Hi folks, I care about it as well. IMQ is great and have to be maintained, but myself I'm not strong with programming. But I can help with testing. If somebody can take care about IMQ (or new project of it) would be very nice. I prefer to have everything on one box - traffic shaping (ingress and outgress), NAT and firewall. So guys can anybody can maintain/update IMQ (specially for 2.6 kernel), please? Thanks Remus ----- Original Message ----- From: "Andre Correa" <andre.correa@xxxxxxxxx> To: <peter@xxxxxxxxxxxxxxxxxxxxxxx> Cc: "Netfilter Mailing List" <netfilter@xxxxxxxxxxxxxxxxxxx> Sent: Monday, January 19, 2004 1:33 PM Subject: Re: IMQ Replacement > > I've being using IMQ with 2.4.23 here with no probs too. I've already > tried to discuss this topic here and in other lists but got no answer. > Writting to the last maintainner didn't helped much... > > Some time ago there was a posting, maybe in LARTC, about a guy that > started a parallell IMQ project, but I've never heard about hin anymore... > > I would like to contribute/coordinate in an effort to put IMQ back on > its track. It is an important feature that is being left behind. > > Peter, maybe we should join, put a new site up with documentation > updates, maintain 2.4.x patchs and later think what to do with 2.6. We > may find other people interrested in coding and helping... > > Anybody else care about it? > > let us know... > > cheers > > Andre > > > > Peter Frischknecht wrote: > > I have been using IMQ with great success for the last few months. The > > particular network where this is used has 400+ nodes, and IMQ performs > > flawlessly. > > > > As anyone who uses IMQ will tell you, the only alternatives are: > > 1 - Install another router, just to perform NAT > > 2 - Install 2 additional NICs in the Linux gateway with a Xover cable > > between them > > > > Is there another piece of software that performs the job of IMQ? > > It is great being able to shape traffic AND masquerade packets in one > > box. How else can it be done? > > > > We shape traffic individually for each client. Generic traffic shapping > > rules that regulate the entire pipe would not work for us. > > > > The fact that IMQ is unmaintained is nerve wrecking. It is difficult > > enough to install the thing with the sources and documentation > > available. It will only become more difficult as the kernel evolves and > > we move to 2.6.x. > > > > I don't think I have the knowledge to mainatin the source code. But I > > believe in the concept so much that I might just take up the code > > maintenance. Hoepefully some other folks will join in. Maybe we can > > convince enough folks to incorporate it into 2.6.x. > > > > Any help, comments or even flames will be appreciated. It is time for a > > solution to this problem. > > > > Peter > > > > > > > >
