Pavan Deolasee escribió:
> >> Also I am not sure if the issue is big enough to demand the change.
> >
> > I think it is, effectively what we have now is "your DDL could fail randomly
> > for reasons that are out of your control" :(
>
> Yeah. I think we better fix this, especially given the above mentioned scenario.
The pg_shdepend code has code to grab a lock on the object being
dropped, which is also grabbed by someone who wants to add a dependency
on the object. Perhaps the pg_depend code should do the same.
I don't think this closes the original report though, unless we ensure
that the lock taken by vacuum conflicts with that one.
--
Alvaro Herrera http://www.CommandPrompt.com/
The PostgreSQL Company - Command Prompt, Inc.