Subject: Re: [vserver] Experimental Branch usable in production?
From: Herbert Poetzl <herbert@13thfloor.at>
Date: Thu, 18 Jun 2009 19:57:27 +0200

On Thu, Jun 18, 2009 at 07:29:45AM -0400, John A. Sullivan III wrote:
> On Thu, 2009-06-18 at 09:50 +0200, randall wrote:
> > Oliver Welter wrote:
> > > -----BEGIN PGP SIGNED MESSAGE-----
> > > Hash: SHA1
> > >
> > > Hi Folks,
> > >
> > > I am using vserver now for around 5 years and the stability is
> > > awesome - thanks folks!
> > >
> > > I am currently running 2.6.22 with vs 2.2.0.7 but due to some
> > > enhancements in the recent kernels I really want to go up to a more
> > > recent one (at least 2.6.26). Anybody here running the experimental
> > > stuff already in production systems?
> > >
> > > regards
> > >
> > > Oliver
> > > - --
> > > Protect your environment -  close windows and adopt a penguin!
> > > PGP-Key: 3B2C 8095 A7DF 8BB5 2CFF  8168 CAB7 B0DD 3985 1721
> > > -----BEGIN PGP SIGNATURE-----
> > > Version: GnuPG v1.4.9 (GNU/Linux)
> > > Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
> > >
> > > iEYEARECAAYFAko54OsACgkQyrew3TmFFyHszACfcsHVvbM32Tc91p1zQvmkrqp8
> > > tukAniNVvBsCOAA+Ckiq0Uxqx45Fjxyh
> > > =bkIa
> > > -----END PGP SIGNATURE-----
> > >   
> > hi there,
> > 
> > i'm currently using the stock Debian Lenny version and have not 
> > experienced any stability or other issues the last few months.
> > 
> > 
> > 2.6.26+17      2.3.0.35
> > util-vserver: 0.30.216-pre2772;
> > 
> > there seem to be some issues mentioned on the wiki with this kernel, not 
> > sure if its debian specific.
> > 
> > http://linux-vserver.org/Installation_on_Debian#Issues_with_the_current_2.6.26_Kernel
> > 
> There are some issues which more experienced folks than I pointed out to
> me on this list.  We are required to use newer kernels (I believe >=.26)
> because of a serious bug in iSCSI in earlier kernels.  However, in
> both .27 and .28, we have picked up an extremely serious bug that has me
> eagerly awaiting stable vserver for .29.  

well, time to start contributing (time or money), otherwise
the testing required for a 'stable' release will not happen
in the near future ...

> Every once and a while, a process sends the CPU into some kind of
> loop. The process cannot be killed. The only option is a reboot of the
> vserver host, not the individual vserver.

which kernels did you test with? what kernels did show this
behaviour? and what messages ended up in the kenel log?

> We saw this regularly with the OSSEC rootkit product but it is not
> specific to OSSEC; we did see one other non-OSSEC

a test case would be good to start with!

best,
Herbert

> instance - John
> -- 
> John A. Sullivan III
> Open Source Development Corporation
> +1 207-985-7880
> jsullivan@opensourcedevel.com
> 
> http://www.spiritualoutreach.com
> Making Christianity intelligible to secular society