[Linux-ha-dev] Future of /proc/ha

[Alan Robertson]

"Stephen C. Tweedie" wrote:
> 
> Hi,
> 
> On Thu, 28 Oct 1999 21:16:00 -0600, Alan Robertson <alanr@bell-labs.com>
> said:
> 
> > Volker,
> 
> > My only concern about the Berkley DB code is that most databases are
> > *highly* prone to corruption during crashes.  I would recommend that
> > no permanent state be kept in a database.
> 
> libdb2 supports journaled updates to the database.

But, if your filesystem has corrupted your base database file (as sometimes
happens), then you have made a nightmare out of startup.  Folks like Oracle
spend way more effort on getting this right than I ever intend to.

This can be due to database bugs, application bugs, filesystem bugs, fsck
bugs, or some combination of the three.

This is not to say that you *can't* make this work but that you are
well-advised to avoid it if you can.

All other things being equal, a design which doesn't depend on this is
superior to one that does.

And, as Volker has made clear, that wasn't his intention anyway.


	-- Alan Robertson
	   alanr@bell-labs.com