> >> As triggers are executed in order of their names, we've prefixed the
> >> trigger names with "zz". This should work fine as long as no-one uses
> >> trigger-name which starts with "zz".
> this seems a lot fragile... why system generated triggers has to be
> executed following the same rules (talking about order of execution)
> as user triggers? can't we simply execute them first or last or maybe
> be clever and mark one to be executed first and others last?
AFAICS, we do not have any category like system triggers. So yeah, it would
have been nice to generate triggers with names (starting with __ for
example) for such special triggers. But I don't think we disallow
user-triggers starting with underscores etc.
So some of the options could be:
- to add a new column in pg_trigger to indicate special or system triggers
which can be executed last (sorted order if multiple entries)
- invent a prefix "__partition__" or something and disallow user triggers to
use such a prefix for their names, plus introduce logic to execute them
(again sorted order if multiple entries) last.
In response to
pgsql-hackers by date
|Next:||From: Alvaro Herrera||Date: 2009-03-31 14:26:20|
|Subject: Re: [GENERAL] pgstattuple triggered checkpoint failureand database outage?|
|Previous:||From: Tom Lane||Date: 2009-03-31 13:57:46|
|Subject: Re: [GENERAL] pgstattuple triggered checkpoint failure and database outage? |