Simon Riggs wrote:
>
> Recent changes to parameters aren't yet sufficiently well documented and
> I'd like to see this improved by the authors of those patches. I accept
> the behaviour changes, but we need full docs to explain them.
>
> * hot_standby doesn't mention that wal_level = hot_standby is also
> required, nor is there a xref.
>
> * wal_level = 'hot_standby' doesn't mention that the second parameter
> also needs to be set, nor is there a xref.
>
> * wal_level doesn't describe what the impacts are on a standby if the
> level is changed on the primary, nor is there a caution or a warning of
> any kind. For example, if a standby is setup with hot_standby = on and
> the primary is set wal_level = archive, does the standby start working
> if the primary changes wal_level = hot_standby? What happens if the
> primary is set wal_level = hot_standby and is then changed to archive?
>
> * wal_level doesn't explicitly describe that the levels are in sequence
> and that hot_standby is a superset of archive. The comment "so at least
> archive level must be used to enable .... streaming replication." can be
> misinterpreted to mean that hot_standby level cannot be used with SR,
> unless the "at least archive level" is successfully interpreted with the
> understanding that minimal < archive < hot_standby. So would like to see
> that made explicit.
I totally agree. Also, there is the new requirement to set wal_level
for PITR, which is not documented as a necessary setup step yet.
-- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB
http://enterprisedb.com