[Dirvish] Hard Drive Died, and I was shocked to discover...

Keith Lofstrom keithl at kl-ic.com
Mon Feb 13 03:08:54 EST 2006


On Sun, Feb 12, 2006 at 10:24:46PM -0800, Steve Ramage wrote:
...
> hard disk. When I did I went into my vault and looked at the latest
> folder to see rsync_error.

...
> I highlighted Decembers, and Novembers, each 100 KB. Eventually I got to
> Octobers and found a copyy. It turns out while testing and writing the
> SSH Guide, I never bothered to switch over the real one. It was of
> little comfort to know that the test folder had backed up every day
> since then.
> 
> Anyway the whole point of my sad story is that I wish Dirvish had been
> e-mailing me to a user-definable e-mail address, to let me know. Also I
> wish there was a latest-good folder to so that I could get to files very
> easily.
...

It is good to check the daily output of dirvish, of course, but that does
not tell how much is getting backed up.  If a divish vault accidentally
has too many excludes set, then that will not show up either.  The
only real way to prove that dirvish is working is to restore a disk
from it.  Onto that spare drive you keep around for emergencies...

One of the things I do with my post-dirvish script is a "df" of the
target drive, which I run through a little perl script then store in
a log file in /var/log.  I look at the log file fairly frequently,
to make sure I am not getting too close to filling a backup disk.
Of course, I also notice if the disk is not slowly filling up.  
This is probably a good thing to make a part of dirvish proper, if
I ever get off my butt and finish fixing 1.3.2 .  

Of course, if somebody else would like to start hacking on 1.3.2 ,
that is what open source is all about...

Keith

-- 
Keith Lofstrom          keithl at keithl.com         Voice (503)-520-1993
KLIC --- Keith Lofstrom Integrated Circuits --- "Your Ideas in Silicon"
Design Contracting in Bipolar and CMOS - Analog, Digital, and Scan ICs


More information about the Dirvish mailing list