Fix thinko in previous commit.
We must still initialize minRecoveryPoint if we start straight with archive
recovery, e.g when recovering from a normal base backup taken with
pg_start/stop_backup. Otherwise we never consider the system consistent.
Branch
--
master
Details
---
http://git.p
Fix thinko in previous commit.
We must still initialize minRecoveryPoint if we start straight with archive
recovery, e.g when recovering from a normal base backup taken with
pg_start/stop_backup. Otherwise we never consider the system consistent.
Branch
--
REL9_2_STABLE
Details
---
http:
Alvaro Herrera writes:
> Tom Lane wrote:
>> Looking again at the spoonbill failure, it's striking that it gets as
>> far as plpython before choking. The symptoms are consistent with the
>> theory that USE_INLINE is getting #define'd by some Python header or
>> other, after having *not* been defin
Tom Lane wrote:
> I wrote:
> > But having said that, it seems there's something wrong with the
> > STATIC_IF_INLINE patch for the case where USE_INLINE doesn't get set.
> > I'm too tired to investigate right now, but if you manually disable
> > USE_INLINE in pg_config.h after configuring, do you se
I wrote:
> But having said that, it seems there's something wrong with the
> STATIC_IF_INLINE patch for the case where USE_INLINE doesn't get set.
> I'm too tired to investigate right now, but if you manually disable
> USE_INLINE in pg_config.h after configuring, do you see a problem?
I summoned t
Alvaro Herrera writes:
> (It seems to me that spoonbill is disabling USE_INLINE because of new
> warnings that are probably caused by the FORTIFY_SOURCE stuff. This
> seems to be interacting strangely with the new uses of USE_INLINE in the
> headers I modified.)
No, those aren't from _FORTIFY_SO
Alvaro Herrera wrote:
> Fix thinko in previous commit
>
> Since postgres.h includes palloc.h, definitions that affect the latter
> must be present before the former is included.
Spoonbill is still failing. I don't really understand what's going on
there -- I'll research further as best as I can.
Fix thinko in previous commit
Since postgres.h includes palloc.h, definitions that affect the latter
must be present before the former is included.
Per buildfarm results
Branch
--
master
Details
---
http://git.postgresql.org/pg/commitdiff/878daf2e72755feadbfb8d21aad26dafd8658086
Modifi
