Re: track_io_timing default setting

2021-12-22 Thread Peter Geoghegan
On Wed, Dec 22, 2021 at 11:16 AM Stephen Frost wrote: > > I set "track_io_timing" to "on" all the time, same as "log_lock_waits", > > so I'd want them both on by default. > > Same. I'd also push back and ask what modern platforms still require a > kernel call for gettimeofday, and are we really d

Re: track_io_timing default setting

2021-12-22 Thread Stephen Frost
Greetings, * Laurenz Albe (laurenz.a...@cybertec.at) wrote: > On Fri, 2021-12-10 at 10:20 -0500, Tom Lane wrote: > > Jeff Janes writes: > > > Can we change the default setting of track_io_timing to on? > > > > That adds a very significant amount of overhead on some platforms > > (gettimeofday is

RE: [EXTERNAL] Re: track_io_timing default setting

2021-12-10 Thread Godfrin, Philippe E
>-Original Message- >Sent: Friday, December 10, 2021 9:20 AM >To: Jeff Janes >Cc: PostgreSQL Hackers >Subject: [EXTERNAL] Re: track_io_timing default setting >Jeff Janes writes: > Can we change the default setting of track_io_timing to on? >That adds a ve

Re: track_io_timing default setting

2021-12-10 Thread Tomas Vondra
On 12/10/21 17:22, Laurenz Albe wrote: On Fri, 2021-12-10 at 10:20 -0500, Tom Lane wrote: Jeff Janes writes: Can we change the default setting of track_io_timing to on? That adds a very significant amount of overhead on some platforms (gettimeofday is not cheap if it requires a kernel call).

Re: track_io_timing default setting

2021-12-10 Thread Laurenz Albe
On Fri, 2021-12-10 at 10:20 -0500, Tom Lane wrote: > Jeff Janes writes: > > Can we change the default setting of track_io_timing to on? > > That adds a very significant amount of overhead on some platforms > (gettimeofday is not cheap if it requires a kernel call).  And I > doubt the claim that t

Re: track_io_timing default setting

2021-12-10 Thread Tom Lane
Jeff Janes writes: > Can we change the default setting of track_io_timing to on? That adds a very significant amount of overhead on some platforms (gettimeofday is not cheap if it requires a kernel call). And I doubt the claim that the average Postgres user needs this, and doubt even more that t

RE: track_io_timing default setting

2021-12-10 Thread Jakub Wartak
> Can we change the default setting of track_io_timing to on? +1 for better observability by default. > I can't imagine a lot of people who care much about its performance impact > will be running the latest version of PostgreSQL on ancient/weird systems  > that have slow clock access. (And the

track_io_timing default setting

2021-12-09 Thread Jeff Janes
Can we change the default setting of track_io_timing to on? I see a lot of questions, such as over at stackoverflow or dba.stackexchange.com, where people ask for help with plans that would be much more useful were this on. Maybe they just don't know better, maybe they can't turn it on because th