Subject: Re: RecoveryWalAll and RecoveryWalStream wait events
To: Atsushi Torikoshi <atorik@gmail.com>
Cc: Kyotaro Horiguchi <horikyota.ntt@gmail.com>,
 PostgreSQL-development <pgsql-hackers@postgresql.org>
References: <124997ee-096a-5d09-d8da-2c7a57d0816e@oss.nttdata.com>
 <20200218.142016.2107516686265665330.horikyota.ntt@gmail.com>
 <2ea0c6f1-adcd-f2a2-fe5a-92376ad38432@oss.nttdata.com>
 <CACZ0uYE1UJomZ1tiB1vLwJSo_a2HAoXA6AF4=GRNfEkz-J1kNA@mail.gmail.com>
From: Fujii Masao <masao.fujii@oss.nttdata.com>
Message-ID: <aab7b94f-8140-48eb-3e31-698be55275d5@oss.nttdata.com>
Date: Tue, 17 Mar 2020 11:55:43 +0900
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.14; rv:68.0)
 Gecko/20100101 Thunderbird/68.6.0
MIME-Version: 1.0
In-Reply-To: 
 <CACZ0uYE1UJomZ1tiB1vLwJSo_a2HAoXA6AF4=GRNfEkz-J1kNA@mail.gmail.com>
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Language: en-US
Content-Transfer-Encoding: 8bit
Precedence: bulk


On 2020/03/15 0:06, Atsushi Torikoshi wrote:
> On 2020/02/19 21:46 Fujii Masao <masao.fujii@oss.nttdata.com <mailto:masao.fujii@oss.nttdata.com>>:
>  >> I agree to the former, I think RecoveryWalInterval works well enough.
>  >RecoveryWalInterval sounds confusing to me...
> 
> IMHO as a user, I prefer RecoveryRetrieveRetryInterval because
> it's easy to understand this wait_event is related to the
> parameter 'wal_retrieve_retry_interval'.
> 
> Also from the point of balance, the explanation of
> RecoveryRetrieveRetryInterval is lengthy, but I
> sometimes feel explanations of wait_events in the
> manual are so simple that it's hard to understand
> well.

+1 to document them more. It's not easy task, though..

>  >    Waiting when WAL data is not available from any kind of sources
>  >    (local, archive or stream) before trying again to retrieve WAL data,
> 
> I think 'local' means pg_wal here, but the comment on
> WaitForWALToBecomeAvailable() says checking pg_wal in
> standby mode is 'not documented', so I'm a little bit worried
> that users may be confused.

This logic seems to be documented in high-availability.sgml.
But, anyway, you think that "pg_wal" should be used instead of "local" here?

Regards,

-- 
Fujii Masao
NTT DATA CORPORATION
Advanced Platform Technology Group
Research and Development Headquarters