[Pgpool-general] Healt Check issues

[J. Carlos Muro]

2009/7/22 Tatsuo Ishii <ishii at sraoss.co.jp>

> > Every time a health check is launched i see these lines in PostgresSQL's
> > logs:
> >
> >     2009-07-21 14:01:15 MSD LOG:  could not receive data from client:
> > Connection reset by peer
> >     2009-07-21 14:01:15 MSD LOG:  unexpected EOF within message length
> word
>
> pgpool-II version? What is your pgpool-II and PostgreSQL
> configuration? Are they connected by TCP/IP?
>

Sorry, I didn't point that though I see that log lines healthcheck seems
anyways to work. I get connected from psql and execute queries normally:

    Jul 21 19:33:32 vapp1a pgpool: 2009-07-21 19:33:32 DEBUG: pid 28060:
starting health checking
    Jul 21 19:33:32 vapp1a pgpool: 2009-07-21 19:33:32 DEBUG: pid 28060:
health_check: 0 th DB node status: 2

PostgreSQL 8.3.7 on x86_64-pc-linux-gnu (from Ubuntu's repositories)
Pgpool 2.2.2 (from CVS at July 21th 2009) via TCP/IP to PostgreSQL
I reduced configuration to a simple case of just connection pooling:

     port = 5431

     replication_mode = false
     load_balance_mode = false
     replication_stop_on_mismatch = false
     replicate_select = true

     reset_query_list = 'ABORT; DISCARD ALL'
     master_slave_mode = false
     connection_cache = true

     health_check_timeout = 20
     health_check_period = 60
     health_check_user = 'pgpool'

     log_statement = true

     backend_hostname0 = 'vpgsql1a'
     backend_port0 = 5432
     backend_weight0 = 1
     backend_data_directory0 = '/var/lib/postgresql/8.3/main/'

It is just curious why those lines appear in PosgreSQL after every
healthcheck. They don't appear if I connect PostgreSQL from psql via pgpool.


Thanks!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://pgfoundry.org/pipermail/pgpool-general/attachments/20090722/e0176938/attachment.html>