[pgpool-general: 2877] Re: Fail over

[Yugo Nagata]

Hi,

On Fri, 16 May 2014 16:32:26 +0000
Wolf Schwurack <wolf at uen.org> wrote:

> I'm trying to test the failover to the replication database
> 
> I stop the master database, here's whats in the pgpool log file. Pgpool doesn't execute fail over script.
> 
> 2014-05-16 10:07:18 ERROR: pid 3606: health check failed. 0 th host pgtest-db at port 5432 is down
> 2014-05-16 10:07:18 LOG:   pid 3606: health check retry sleep time: 30 second(s)

I think, you specified health_check_max_retries and health_check_retry_delay, so
failover will be exeuted after (health_check_max_retries * health_check_retry_delay) 
seconds.

> 2014-05-16 10:07:25 ERROR: pid 3748: connect_inet_domain_socket: getsockopt() detected error: Connection refused
> 2014-05-16 10:07:25 ERROR: pid 3748: make_persistent_db_connection: connection to pgtest-db(5432) failed
> 2014-05-16 10:07:25 ERROR: pid 3748: check_replication_time_lag: could not connect to DB node 0, check sr_check_user and sr_check_password
> 2014-05-16 10:07:35 ERROR: pid 3748: connect_inet_domain_socket: getsockopt() detected error: Connection refused
> 2014-05-16 10:07:35 ERROR: pid 3748: make_persistent_db_connection: connection to pgtest-db(5432) failed
> 2014-05-16 10:07:35 ERROR: pid 3748: check_replication_time_lag: could not connect to DB node 0, check sr_check_user and sr_check_password
> 
> Postgres 9.2.6
> Pgpool 3.3.3
> Master with streaming replication
> 
> Pgpool.conf file setting
> Failover_command= '/usr/lib/postgresql/9.2/bin/failover_stream.sh %d %H /tmp/trigger_file'
> 
> 
> 
> Wolfgang Schwurack
> Database/System Administrator
> Utah Education Network
> 801-587-9444
> wolf at uen.org
> 


-- 
Yugo Nagata <nagata at sraoss.co.jp>