problem.
> My other problem is still using the main director when the backup
> takes over. Supposing the backup redirects a connection to the
> director's web server (using the NATed network), how will this box
> be able to answer whereas it does not have the correct gateway
> (which is set to the external network, not the NATed one).
You use LVS_NAT for your RS and LVS_DR for your director maybe?
> Having 127.0.0.1 in the conf file instead of the director internal
> IP is not a better choice, because the director and the backup conf
> files would have to be different :(
I don't know about piranha, so I cannot help you here. The piranha
mailing list is best suited for that. But for your problem above I would
like to see the output of the things I requested. Like that we can
decide if there is really a problem with a specific scheduler or LVS.
Best regards,
Roberto Nibali, ratz
|
