Re: [HACKERS] [PATCH] A hook for session start - Mailing list pgsql-hackers

From Fabrízio de Royes Mello
Subject Re: [HACKERS] [PATCH] A hook for session start
Date
Msg-id CAFcNs+q_Fv42E7_Rf2+TWJ2q3jsmaqLCGCVES=rUFp8SuiquiQ@mail.gmail.com
Whole thread Raw
In response to Re: [HACKERS] [PATCH] A hook for session start  (Michael Paquier <michael.paquier@gmail.com>)
Responses Re: [HACKERS] [PATCH] A hook for session start  (Michael Paquier <michael.paquier@gmail.com>)
List pgsql-hackers

On Fri, Nov 3, 2017 at 11:19 AM, Michael Paquier <michael.paquier@gmail.com> wrote:
>
>      /*
> +     * Setup handler to session end hook
> +     */
> +    if (IsUnderPostmaster)
> +        on_proc_exit(do_session_end_hook, 0);
> I think that it would be better to place that in ShutdownPostgres.
> This way it is possible to take actions before any resource is shut
> down.
>

Hmmm... ok but I have some doubt... ShutdownPostgres make sure to abort any active transaction (AbortOutOfAnyTransaction) and release all locks (LockReleaseAll). If we hook session end at this point we'll do that after this two steps and after that run again... something like:

...
    /* Make sure we've killed any active transaction */
    AbortOutOfAnyTransaction();

    /*   
     * User locks are not released by transaction end, so be sure to release
     * them explicitly.
     */
    LockReleaseAll(USER_LOCKMETHOD, true);

    if (session_end_hook) {
        (*session_end_hook) (port->database_name, port->user_name);

        /* Make sure session end hook doesn't leave trash behind */
        AbortOutOfAnyTransaction();
        LockReleaseAll(USER_LOCKMETHOD, true);
    }
...


> Passing the database name and user name does not look much useful to
> me. You can have access to this data already with CurrentUserId and
> MyDatabaseId.


This way we don't need to convert oid to names inside hook code.

Regards,

--
Fabrízio de Royes Mello
Consultoria/Coaching PostgreSQL
>> Timbira: http://www.timbira.com.br
>> Blog: http://fabriziomello.github.io
>> Linkedin: http://br.linkedin.com/in/fabriziomello
>> Twitter: http://twitter.com/fabriziomello
>> Github: http://github.com/fabriziomello

pgsql-hackers by date:

Previous
From: Fabrízio de Royes Mello
Date:
Subject: Re: [HACKERS] [PATCH] A hook for session start
Next
From: Tom Lane
Date:
Subject: Re: [HACKERS] Re: PANIC: invalid index offnum: 186 when processing BRIN indexes in VACUUM