Re: [rdiff-backup-users] rdiff-backup keepalive

[Dave Kempe]

On Fri, 2003-03-21 at 06:01, Ben Escoto wrote:
> Yeah, --resume was problematic so was removed and there is a new
> system for 0.11.2 and later.  But what do you mean "retry"?  Keep
> starting up ssh or whatever?  I am really hesitant to even start
> thinking about this issue.  Right now the requirements are simple:

Yeah I was just wondering how hard it might be to get rdiff-backup to
see if the ssh connection reports a timeout and then just restart the
ssh session.
Perhaps if resume is fixed in 11.4 or later then I can write some sort
of script to monitor the connection and restart it myself?
Or is rerunning the backup straight away a resonable alternative to
--resume.



> rdiff-backup requires a bidirectional pipe to the destination.  That
> seems to be a reasonable division of labor.  I don't want to have to
> start thinking about networking issues, failed connections, resending
> corrupted packets, or anything like that.

I can understand that completely. Do you have any suggestions off the top 
of your head as to how I could monitor a timeout from a script myself?

thanks

dave