On Mon, Jan 30, 2006 at 09:49:48AM +1000, Anton Umnikov wrote: > On Sun, 29 Jan 2006 09:37:14 +0100 > Martin Koniczek <[EMAIL PROTECTED]> wrote: > > Mike O'Connor wrote: > > > So which patch is safe for a 2.6.15 kernel and to the best of knowledge > > > works correctly ? > > > > For this cases, Herbert provided a special Patch Version for 2.1.0.5 > > (just a minor cleanup and the bug mentioned in this thread fixed) > > > > http://vserver.13thfloor.at/Experimental/patch-2.6.15-vs2.1.0.5.1.diff > > > > (the front page http://linux-vserver.org/ got updated accordingly) > > > > It works nice for my test scenarios at least, but it is of course still > > a patch of the "experimental" tree, and becomes outdated quickly. > > > > > I ask this because 2.6.15 is the only stable kernel which as support for > > > ATA path thought for SATA, which is important for server reliability. > > > > > > I suppose the best thing would be someone with more skill then me to > > > take a stable vs2.0 patch and get it working with the current stable > > kernel. > > > > http://vserver.13thfloor.at/Experimental/patch-2.6.15-vs2.0.1.2.diff > > is the latest patch of the stable version and is reported to work?
> Yes, it's work, but with some strange events/ For example when I'd > made piix as module, all PATA disks runned very slow. I'm not sure > that its vserver related issue. But anyway, it's run: you probably have to enable DMA and 32bit mode when you load it as module with something like: hdparm -c 1 -d 1 -m 16 -u 1 /dev/hd? > dory:backups# dory:vservers# uname -a > Linux miskey.ic.dvgu.ru 2.6.15.1-vs2.0.1.2 #6 SMP PREEMPT Thu Jan 26 16:51:24 > VLAT 2006 i686 i686 i386 GNU/Linux > miskey:vservers# vst > CTX PROC VSZ RSS userTIME sysTIME UPTIME NAME > 0 60 114.3M 37M 35m43s31 42m18s16 2d23h36 root server > 49153 4 8.3M 3.9M 3m57s25 5m36s20 2d23h36 log > 49154 11 214.6M 78.3M 20m24s50 2m34s00 2d23h36 tower > 49155 3 5.4M 2.1M 5m42s78 4m36s93 2d23h36 arc > 49163 4 8.3M 3.5M 0m00s64 0m00s88 1d18h36 castle sidenote: better start using static context, the dynamic ones will go away sooner or later ... best, Herbert > > cu, > > cohan > > > > > > _______________________________________________ > > Vserver mailing list > > [email protected] > > http://list.linux-vserver.org/mailman/listinfo/vserver > > > _______________________________________________ > Vserver mailing list > [email protected] > http://list.linux-vserver.org/mailman/listinfo/vserver _______________________________________________ Vserver mailing list [email protected] http://list.linux-vserver.org/mailman/listinfo/vserver
