[Pgpool-general] Cannot trigger out-of-sync

Tatsuo Ishii ishii at sraoss.co.jp
Sun Jan 17 09:00:51 UTC 2010


pcp commands do not tell DB nodes are in out-of-sync status. So I'm
not sure what he is expecting here but I think it might be better to
trigger failover in this case (ERROR: pgpool detected difference of
the number of inserted, updated or deleted tuples). Opinions?
--
Tatsuo Ishii
SRA OSS, Inc. Japan

> I'm also getting this exact behavior. Any clue?
> 
> Thanks,
> Daniel
> 
> > -----Original Message-----
> > From: pgpool-general-bounces at pgfoundry.org [mailto:pgpool-general-
> > bounces at pgfoundry.org] On Behalf Of Ger Apeldoorn
> > Sent: Thursday, January 14, 2010 2:56 AM
> > To: pgpool-general at pgfoundry.org
> > Subject: [Pgpool-general] Cannot trigger out-of-sync
> > 
> > Hi,
> > 
> > I want to trigger an out-of-sync for a pgpool cluster in our
> > test-environment, but am unable to do it.
> > 
> > As I understood it, if pgpools get a different number of rows back
> from
> > a query, it should trigger an out of sync for one of the databases.
> > 
> > This is what I did:
> > 1) Insert new record directly in the local postgres database:
> > psql -p5433 my_database -c "insert into users (id, name, description,
> > password, field) values (10000,'gap','Ger','hihaho', 1);"
> > 
> > 2) Update this record through PGPool:
> > psql -p9999 my_database -c "update users set name = 'Harry' where id =
> > 10000;"
> > 
> > This generates the following error:
> > ERROR:  pgpool detected difference of the number of inserted, updated
> > or
> > deleted tuples. Possible last query was: "update users set name =
> > 'Harry' where id = 10000;"
> > HINT:  check data consistency between master and other db node
> > 
> > Unfortunately, no out-of-sync status is being set:
> > [root at db3 ~]# pcp_node_info 30 node1 9898 pgpool fake_password 1
> > node2 5433 2 1073741823.500000
> > [root at node2 ~]# pcp_node_info 30 node1 9898 pgpool fake_password 0
> > node1 5433 2 1073741823.500000
> > 
> > How can this be? Is there another (better) way to trigger a failure?
> > 
> > Thanks again,
> > Ger.
> > 
> > 
> > 
> > _______________________________________________
> > Pgpool-general mailing list
> > Pgpool-general at pgfoundry.org
> > http://pgfoundry.org/mailman/listinfo/pgpool-general
> _______________________________________________
> Pgpool-general mailing list
> Pgpool-general at pgfoundry.org
> http://pgfoundry.org/mailman/listinfo/pgpool-general


More information about the Pgpool-general mailing list