This is the mail archive of the
mailing list for the Cygwin project.
Re: cygwin to backup Windows data
Thanks for the feedback.
At 06:34 PM 8/4/2003 -0700, Brian Dessent wrote:
>This is too generic a question. Cygwin is just a set of libraries that
>provide a Unix-like environment to programs... the behavior you describe
>would depend entirely on how those programs are written. Without any
>details of the proposed solutions, no one will be able to tell you
My guess is it works something like this:
1) during the day, chemistry machines Chem-AA through Chem-ZZ get some
data and the save it to their local disk
2) at some preset interval (hours, days, weeks?), the backup server
backups the data on each machine
I imagine that Cygwin would be involved in step 2. Via a cron job, a bash
script is run on the Windows clients which checks for a file. If the file
exists and is not currently open, rsync backs it up to a remote server via
An alternative solution would be for the FreeBSD or Linux server to run
Samba. The Windows clients map the Samba share and run a script via the
Windows scheduler, copying the original data file to the Samba share. This
could be easily accomplished with a DOS batch file or perl without
>For example, if rsync is being used to copy files to the
>backup server then it would depend on how rsync handles files that are
>open and being written to by other processes...and that would be a
>question for the rsync mailing list, or for the vendors that are
>offering those solutions to the bid.
I would imagine this rsync-based backup scheme would involve a number of
different programs: cygwin itself, crond, ssh, ps, and bash at a minimum in
addition to rsync. That's a lot of different programs.
I'm beginning to think a non-Cygwin solution may be their better fit and
may be the one they are leaning towards.
Again, thanks for the feedback.
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problems.html