[pgpool-general: 1240] Dropping template1 through PGPool not possible

[Maik Kulbe]

Hello List,

on my PGPool 3.2.1 Cluster, replicating with load balanced mode on Postgres
9.2 Nodes, I tried to change the template1 database to UTF-8.

Changeing the database metadate so I can drop the database worked just
fine, but when dropping the database once, I get a mismatch, when trying to
drop it twice the cluster just gets completely disconnected(I had four
online nodes when I started).

The following is what happened in the psql console:


postgres=# UPDATE pg_database SET datallowconn = TRUE WHERE datname =
'template0';
UPDATE 1
postgres=# \c template0
You are now connected to database "template0" as user "postgres".
template0=# DROP DATABASE template1;
ERROR:  cannot drop a template database
template0=# UPDATE pg_database SET datistemplate = FALSE WHERE datname =
'template1';
UPDATE 1
template0=# DROP DATABASE template1;
ERROR:  kind mismatch among backends. Possible last query was: "DROP
DATABASE template1;" kind details are: 0[C] 1[E: database "template1" is
being accessed by other users] 2[E: database "template1" is being accessed
by other users] 3[C]
HINT:  check data consistency among db nodes
ERROR:  kind mismatch among backends. Possible last query was: "DROP
DATABASE template1;" kind details are: 0[C] 1[E: database "template1" is
being accessed by other users] 2[E: database "template1" is being accessed
by other users] 3[C]
HINT:  check data consistency among db nodes
The connection to the server was lost. Attempting reset: Succeeded.
template0=# DROP DATABASE template1;
ERROR:  kind mismatch among backends. Possible last query was: "DROP
DATABASE template1;" kind details are: 0[E: database "template1" does not
exist] 1[C] 3[E: database "template1" does not exist]
HINT:  check data consistency among db nodes
ERROR:  kind mismatch among backends. Possible last query was: "DROP
DATABASE template1;" kind details are: 0[E: database "template1" does not
exist] 1[C] 3[E: database "template1" does not exist]
HINT:  check data consistency among db nodes
The connection to the server was lost. Attempting reset: Failed.
!>


Shouldn't there be at least some kind of mechanism in PGPool to check, if
this DROP operation is possible instead of degrading the whole cluster with
the failure of it?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.sraoss.jp/pipermail/pgpool-general/attachments/20121206/ecda0864/attachment.html>