Hi everybody,
I have a two nodes master/slave slony replication. It replicates
perfectly. But I have a problem doing the switchover. I have tried
some scripts to do this, but it always cause the same error.
Here is the output:
$ ./switchover
<stdin>:13: PGRES_FATAL_ERROR select "_slony_cluster".moveSet(1, 2);
- server closed the connection unexpectedly
This probably means the server terminated abnormally
before or while processing the request.
<stdin>:13: rollback transaction;
$vi switchover
#!/bin/sh
. config_vars
#!/bin/sh
slonik << _END_
cluster name = $CLUSTER;
node 1 admin conninfo = 'dbname=$DBNAME1 host=$HOST1 port=$PORT1
user=$REPLICATIONUSER';
node 2 admin conninfo = 'dbname=$DBNAME2 host=$HOST2 port=$PORT2
user=$REPLICATIONUSER';
#
# lock and move set
#
lock set (id=1, origin=1);
move set (id=1, old origin=1, new origin=2);
# subscribe set if required
subscribe set (id=1, provider=2, receiver=1, forward=yes);
_END_
************************
After this, the master database has locked the tables, and I only can
drop the database and start from scratch to test again. When I try to
insert into the Master, it says: ERROR: Slony-I: Table tellers is
currently locked against updates because of MOVE_SET operation in
progress. And the slave remains being slave. (I'm using pgbench to
test).
Thank you,
Daniel
_______________________________________________
Slony1-general mailing list
[email protected]
http://gborg.postgresql.org/mailman/listinfo/slony1-general
