Horton

Thread

Commits

  1. Fix single-record reads to use restore_command if available in pg_rewind

  2. Add -c/--restore-target-wal to pg_rewind

  1. forgotten initalization of a variable

    Kyotaro Horiguchi <horikyota.ntt@gmail.com> — 2020-04-21T06:08:30Z

    Hello.

The commit a7e8ece41c adds a new member restoreCommand to
XLogPageReadPrivate. readOneRecord doesn't make use of it but forgets
to set NULL. That can lead to illegal pointer access.

regards.

-- 
Kyotaro Horiguchi
NTT Open Source Software Center

  2. Re: forgotten initalization of a variable

    Michael Paquier <michael@paquier.xyz> — 2020-04-21T08:34:26Z

    On Tue, Apr 21, 2020 at 03:08:30PM +0900, Kyotaro Horiguchi wrote:
> The commit a7e8ece41c adds a new member restoreCommand to
> XLogPageReadPrivate. readOneRecord doesn't make use of it but forgets
> to set NULL. That can lead to illegal pointer access.

That's an oversight of the original commit.  Now, instead of failing
even if there is a restore command, wouldn't it be better to pass down
the restore_command to readOneRecord() so as we can optionally
improve the stability of a single record lookup?  This only applies to
a checkpoint record now, but this routine could be called elsewhere in
the future.  Please see the attached.
--
Michael

  3. Re: forgotten initalization of a variable

    Kyotaro Horiguchi <horikyota.ntt@gmail.com> — 2020-04-21T09:09:30Z

    At Tue, 21 Apr 2020 17:34:26 +0900, Michael Paquier <michael@paquier.xyz> wrote in 
> On Tue, Apr 21, 2020 at 03:08:30PM +0900, Kyotaro Horiguchi wrote:
> > The commit a7e8ece41c adds a new member restoreCommand to
> > XLogPageReadPrivate. readOneRecord doesn't make use of it but forgets
> > to set NULL. That can lead to illegal pointer access.
> 
> That's an oversight of the original commit.  Now, instead of failing
> even if there is a restore command, wouldn't it be better to pass down
> the restore_command to readOneRecord() so as we can optionally
> improve the stability of a single record lookup?  This only applies to

Oops! You're right.

> a checkpoint record now, but this routine could be called elsewhere in
> the future.  Please see the attached.

It looks fine to me.

regards.

-- 
Kyotaro Horiguchi
NTT Open Source Software Center

  4. Re: forgotten initalization of a variable

    Michael Paquier <michael@paquier.xyz> — 2020-04-21T23:13:02Z

    On Tue, Apr 21, 2020 at 06:09:30PM +0900, Kyotaro Horiguchi wrote:
> At Tue, 21 Apr 2020 17:34:26 +0900, Michael Paquier <michael@paquier.xyz> wrote in 
>> a checkpoint record now, but this routine could be called elsewhere in
>> the future.  Please see the attached.
> 
> It looks fine to me.

Fixed this way, then.  Thanks for the report!
--
Michael

  5. Re: forgotten initalization of a variable

    Kyotaro Horiguchi <horikyota.ntt@gmail.com> — 2020-04-22T00:13:47Z

    At Wed, 22 Apr 2020 08:13:02 +0900, Michael Paquier <michael@paquier.xyz> wrote in 
> On Tue, Apr 21, 2020 at 06:09:30PM +0900, Kyotaro Horiguchi wrote:
> > At Tue, 21 Apr 2020 17:34:26 +0900, Michael Paquier <michael@paquier.xyz> wrote in 
> >> a checkpoint record now, but this routine could be called elsewhere in
> >> the future.  Please see the attached.
> > 
> > It looks fine to me.
> 
> Fixed this way, then.  Thanks for the report!

Thans for fixing this!

regards.

-- 
Kyotaro Horiguchi
NTT Open Source Software Center