Re: pg_rewind docs correction

On Mon, Mar 09, 2020 at 09:26:17AM -0400, James Coleman wrote:
>> - <filename>pg_stat_tmp/</filename>, and
>> - <filename>pg_subtrans/</filename> are omitted from the data copied
>> - from the source cluster. Any file or directory beginning with
>> - <filename>pgsql_tmp</filename> is omitted, as well as are
>> + <filename>pg_stat_tmp/</filename>, and
>> <filename>pg_subtrans/</filename>
>> + are omitted from the data copied from the source cluster. The files
>>
>> This is just reorganizing an existing list, why?
>>
>
> The grammar seemed a bit awkward to me, so while I was already reworking
> this paragraph I tried to clean that up a bit.

Thanks for the new patch, and sorry for the delay.

Okay, I saw what you were coming at here, with one sentence for
directories, and one for files.

> Still ongoing, correct? I guess I mentally think of them as being only one
> month, but I guess that's not actually true. Regardless I'm not sure what
> policy is for patches that have been in flight in hackers for a while but
> just missed being added to the CF app.

This is a documentation patch, so improving this part of the docs now
is fine by me, particularly as this is an improvement. Here are more
notes from me:
- I have removed the "As with a base backup" at the beginning of the
second paragraph you modified. The first paragraph modified already
references a base backup, so one reference is enough IMO.
- WAL replay does not happen from the WAL position where WAL diverged,
but from the last checkpoint before WAL diverged.
- Did some tweaks about the new part for configuration files, as it
may actually not be necessary to update the configuration for recovery
to complete (depending on the settings of the source, the target may
just require the creation of a standby.signal file in its data
directory particularly with a common archive location for multiple
clusters).
- Some word-smithing in the step-by-step description.

Is the updated version fine for you?
--
Michael