Hi Jason, Ref: http://www.austintek.com/LVS/LVS-HOWTO/HOWTO/LVS-HOWTO.LVS-DR.html#director_as_client_in_LVS-DR Jason Stubbs a écrit : >>>> Reading the "7.10. Director as client in LVS-DR" >>>> >>> Yes, the director cannot access the LVS service with an unpatched kernel. >>> With the patch mentioned in the FAQ, your configuration should work fine >>> as far as I can see. >>> >> The patch seems pretty old, is it still "patchable" ? >> any kernel compatibly I should watch? >> >> The patch is pretty simple, I could probably correct the code also... >> > > It won't apply directly as the code has changed slightly. As for > compatibility > and what not, you're pretty much on your own... I'm testing that patch and a > few other changes at the moment but I haven't put it in production yet or > stress-tested. It might OOPS after several hours of running for all I know. > What does it means "OOPS" ? > Having said that, there's no other way to do what you want at the moment. If > you really want/need to do it, patching and testing is about all you can do. > I really need to loadbalance squid "output" and detect failure on realserver backend. Having playing with LVS showed that this way seem interesting to test. I can also test squid internal loadbalancing, but the pound are pretty bugged for the configuration I've tested. That will be my next path.
Thank for the links. Regards, Sylvain. _______________________________________________ LinuxVirtualServer.org mailing list - [email protected] Send requests to [EMAIL PROTECTED] or go to http://lists.graemef.net/mailman/listinfo/lvs-users
