Hi,

Thanks for the replies, I'll look at the statistics module (but I think I 
already tried and it wasn't there either) - actually, I'm
not coming from any iptables version, I just read the manual available on the 
machines and attempted to use what's in that manual...
Clusterip is in fact present, but using it will require more work because of 
our node management system which might not be very
flexible, it's why I was looking for more basic modules ;)

Also, I find it strange that the balance module wouldn't be in the sources but 
that the iptables command would allow to "use" it ?

I forgot to say I tried to test using an ip range with the DNAT target : the 
manual says that if working, this would be the
equivalent of a simple round robin... but in fact, my tests showed that this 
just takes the first IP of the range, the others (I
tried 2 IP) are never used... too bad too.

Regards

> -----Message d'origine-----
> De : owner-scientific-linux-us...@listserv.fnal.gov 
> [mailto:owner-scientific-linux-us...@listserv.fnal.gov] De la part de Tom H
> Envoyé : vendredi 27 janvier 2012 21:45
> À : SL Users
> Objet : Re: missing iptables modules and filters in all SL distributions
> 
> On Fri, Jan 27, 2012 at 3:06 PM, Tom H <tomh0...@gmail.com> wrote:
> > On Fri, Jan 27, 2012 at 2:45 PM, Orion Poplawski <or...@cora.nwra.com> 
> > wrote:
> >> On 01/27/2012 10:22 AM, SCHAER Frederic wrote:
> >>>
> >>> Looking for simple round robin solutions, I tried several times today to
> >>> use
> >>> iptables functions without success.
> >>>
> >>> The cause is that some targets and some matching filters are not available
> >>> as
> >>> modules and not compiled in the kernel.
> >>>
> >>> For instance, the BALANCE target isn’t available. Too bad.
> >>>
> >>> Then, the “-nth” or“-random” parameters aren’t available neither.
> >>
> >> I think you must be coming from a very old version of iptables.
> >
> > +1
> >
> > For “-nth” or“-random”, take a look at "statistic" in "man iptables".
> 
> Replied too quickly.
> 
> I googled for "balance target" because I've never seen it before.
> "clusterip" might be the replacement but since I've never used either,
> I'm not sure...

Attachment: smime.p7s
Description: S/MIME cryptographic signature

Reply via email to