Re: [lxc-devel] use defined rootfs mount point regression?

2010-05-26 Thread Ferenc Wagner
Daniel Lezcano daniel.lezc...@free.fr writes:

 On 05/21/2010 02:20 PM, Nathan Lynch wrote:

 On Fri, 2010-05-21 at 09:56 +0200, Daniel Lezcano wrote:

 On 05/20/2010 10:40 PM, Nathan Lynch wrote:
  
 lxc-execute: No such file or directory - failed to access to 
 '/usr/lib64/lxc', check it is present
 lxc-execute: failed to set rootfs for 'truetest-19794'
 lxc-execute: failed to setup the container

 /usr/lib64/lxc does not exist on the host.  Is this the intended
 behavior?

 Yes, you have to create it. I expect the distro maintainers to update
 their package %post_install section to create the directory.

 Not to bikeshed this further, but /usr/$libdir/lxc is probably not a
 good default value.  FHS says applications may use only a single
 subdirectory in /usr/lib, and I think it was pointed out on the earlier
 thread that Debian and related distros already place lxc components in
 this directory.

 If I refer to http://en.wikipedia.org/wiki/Filesystem_Hierarchy_Standard

:) If you dared to mention the FHS: it does not know such things as
libexec.  So my recommendation is still ditching /usr/libexec in favor
of /usr/lib/lxc/lxc-init and /usr/lib/lxc/rootfs.  To quote the FHS 2.3:

   /usr/lib includes object files, libraries, and internal binaries that
   are not intended to be executed directly by users or shell scripts.

   Applications may use a single subdirectory under /usr/lib. If an
   application uses a subdirectory, all architecture-dependent data
   exclusively used by the application must be placed within that
   subdirectory.

I see that RedHat is somewhat committed to libexec and the concept
indeed makes sense to me.  So you'll probably want to keep it, and other
distros will continue coalescing them via configure options.  So you can
use $libdir/lxc or even $localstatedir/lxc (ie. /var/lib/lxc), since it
isn't a problem if some lxc internal stuff is temporarily shadowed
during container setup.

 /var/tmp/lxc could be a good candidate no ?

I checked some Debian systems around here, /var/tmp is generally empty,
not a single package owns a directory there.  But technically it should
work.  I wouldn't call this mount point itself temporary, though.
-- 
Cheers,
Feri.

--

___
Lxc-devel mailing list
Lxc-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/lxc-devel


Re: [lxc-devel] lxc-unshare woes and signal forwarding in lxc-start

2010-05-26 Thread Ferenc Wagner
Daniel Lezcano daniel.lezc...@free.fr writes:

 On 05/13/2010 02:22 PM, Ferenc Wagner wrote:

 I attached a proof-of-concept patch which seems to work good enough for
 me.  The function names are somewhat off now, but I leave that for later

 do you have definitive version for this ?
 I have some modifications in the start function and they may conflict
 with your patch.

No, I didn't work on this any further.  I'm not sure about the realtime
signals... maybe it would be best to reverse the logic and use
sigfillset and sigdelset instead?
-- 
Cheers,
Feri.

--

___
Lxc-devel mailing list
Lxc-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/lxc-devel