Re: pg_verify_checksums review - Mailing list pgsql-hackers

From Magnus Hagander
Subject Re: pg_verify_checksums review
Date
Msg-id CABUevEzEP7S3TxfHHe2Y82KvyZwpPWyvU80NJGrtQsQEXf-Ucg@mail.gmail.com
Whole thread Raw
In response to pg_verify_checksums review  (Daniel Gustafsson <daniel@yesql.se>)
Responses Re: pg_verify_checksums review
List pgsql-hackers


On Tue, Jun 19, 2018 at 10:25 AM, Daniel Gustafsson <daniel@yesql.se> wrote:
In looking over pg_verify_checksums I found a few small things that I think
would improve on it:

* pg_verify_checksums was placed in the Client Utils section in the docs.
Since it requries physical access to the cluster datafiles it seems to belong
in the Server Utils section.

Makes sense.


* The -D option and supported environment variable wasn’t documented.

* Only -D is supported for specifying the data directory, but most all other
utilities also support --pgdata on top of -D.  To present a consistent user
interface we should probably support --pgdata in pg_verify_checksums as well. 

The latter is I assume too invasive as we are past the freeze, but the first
two docs patches would make sense in 11 IMO as they document whats in the tree.

The attached patches fixes the above mentioned things (I don’t have a docs
toolchain working right now so the docs patches are best effort).

I believe both those are fine for 11, so I've pushed that. I kept it as a separate patch to make it easy enough to revert it if people prefer that :)

--

pgsql-hackers by date:

Previous
From: David Rowley
Date:
Subject: Re: Excessive CPU usage in StandbyReleaseLocks()
Next
From: Magnus Hagander
Date:
Subject: Re: Fix some error handling for read() and errno