Re: Unexpected VACUUM FULL failure

Gregory Stark <stark(at)enterprisedb(dot)com> writes:
> "Simon Riggs" <simon(at)2ndquadrant(dot)com> writes:
>> It would be better to release 8.3 with a new, clean, fast implementation
>> of VF, than to release it with code that we *think* is right, but has so
>> far proved a source of some truly obscure bugs.

> Well building a suitable replacement for VACUUM FULL is more work than I was
> proposing.

Reimplementing VACUUM FULL for 8.3 is right out :-(. I do want to ship
a release in calendar 2007 ...

> I also think it's a safer course of action to create a new command, spend one
> or two releases improving it until we're sure it meets everyone's needs, then
> drop the old command.

Agreed. While I'd like to see V.F. dead, we need a proven replacement
first. (At the same time, we shouldn't spend more effort on V.F. than
we have to.)

regards, tom lane