Re: [HACKERS] Proposal to sync SET ROLE and pg_stat_activity

2008-08-28 Thread Grant Finnemore
Hi Alvaro, Alvaro Herrera wrote: Grant Finnemore wrote: Well, pg_stat_activity isn't really the problem here, because as you point out, it's just a view, and I could certainly redefine the view. The limiting factor is that the backend doesn't push the role name changes to the stats subsystem f

Re: [HACKERS] Proposal to sync SET ROLE and pg_stat_activity

2008-08-28 Thread Alvaro Herrera
Grant Finnemore wrote: > Well, pg_stat_activity isn't really the problem here, because as you > point out, it's just a view, and I could certainly redefine the view. > The limiting factor is that the backend doesn't push the role name > changes to the stats subsystem for either SET ROLE or SET SES

Re: [HACKERS] Proposal to sync SET ROLE and pg_stat_activity

2008-08-28 Thread Grant Finnemore
Hi Bernd, Bernd Helmle wrote: --On Mittwoch, August 27, 2008 09:35:03 +0200 Grant Finnemore <[EMAIL PROTECTED]> wrote: I have a session pool, where all connections to the database are obtained as a superuser. On issuing connections to the client, we invoke either SET ROLE or SET SESSION AUTHO

Re: [HACKERS] Proposal to sync SET ROLE and pg_stat_activity

2008-08-28 Thread Bernd Helmle
--On Mittwoch, August 27, 2008 09:35:03 +0200 Grant Finnemore <[EMAIL PROTECTED]> wrote: I have a session pool, where all connections to the database are obtained as a superuser. On issuing connections to the client, we invoke either SET ROLE or SET SESSION AUTHORIZATION and switch to a role wi

Re: [HACKERS] Proposal to sync SET ROLE and pg_stat_activity

2008-08-27 Thread Grant Finnemore
Hi Euler, Euler Taveira de Oliveira wrote: Grant Finnemore escreveu: Invoking pg_stat_activity after the SET ROLE is changed will however leave the usename unchanged. You're right. Because, as you spotted, usename is synonym of session usename. > The one problem with this mapping is that per

Re: [HACKERS] Proposal to sync SET ROLE and pg_stat_activity

2008-08-26 Thread Euler Taveira de Oliveira
Grant Finnemore escreveu: > Invoking pg_stat_activity after the SET ROLE is changed will however > leave the usename unchanged. > You're right. Because, as you spotted, usename is synonym of session usename. > SET SESSION AUTHORIZATION behaves similarly, although in that case, > it's documented t

[HACKERS] Proposal to sync SET ROLE and pg_stat_activity

2008-08-25 Thread Grant Finnemore
Hi, In the manual for SET ROLE, it's noted that an invocation of SET ROLE will leave the session_user unchanged, but will change the current_user. Invoking pg_stat_activity after the SET ROLE is changed will however leave the usename unchanged. (Also from the manual we note that a snapshot is ta