Subject: Re: [vserver] test patch for 2.6.30.7
From: Natanael Copa <natanael.copa@gmail.com>
Date: Mon, 28 Sep 2009 14:00:23 +0200

I just found out that 2.6.30.8 is out.... sigh...

On Mon, 2009-09-28 at 12:51 +0200, Herbert Poetzl wrote:
> > > nope, that would be untested and that's not the way we
> > > do it .. besides anybody can ignore/fix a kernel version
> > > change easily ...
> 
> > It helps people who build precompiled packages for distros.
> 
> how would uploading untested patches help them?

1. It reduces number of modifications of changes needed in buildscripts.

2. It increases chances that more people will test. Not uploading means
only the package maintainer and that specific distro will test.

> > > 2.6.30.x is not long term supported (yet) and so we
> > > moved on to the 2.6.31 kernel, why not use that version
> > > instead?
> 
> > Because upgrading to 2.6.x releases will almost for sure break things
> > for users. Upgrading from 2.6.x.y will normally fix more things for
> > users than it breaks things.
> 
> well, it is correct that some 2.6.x releases are
> not the perfect thing, but without testing, there
> will be no 2.6.x.y release either :)

I will start test 2.6.31 stuff too soon, but in parallell i need to
provide something "stable". 2.6.30.x have had some testing time by now
and many issues are fixed. 

> > I don't switch from 2.6.30.y to 2.6.31.y just before a new distro
> > release, alpine 1.9.
> 
> that means, you are targeting (as debian usually does)
> a kernel which will be gone and forgotten in a few
> weeks from now ... interesting aproach ...

its hardly forgotten if debian (or other distro) is supporting it. I
don't intend to support it forever, just long enough so I have had time
to test the next one during a reasonable period.

> > >> It builds here atleast.
> > > that's good to know, does it boot and work too?
> > I haven't rolled it out for production yet but the testsript works:
> > .
> > alpine-netbook:~$ sudo bash testme.sh
> > Linux-VServer Test [V0.17] Copyright (C) 2003-2006 H.Poetzl
> > chcontext is working.
> > chbind is working.
> > Linux 2.6.30.7-vs2.3.0.36.14-pre8 #1 SMP Thu Sep 24 11:52:00 UTC 2009 i686
> > Ea 0.30.216 273/glibc (Sa) <v13,net,v21,v22,v23,netv2>
> > VCI: 0002:0304 273 13000fb1 (TbsPHIW)
> > ---
> > [000]# succeeded.
> > [001]# succeeded.
> > [011]# succeeded.
> > [031]# succeeded.
> > [101]# succeeded.
> > [102]# succeeded.
> > [201]# succeeded.
> > [202]# succeeded.
> 
> > Btw.... that glibc notice up there is wrong. It is uclibc.
> 
> what does 'vserver-info - SYSINFO' give?

alpine-netbook:~# vserver-info - SYSINFO
Versions:
                   Kernel: 2.6.30.7-vs2.3.0.36.14-pre8
                   VS-API: 0x00020304
             util-vserver: 0.30.216-pre2841; Sep 24 2009, 13:45:00

Features:
		       CC: ccache gcc, gcc (GCC) 4.4.1
                      CXX: ccache g++, g++ (GCC) 4.4.1
                 CPPFLAGS: '-march=i486 -Os -fomit-frame-pointer'
                   CFLAGS: '-march=i486 -Os -fomit-frame-pointer -std=c99 -Wall -pedantic
-W -funit-at-a-time'
                 CXXFLAGS: '-march=i486 -Os -fomit-frame-pointer -ansi -Wall -pedantic
-W -fmessage-length=0 -funit-at-a-time'
               build/host: i686-pc-linux-gnu/i686-pc-linux-gnu
             Use dietlibc: no (you have been warned)
       Build C++ programs: yes
       Build C99 programs: yes
           Available APIs: v13,net,v21,v22,v23,netv2
            ext2fs Source: e2fsprogs
    syscall(2) invocation: alternative
      vserver(2) syscall#: 273/glibc
               crypto api: beecrypt
          python bindings: no
   use library versioning: yes

Paths:
                   prefix: /usr
        sysconf-Directory: /etc
            cfg-Directory: /etc/vservers
         initrd-Directory: $(sysconfdir)/init.d
       pkgstate-Directory: /var/run/vservers
          vserver-Rootdir: /vservers


-nc