Re: too much pgbench init output

On Tue, Sep 4, 2012 at 8:35 PM, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
> Peter Eisentraut <peter_e(at)gmx(dot)net> writes:
>> On 9/1/12 6:30 AM, Robert Haas wrote:
>>> On Sat, Sep 1, 2012 at 12:00 AM, Peter Eisentraut <peter_e(at)gmx(dot)net> wrote:
>>>> When initializing a large database, pgbench writes tons of "%d tuples
>>>> done" lines. I propose to change this to a sort of progress counter
>>>> that stays on the same line, as in the attached patch.
>
>>> I'm not sure I like this - what if the output is being saved off to a file?
>
>> I suppose we could print \n instead of \r then.
>
> Possibly off-the-wall idea: we could fix the "too much output" problem
> once and for all by going to a log scale.
>
> 10 tuples done
> 100 tuples done
> 1000 tuples done
> 10000 tuples done
> 100000 tuples done
> ...

I don't like that, because one of the things you can see by following
the current output is where the checkpoint stalls are happening during
the load. You'd lose the ability to notice any kind of slowdown after
the first few tuples with this kind of format.

Actually, this whole things seems like a solution in search of a
problem to me. We just reduced the verbosity of pgbench -i tenfold in
the very recent past - I would have thought that enough to address
this problem. But maybe not.

--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company