From: Thomas Munro <thomas.mu...@gmail.com>
> We have automated tests for many specific replication and recovery scenarios,
> but nothing that tests replay of a wide range of records.
> People working on recovery code often use installcheck (presumably along
> with other custom tests) to exercise it, sometimes with
> wal_consistency_check enabled. So, why don't we automate that? Aside
> from exercising the WAL decoding machinery (which brought me here), that'd
> hopefully provide some decent improvements in coverage of the various redo
> routines, many of which are not currently exercised at all.
>
> I'm not quite sure where it belongs, though. The attached initial sketch
> patch
I think that's a good attempt. It'd be better to confirm that the database
contents are identical on the primary and standby. But... I remember when I
ran make installcheck with a standby connected, then ran pg_dumpall on both the
primary and standby and compare the two output files, about 40 lines of
difference were observed. Those differences were all about the sequence
values. I didn't think about whether I should/can remove the differences.
+# XXX The tablespace tests don't currently work when the standby shares a
+# filesystem with the primary due to colliding absolute paths. We'll skip
+# that for now.
Maybe we had better have a hidden feature that creates tablespace contents in
/tablespace_location/PG_..._<some_name>/
<some_name> is the value of cluster_name or application_name.
Or, we may provide a visible feature that allows users to store tablespace
contents in a specified directory regardless of the LOCATION value in CREATE
TABLESPACE. For instance, add a GUC like
table_space_directory = '/some_dir'
Then, the tablespace contents are stored in /some_dir/<tablespace_name>/. This
may be useful if a DBaaS provider wants to offer some tablespace-based feature,
say tablespace transparent data encryption, but doesn't want users to specify
filesystem directories.
Regards
Takayuki Tsunakawa
