Re: Why are we waiting? - Mailing list pgsql-hackers

On Wed, 2008-02-06 at 15:30 +0100, Staale Smedseng wrote:
> On Mon, 2008-02-04 at 19:46, Simon Riggs wrote:
> > We've got various facilities now for looking at LWLock waits, but I'd
> > like to have more information about the *reasons* for lock waits.
> > 
> > I know its possible to get backtraces in Dtrace at various tracepoints
> > but that can be fairly hard to interpret.

Thanks for this Staale.

>              Lock Id            Mode           Count
>        ProcArrayLock          Shared            1821
>        ProcArrayLock       Exclusive           49762
> 
>              Lock Id            Mode   Combined Time (ns)
>        ProcArrayLock          Shared          31114676303
>        ProcArrayLock       Exclusive         888356047549

Which looks like a mean service time of 17ms for X lock waiters.

What is the frequency distribution of lock wait time on ProcArrayLock?
Does the distribution vary over time?

--  Simon Riggs 2ndQuadrant  http://www.2ndQuadrant.com