Re: [PATCHES] enable logging of start time/cookie for all backend processes

2007-08-02 Thread Andrew Dunstan
Alvaro Herrera wrote: Andrew Dunstan wrote: The attached patch makes a very small but useful change to the behaviour of log_line_prefix, by enabling the start time (%s) and cookie (%c) logging to occur for all backends rather than just for session processes (i.e. backends started for a

Re: [PATCHES] enable logging of start time/cookie for all backend processes

2007-08-02 Thread Andrew Dunstan
Andrew Dunstan wrote: Alvaro Herrera wrote: Andrew Dunstan wrote: The attached patch makes a very small but useful change to the behaviour of log_line_prefix, by enabling the start time (%s) and cookie (%c) logging to occur for all backends rather than just for session processes (i.e.

Re: [PATCHES] enable logging of start time/cookie for all backend processes

2007-08-01 Thread Andrew Dunstan
Andrew Dunstan wrote: The attached patch makes a very small but useful change to the behaviour of log_line_prefix, by enabling the start time (%s) and cookie (%c) logging to occur for all backends rather than just for session processes (i.e. backends started for a client connection). We

Re: [PATCHES] enable logging of start time/cookie for all backend processes

2007-08-01 Thread Magnus Hagander
Andrew Dunstan wrote: Andrew Dunstan wrote: The attached patch makes a very small but useful change to the behaviour of log_line_prefix, by enabling the start time (%s) and cookie (%c) logging to occur for all backends rather than just for session processes (i.e. backends started for a

Re: [PATCHES] enable logging of start time/cookie for all backend processes

2007-08-01 Thread Alvaro Herrera
Andrew Dunstan wrote: Andrew Dunstan wrote: The attached patch makes a very small but useful change to the behaviour of log_line_prefix, by enabling the start time (%s) and cookie (%c) logging to occur for all backends rather than just for session processes (i.e. backends started for a

Re: [PATCHES] enable logging of start time/cookie for all backend processes

2007-08-01 Thread Andrew Dunstan
Alvaro Herrera wrote: I am also wondering if setting it in InitAuxiliaryProcess is redundant (or rather, whether the extra set in each auxiliary process is redundant with the call in InitAuxiliaryProcess). Well, it's set right at the front of SubPostmasterMain, immediately after setting

Re: [PATCHES] enable logging of start time/cookie for all backend processes

2007-08-01 Thread Alvaro Herrera
Andrew Dunstan wrote: Alvaro Herrera wrote: I am also wondering if setting it in InitAuxiliaryProcess is redundant (or rather, whether the extra set in each auxiliary process is redundant with the call in InitAuxiliaryProcess). Well, it's set right at the front of SubPostmasterMain,

Re: [PATCHES] enable logging of start time/cookie for all backend processes

2007-08-01 Thread Andrew Dunstan
Alvaro Herrera wrote: Andrew Dunstan wrote: Alvaro Herrera wrote: I am also wondering if setting it in InitAuxiliaryProcess is redundant (or rather, whether the extra set in each auxiliary process is redundant with the call in InitAuxiliaryProcess). Well, it's set right at

Re: [PATCHES] enable logging of start time/cookie for all backend processes

2007-08-01 Thread Alvaro Herrera
Andrew Dunstan wrote: Alvaro Herrera wrote: Andrew Dunstan wrote: Alvaro Herrera wrote: I am also wondering if setting it in InitAuxiliaryProcess is redundant (or rather, whether the extra set in each auxiliary process is redundant with the call in InitAuxiliaryProcess). My

Re: [PATCHES] enable logging of start time/cookie for all backend processes

2007-08-01 Thread Andrew Dunstan
Alvaro Herrera wrote: My original thought was that it needed to be set anywhere MyProcPid was set. That still seems to make sense, it just looks like I missed a few places, which I should fix. +1 because of cleanliness, even if some calls turn out to be redundant. That was in fact

Re: [PATCHES] enable logging of start time/cookie for all backend processes

2007-08-01 Thread Alvaro Herrera
Andrew Dunstan wrote: The attached patch makes a very small but useful change to the behaviour of log_line_prefix, by enabling the start time (%s) and cookie (%c) logging to occur for all backends rather than just for session processes (i.e. backends started for a client connection). We