[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: crontab use in chroot 32-bit environment ?

More specifically, here is my problem:

Under 32-bit chroot environment, I issue the command to backup my system(s) using amanda:

/usr/bin/amdump new

and all works as expected.

Under the 64-bit environment, I issue the following command

dchroot -c ia32 /usr/bin/amdump new

However, the command fails and the system hangs and behaves in exactly the same manner that it would fail as if the command were run in 64-bit native mode.  Note, I have submitted a bug report to this effect under the amanda-server package.  What I am trying to do, in effect, is a work around to this bug.

What I would like to do is set this up so that I do not have to run amdump interactively in order to work, i.e. crontab from the 64-bit environment.

I suspect that amdump must be subsequently calling something from the 64-bit; i.e., by entering dchroot, do all subsequent calls to executables remain in the chroot environment ?

On Wed, 2005-11-09 at 11:47 -0500, Lennart Sorensen wrote:
On Wed, Nov 09, 2005 at 11:02:13AM -0500, James D. Freels wrote:
> I set up a crontab for user backup (to run amanda) in the chroot 32-bit
> environment.  It does not seem to run.  Is it a requirement to run
> crontab from the 64-bit home environment and then include the 32-bit
> instructions within that crontab  (from root) ?  There are some bugs in
> the ananda-server package that require it to run in 32-bit mode.  I am
> getting ready to submit a bug report about this.

Normally there are no daemons running in the chroot, since well the init
scripts are never run for the chroot.

You can certainly add things to cron in the main system that runs chroot
things.  Something like this might work:

1 2 * * * dchroot -options command

With appropriate dchroot setup and options for running the command in
the chroot.

The crond doesn't know about the chroot, and the crond in the chroot
isn't running, so your system is behaving normally.

Len Sorensen
James D. Freels, Ph.D.
Oak Ridge National Laboratory

Reply to: