Re: Server instrumentation for 8.1 - Mailing list pgsql-hackers

From Magnus Hagander
Subject Re: Server instrumentation for 8.1
Date
Msg-id 6BCB9D8A16AC4241919521715F4D8BCE6C7492@algol.sollentuna.se
Whole thread Raw
In response to Server instrumentation for 8.1  (Andreas Pflug <pgadmin@pse-consulting.de>)
List pgsql-hackers
> > Not kill -9. Kill -9 is "safe" because it causes a complete
> restart of
> > the postmaster (it's the same as a backend crash, really).
> Kill -INT
> > is also safe, because it does a simlpe query cancel.
>
> I don't recall exactly; AFAIR this was discussed between Dave
> and Tom.
> Actually, *both* should be accessible (maybe with some
> restriction making it difficult to call it from an sql cmd line)

Actually, I take the blame for starging that discussion - it started
when I craeted the pg_cancel_backend / pg_kill_backend questions.

Now, I *want* these functions. But I want them to be safe :-) (I want
kill -TERM on the commandline to be safe as well!)

//Magnus


pgsql-hackers by date:

Previous
From: Andrew - Supernews
Date:
Subject: Re: Server instrumentation for 8.1
Next
From: "John Hansen"
Date:
Subject: Re: SQL_ASCII vs. 7-bit ASCII encodings