agmentation from pgstatindex('idx_test2');
-[ RECORD 1 ]--+--
leaf_fragmentation | 0 <- OK
test=# INSERT INTO test2 select generate_series(1,1000);
INSERT 0 1000
test=# SELECT leaf_fragmentation from pgstatindex('idx_test2');
-[ RECORD 1 ]--+--
leaf_fragmentation | 28
not?
Yeah, I think so.
> Incidentally, the return value seems a bit useless.
Exactly. "Void" is enough.
--
NTT OSS Center
Tatsuhito Kasahara
kasahara.tatsuhito _at_ oss.ntt.co.jp
---(end of broadcast)---
TIP 5: don't forget to increase your free space map settings
ages | 0
avg_leaf_density | 90.64
leaf_fragmentation | 0
I also fix *_pages variables just in case.
Please confirm this.
Best regards.
--
NTT OSS Center
Tatsuhito Kasahara
kasahara.tatsuhito _at_ oss.ntt.co.jp
diff -crN postgresql-8.3.0.org/contrib/pgstattuple/pgstatindex.c
postgresql
uot;free_space" to double.
Best regards.
--
NTT OSS Center
Tatsuhito Kasahara
kasahara.tatsuhito _at_ oss.ntt.co.jp
*** postgresql-8.3.0.org/contrib/pgstattuple/pgstatindex.c 2007-11-16
06:14:31.0 +0900
--- postgresql-8.3.0/contrib/pgstattuple/pgstatindex.c
Tatsuhito Kasahara wrote:
> I fix the patch.
Oops, I forgot to attach the patch for pgstattuple.sql.
I send it again.
Best regards.
--
NTT OSS Center
Tatsuhito Kasahara
kasahara.tatsuhito _at_ oss.ntt.co.jp
diff -crN postgresql-8.3.0.org/contrib/pgstattuple/pgstatindex.c
postgresql-8.