Re: Open items - Mailing list pgsql-hackers

From Satoshi Nagayasu
Subject Re: Open items
Date
Msg-id 42C36726.2060100@nttdata.co.jp
Whole thread Raw
In response to Re: Open items  (Tom Lane <tgl@sss.pgh.pa.us>)
Responses Re: Open items
List pgsql-hackers
Tom Lane wrote:
> Josh Berkus <josh@agliodbs.com> writes:
> 
>>Satoshi,
>>
>>>sort=# select * from pg_stat_sorts ;
>>>� datname �| heap_all | index_all | heap_tape | index_tape | max_size
> 
> 
>>Good for me, if you explain the column names?   
> 
> 
> I was wondering about that too ... temporary sort files haven't got
> indexes ...

Sorry. It's my misunderstanding. index_tape will be zero forever...

My patch counts inittapes(), tuplesort_begin_heap() and tuplesort_begin_index(),
and collect them, and sum them through the stat collector.

I'm ready to rewrite if it is required.

Thanks.
-- 
NAGAYASU Satoshi <nagayasus@nttdata.co.jp>


pgsql-hackers by date:

Previous
From: Josh Berkus
Date:
Subject: Re: Checkpoint cost, looks like it is WAL/CRC
Next
From: Tom Lane
Date:
Subject: Re: Checkpoint cost, looks like it is WAL/CRC