[svn.haxx.se] · SVN Dev · SVN Users · SVN Org · TSVN Dev · TSVN Users · Subclipse Dev · Subclipse Users · this month's index

Re: Linux Kernel Summit

From: Chris D. Sloan <cds_at_cs.hmc.edu>
Date: 2001-04-03 02:38:30 CEST

I'm extremely in favor of checksumming the repository. Being able to
*verify* its integrity once a month (or something) would be of huge
value. Sure, some open source authors may conclude that repository
corruption is not a concern for them, but I would have expected that
most would find their data quite valuable even if they can't establish
a monetary value for it.

As for checksumming the data on the wire, I think that obviously, all
things being equal it would be a good thing. I'm personally not *as*
concerned, however, because if my current working versions become
corrupted, I'm likely to figure it out before too much damage has been
done. If on the other hand, my history is silently corrupted, I may
not figure it out until it is too late.

In other words, I think that checksumming the transmission has value
(if it doesn't slow things down too much) and I think that
checksumming the repository has huge value.


On Mon, Apr 02, 2001 at 11:45:49AM -0700, Brian Behlendorf wrote:
> On Mon, 2 Apr 2001, Greg Hudson wrote:
> > For checksums in the DB: Larry makes a good argument that if you're
> > storing very valuble data, you want checksums (or some other form of
> > redundancy) in your permanent storage, and you want to periodically
> > verify that none of the data is corrupt. On the other hand, we're not
> > particularly targeting very expensive data; we're targeting the open
> > source community, according to our web pages. So, I would say: we
> > should ensure that our DB format is extensible enough to support
> > adding per-revision fields, and we should treat this as a "some day"
> > feature unless someone feels inspired to implement it right away.
> I'm quite certain that I'd be in the doghouse if the CVS repository for
> apache.org became corrupted in a non-recoverable way. I would consider
> the db of any long-running open source project to be very expensive, just
> a different form of expense.
> Brian

       Chris Sloan aka. christop aka. Bob aka. Dennis aka. ...
     Used to be a Harvey Mudd College Senior in Computer Science
     Ex-Member, CS Dept. Student Staff - Still a Purveyor of Puns
             BAIT (BIG's Artificial Intelligence Trainer)
Received on Sat Oct 21 14:36:27 2006

This is an archived mail posted to the Subversion Dev mailing list.