Re: Packing, again

On Sun, 2004-09-05 at 12:04 +0100, Toby Dickenson wrote:
> On Sunday 05 Sep 2004 10:50, Toby Dickenson wrote:
> 
> > I'm very confident that an interrupted
> > pack wont corrupt the current version of the data, and pretty confident
> > that it wont cause problems when looking at historical revisions of
> > objects, or doing an undo etc. But there could still be a corner case bug
> > where an interrupted pack can cause trouble for a subsequent pack. 
> 
> There _are_ several corner cases that can be caused by an interrupted pack 
> because the bulk file deletion is not handled atomically. It is possible for 
> an interruption in phase 4 to leave some objects 'half deleted'. This can 
> cause checkds to report problems.

Ah, that's what I had thought. Does that mean if I restore the one file
it's complaining about missing, it might also fix the problem?

> *all* of these problems can be fixed by repeating the pack and allowing it to 
> finish. However you do need the pack time to be no longer than in the 
> interrupted pack. It is possible for the second pack in the following 
> scenario to fail.... does this apply to you?
> 
> a. many things were deleted 2 days ago
> b. start packing keeping 1 day of history. interrupt it during phase 4.
> c. start packing keeping 3 days of history.
> 
> To resolve these problem, you need to pack keeping 1 day of history ... the 
> same setting as the pack that was interrupted in step b. 

Hmm. As previously discussed (and to be remedied in all future packs
here), I didn't actually keep history, so how would I remedy using this?


--Richard



-------------------------------------------------------
This SF.Net email is sponsored by BEA Weblogic Workshop
FREE Java Enterprise J2EE developer tools!
Get your free copy of BEA WebLogic Workshop 8.1 today.
http://ads.osdn.com/?ad_id=5047&alloc_id=10808&op=click