From: | Robert Treat <xzilla(at)users(dot)sourceforge(dot)net> |
---|---|
To: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
Cc: | Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us>, pgsql-hackers(at)postgresql(dot)org |
Subject: | Re: Generate user/group sysids from a sequence? |
Date: | 2003-01-17 19:43:00 |
Message-ID: | 1042832581.2124.19.camel@camel |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
On Fri, 2003-01-17 at 14:32, Tom Lane wrote:
> Robert Treat <xzilla(at)users(dot)sourceforge(dot)net> writes:
> > Have we decided it's really too difficult to remove all references to a
> > given sysid when the user is dropped?
>
> Getting at objects in other databases is considerably less practical
> than anything we've discussed here. At a minimum I think it would
> require launching an additional backend to connect into each other
> database. But that is pretty ugly because then the updates in each
> database are distinct transactions --- what if one fails, and you've
> already committed removals in other databases?
>
Perhaps you don't allow the removal unless all databases came up clean.
You can have the db admin go in a clean up each database manually if
need be. Granted this doesn't solve getting at those other databases in
the first place, but it eliminates any transactional issues.
Robert Treat
From | Date | Subject | |
---|---|---|---|
Next Message | Greg Stark | 2003-01-17 20:12:58 | Re: Suggestion for aggregate function |
Previous Message | Steve Crawford | 2003-01-17 19:37:26 | Terrible performance on wide selects |