Re: [HACKERS] pg_resetwal is broken if run from v10 against older version of PG data directory - Mailing list pgsql-hackers

From Tom Lane
Subject Re: [HACKERS] pg_resetwal is broken if run from v10 against older version of PG data directory
Date
Msg-id 1791.1496092371@sss.pgh.pa.us
Whole thread Raw
In response to Re: [HACKERS] pg_resetwal is broken if run from v10 against olderversion of PG data directory  (Michael Paquier <michael.paquier@gmail.com>)
List pgsql-hackers
Michael Paquier <michael.paquier@gmail.com> writes:
> Agreed. Shouldn't this be back-patched? PG_CONTROL_VERSION has not
> been bumped between 9.4 and 9.5. Attached is a patch for HEAD.

Pushed with minor adjustments.  Notably, I didn't take your addition
of canonicalize_path() and referencing the file by absolute path.
That seems like an independent and rather debatable behavioral change.
All the other files that pg_resetwal touches are referenced with
relative paths; if we did want to change the reporting, shouldn't we
change it for all of them?
        regards, tom lane



pgsql-hackers by date:

Previous
From: Mark Kirkwood
Date:
Subject: Re: [HACKERS] logical replication - still unstable after all thesemonths
Next
From: Petr Jelinek
Date:
Subject: Re: [HACKERS] logical replication busy-waiting on a lock