|
Hello,
On Fri, 5 Jul 2002, Damien CLERMONTE wrote:
> > FWMARK-based VS:
> > Client_Net -> FWMARK -> RS
>
> And this is a problem with lblc, the dest ip is not checked, thus persistence
> is wrong.
>
> If I fwmark 10.0.0.0/8, and the client connect to 10.0.0.1, I expect the next
> connection
> to 10.0.0.1 to go to the same server, and it works.
>
> But if the same client connect to 10.0.0.2 just after that, I expect it to be
> load
> balanced to another server (by not finding it in persistence cache, and
> forwarding it to
> lblc scheduler)
"load balanced to another server" is performed from lblc*,
if the persistence does not recommend RS
> It does without persistence, it doesn't with persistence as both client_net
> and fwmark are
> the same and matched with cache. Thats why persistence need to also check VIP
> somewhere,
> at least for lblc*
Hm, why you need persistence when LBLC* are used? Can
you explain why LBLC* without persistence does not work for you?
The proposed from you persistence can return RS that is already
scheduled from LBLC, so it works as a L2 cache considering the
LBLC as L1 cache. In this case your persistence will return
the same RS that LBLC* can return. LBLC* guarantee that for one
target the same RS is used. How you plan to use the both features,
I still can't see the trick? It seems you are trying to replace
LBLC* with the proposed persistence and to avoid using LBLC*.
Because 0/0->TARGET_VIP->RS is similar to LBLC*. Note that there
are other schedulers that you can consider.
> Regards,
>
> Damien
Regards
--
Julian Anastasov <ja@xxxxxx>
|
