Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Skip to content

Commit 74a587a

Browse files
committed
Doc: document cases where queryid is stable
The documents were clear that queryid should not be assumed to be stable between major versions but said nothing about minor versions and left the reader to guess if that was implied by the mention of the instability of queryid between major versions. Here we give minor versions an explicit mention to indicate queryid can generally be assumed stable between minor versions. Reviewed-by: Michael Paquier Discussion: https://postgr.es/m/CAApHDvpYGE6h0cD9UO-eHySPynPj1L3J%3DHxT%2BA7Ud8_Yo6AuzA%40mail.gmail.com Backpatch-through: 12
1 parent cd26f08 commit 74a587a

File tree

1 file changed

+15
-9
lines changed

1 file changed

+15
-9
lines changed

doc/src/sgml/pgstatstatements.sgml

Lines changed: 15 additions & 9 deletions
Original file line numberDiff line numberDiff line change
@@ -292,15 +292,21 @@
292292
</para>
293293

294294
<para>
295-
As a rule of thumb, <structfield>queryid</structfield> values can be assumed to be
296-
stable and comparable only so long as the underlying server version and
297-
catalog metadata details stay exactly the same. Two servers
298-
participating in replication based on physical WAL replay can be expected
299-
to have identical <structfield>queryid</structfield> values for the same query.
300-
However, logical replication schemes do not promise to keep replicas
301-
identical in all relevant details, so <structfield>queryid</structfield> will
302-
not be a useful identifier for accumulating costs across a set of logical
303-
replicas. If in doubt, direct testing is recommended.
295+
Two servers participating in replication based on physical WAL replay can
296+
be expected to have identical <structfield>queryid</structfield> values for
297+
the same query. However, logical replication schemes do not promise to
298+
keep replicas identical in all relevant details, so
299+
<structfield>queryid</structfield> will not be a useful identifier for
300+
accumulating costs across a set of logical replicas.
301+
If in doubt, direct testing is recommended.
302+
</para>
303+
304+
<para>
305+
Generally, it can be assumed that <structfield>queryid</structfield> values
306+
are stable between minor version releases of <productname>PostgreSQL</productname>,
307+
providing that instances are running on the same machine architecture and
308+
the catalog metadata details match. Compatibility will only be broken
309+
between minor versions as a last resort.
304310
</para>
305311

306312
<para>

0 commit comments

Comments
 (0)