Thread: fix CRC algorithm in WAL reliability docs

fix CRC algorithm in WAL reliability docs

From
Nathan Bossart
Date:
I noticed that the "WAL reliability" documentation says that we use CRC-32
for WAL records and two-phase state files, but we've actually used CRC-32C
since v9.5 (commit 5028f22).  I've attached a short patch to fix this that
I think should be back-patched to all supported versions.

I've attached a second patch that standardizes how we refer to these kinds
of algorithms in our docs.  Specifically, it adds dashes (e.g., "CRC-32C"
instead of "CRC32C").  Wikipedia uses this style pretty consistently [0]
[1] [2], and so I think we should, too.  I don't think this one needs to be
back-patched.

Thoughts?

[0] https://en.wikipedia.org/wiki/SHA-1
[1] https://en.wikipedia.org/wiki/SHA-2
[2] https://en.wikipedia.org/wiki/Cyclic_redundancy_check

-- 
nathan

Attachment

Re: fix CRC algorithm in WAL reliability docs

From
Robert Haas
Date:
On Thu, Aug 8, 2024 at 1:51 PM Nathan Bossart <nathandbossart@gmail.com> wrote:
> I noticed that the "WAL reliability" documentation says that we use CRC-32
> for WAL records and two-phase state files, but we've actually used CRC-32C
> since v9.5 (commit 5028f22).  I've attached a short patch to fix this that
> I think should be back-patched to all supported versions.
>
> I've attached a second patch that standardizes how we refer to these kinds
> of algorithms in our docs.  Specifically, it adds dashes (e.g., "CRC-32C"
> instead of "CRC32C").  Wikipedia uses this style pretty consistently [0]
> [1] [2], and so I think we should, too.  I don't think this one needs to be
> back-patched.
>
> Thoughts?

LGTM.

--
Robert Haas
EDB: http://www.enterprisedb.com



Re: fix CRC algorithm in WAL reliability docs

From
Nathan Bossart
Date:
On Thu, Aug 08, 2024 at 03:13:50PM -0400, Robert Haas wrote:
> LGTM.

Committed, thanks.

-- 
nathan