|
Roberto Nibali <ratz@xxxxxx> wrote:
> Hi,
>
> This is my next version and this time I've also implemented the session
> pool, based on an idea and POC implementation forged in a lobby
> somewhere in the Ottawa Convention centre 3 years ago by Horms and me.
> It's against 2.4.x because this is our strategic kernel, but 2.6.x
> should be easy to adapt.
>
> I've run an intesive 72 hour test on an SMP kernel with this patchset
> and it looks really stable and functionally correct. We will definitely
> include this into our kernel and ship it productively.
>
> What's missing? Some schedulers have not been updated yet to cope with
> the session overflow pool semantics, however this is not so tragic. The
> patch otherwise consists of:
>
> o Destination threshold limitation implementation
> o Spillover or session overflow pool
> o Upgraded schedulers: wrr, wlc, sed, rr, lc
> o New schedulers : hprio
> o Some cleanup in the IPVS code generally
>
> The patch is incorrect regarding:
>
> o the changes in the sh and dh schedulers, I'll revert them
> o two hunks have been merged upstream in the newly released kernel
>
> Other than that, I wish someone could comment on it and give some
> feedback. Oh, and of course you need the upgraded ipvsadm source. This
> makes the following patchset:
>
> [1/2]: the kernel part
> [2/2]: the ipvsadm enhanced with the needed functionality
Hi Rats,
I read over your patches a couple of times and I couldn't find anything
terribly disturbing in them. Here are some more general comments,
mostly minor.
* 80 columns wide
(there seems to be one fragment that ever so slightly overflows)
* Can you break out the functionally different parts of the patches
into separate patches, especially the hprio scheduler and the cleanups.
* For upstream I think this will have to be a 2.6 patch
* Seems like I can use your framework to add my quiescent/offline flags
idea. Perhaps ip_vs_overflow should be ip_vs_available or something
like that. Not a big deal, we can always change it later.
* As far as I can see your code won't break binary compatibility with
ipvsadm. But its not much use without an updated ipvsadm.
Wensong, are you processing patches for ipvsadm or should I ?
Lastly, not strictly related, this mail is the last mail
for an unreviewed patch. If I've missed one, please let me know.
Responding to the orignial thread with me CCed should grab my attention.
--
Horms
|
