Re: How To Backup All of CVS?

[Dennis Jones]

You need everything starting at $CVSROOT.  Note that $CVSROOT is NOT the
same thing as the "CVSROOT" directory in the repository.  "CVSROOT" is, in
fact, a sub-directory of $CVSROOT:

$CVSROOT (where $CVSROOT might be defined as, /vol/cvs [for example])
    \CVSROOT
    \ other repository directories

So, here is what I started doing recently:

In the superuser cronjob:

tar cfz repository.tgz  /vol/cvs/

Note that "/vol/cvs/" is equivelent to $CVSROOT on *MY* system, and $CVSROOT
might not be defined within the context of the cronjob, so I explcitly
specify path.  You will want to adjust your cronjob accordingly.

The TGZ file then contains all of the directories, files, and permissions
for everything in the repository.  Since I do not have a backup system on my
Linux box, I must move the TGZ file to a directory where it can be backed up
by an NT machine via a Samba share.  Of course, if you have a local backup
system on your Unix machine, then you could skip that step and backup the
file straightaway.

I hope this helps.

- Dennis

<address@hidden> wrote in message
news:address@hidden
> Hello all,
>
> I would like to backup all of CVS nightly using a cronjob.  I want to be
> able to restore CVS to its original state in case something happens to our
> server.  The problem is, I don't know exactly what I need to grab from CVS
> in order to be able to re-create it.  I believe I need the RCS files (the
> history files with the ,v at the end) but is there other stuff I also
need?
> How do I checkout or grab the RCS files?
>
> If anyone could provide me with some assistance or point me to relevant
> documentation I would greatly appreciate it.
>
> Thank you.
> Rebecca
>
>