Re: BUG #15467: The database subdirectory "pg_tblspc/1932420460/PG_10_201707211/16400" is missing.

> Do you still get the error from just
>
> select oid from pg_database;
>
> or
>
> select oid,dattablespace from pg_database;

That does work!

root=# select oid,dattablespace from pg_database ;
oid | dattablespace
-------+---------------
16402 | 1663
1 | 1663
12291 | 1663
16400 | 1932420460
12292 | 1663
16401 | 1663
(6 rows)

> I just noticed: 1932420460 = 0x732E656C = "le.s" which is suspiciously
> textual.
>
> Any chance you can do this query:
>
> select pg_relation_filepath('pg_database'::regclass);
>
> and then do hexdump -C on that file?

Quite a bit there, but the region around there does look weird:

00001d40 00 00 00 00 00 00 00 00 00 01 00 00 ff ff ff ff |................|
00001d50 00 00 00 00 f4 06 00 00 01 00 00 00 6c 65 2e 73 |............le.s|
00001d60 6f 00 70 79 63 00 65 2e 00 00 00 00 00 00 00 00 |o.pyc.e.........|
00001d70 00 ff 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................|
00001d80 00 00 00 00 00 00 00 00 3a 00 00 ff 00 00 00 00 |........:.......|
00001d90 00 00 00 00 00 00 00 ff 00 00 00 00 00 00 00 00 |................|
00001da0 28 f8 12 00 00 00 00 00 00 00 00 00 00 00 00 00 |(...............|
00001db0 0d 00 0d 80 0a 29 20 00 00 00 00 00 03 30 00 00 |.....) ......0..|
00001dc0 74 65 6d 70 6c 61 74 65 30 00 00 00 00 00 00 00 |template0.......|
00001dd0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................|

The le.s that you noticed actually looks like le.so, and then there's
a pyc and an "e." after it (null-separated). Also weird is that there
are six different sections describing(?) template1:

hexdump -C global/1262 |grep -B1 -A1 template
00001110 0b 00 0d 80 0a 29 20 00 00 00 00 00 01 00 00 00 |.....) .........|
00001120 74 65 6d 70 6c 61 74 65 31 00 00 00 00 00 00 00 |template1.......|
00001130 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................|
--
00001240 0b 00 0d c0 0a 25 20 00 00 00 00 00 01 00 00 00 |.....% .........|
00001250 74 65 6d 70 6c 61 74 65 31 00 00 00 00 00 00 00 |template1.......|
00001260 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................|
--
00001360 00 00 00 00 00 00 00 00 09 00 0d c0 0a 25 20 00 |.............% .|
00001370 00 00 00 00 01 00 00 00 74 65 6d 70 6c 61 74 65 |........template|
00001380 31 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |1...............|
--
00001480 08 00 0d c0 0a 25 20 00 00 00 00 00 01 00 00 00 |.....% .........|
00001490 74 65 6d 70 6c 61 74 65 31 00 00 00 00 00 00 00 |template1.......|
000014a0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................|
--
000015a0 00 00 00 00 00 00 00 00 07 00 0d c0 0a 25 20 00 |.............% .|
000015b0 00 00 00 00 01 00 00 00 74 65 6d 70 6c 61 74 65 |........template|
000015c0 31 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |1...............|
--
00001db0 0d 00 0d 80 0a 29 20 00 00 00 00 00 03 30 00 00 |.....) ......0..|
00001dc0 74 65 6d 70 6c 61 74 65 30 00 00 00 00 00 00 00 |template0.......|
00001dd0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................|
--
00001ee0 06 00 0d c0 0a 25 20 00 00 00 00 00 01 00 00 00 |.....% .........|
00001ef0 74 65 6d 70 6c 61 74 65 31 00 00 00 00 00 00 00 |template1.......|
00001f00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................|

Also, of the two main databases in use, one of them has 8 different
sections, and the other has 6. I'm not seeing anything like that on
other systems, but maybe it's normal?

I can paste in the entire hexdump if it's useful, but there's a whole
ton of stuff there, so I'm leaving it out unless somebody really wants
to see it.