Re: BUG #17928: Standby fails to decode WAL on termination of primary

On Tue, Sep 12, 2023 at 2:12 PM Michael Paquier <michael@paquier.xyz> wrote:
> On Tue, Sep 12, 2023 at 02:08:09PM +1200, Thomas Munro wrote:
> > I'll do that (and update the message).  Thanks for all the help!
>
> Cool, thanks!

I rebased (bumped the test number 038 -> 039), updated the commit
message[1], tested on a big endian machine (AIX/POWER), but then I
found there is something not quite right or stable with the WAL
positioning logic.  On one of my machines, but not another, the
back-patch to 12 fails in one of the tests unless I hack it to insert
a bit more WAL, and I think that's telling me that Perl code to
control the insertion point is not quite bullet proof enough yet.
I'll need to dig into this some more on Monday.

[1] https://github.com/macdice/postgres/commits/fix-12