[pgpool-general: 1761] Re: pgpool-II error and consultation on optimal environment for use

[Tatsuo Ishii]

Every Error detected by pgpool-II is not neccessaily the network
failure. I recommend you to check PostgreSQL logs as well.

> pool_read: read failed (Connection reset by peer)

It is possible PostgreSQL disconnected the connection to pgpool for
some reasons.
--
Tatsuo Ishii
SRA OSS, Inc. Japan
English: http://www.sraoss.co.jp/index_en.php
Japanese: http://www.sraoss.co.jp

> good morning, I have a query that I would like them because I find a pretty
> big question, I have installed postgresql 9.2 + pgpool-II 3.2.4 on a Redhat
> system 6 64 bits and heartbeat. my question is, what should be the
> operating system and database version for the pgpool-II works properly
> optimized, the pgpool-II is certified to work in a specific environment. I'm
> currently testing in the environment mentioned above, but I'm encountering some
> problems, it is an error that is repeated every 4 to 5 hours.
> pool_read: read failed (Connection reset by peer)
> May 16 16:34:56 pg01 pgpool[8196]: degenerate_backend_set: 1 fail over
> request from pid 8196
> May 16 16:34:56 pg01 pgpool[8196]: pool_flush_it: write failed to backend
> (1). reason: Broken pipe offset: 0 wlen: 5
> May 16 16:34:56 pg01 pgpool[12089]: starting degeneration. shutdown host
> 192.169.103.22(5432)
> May 16 16:34:56 pg01 pgpool[12089]: Restart all children
> May 16 16:34:56 pg01 pgpool[12089]: execute command: pgpool-failover
> May 16 16:34:56 pg01 pgpool[12089]: failover: set new primary node: -1
> May 16 16:34:56 pg01 pgpool[12089]: failover: set new master node: 0
> May 16 16:34:56 pg01 pgpool[12089]: failover done. shutdown host
> 192.169.103.22(5432)
> May 16 16:34:56 pg01 pgpool[5422]: worker process received restart request
> May 16 16:34:57 pg01 pgpool[5420]: pcp child process received restart
> request
> May 16 16:34:57 pg01 pgpool[12089]: PCP child 5420 exits with status 256 in
> failover()
> May 16 16:34:57 pg01 pgpool[12089]: fork a new PCP child pid 12124 in
> failover()
> May 16 16:34:57 pg01 pgpool[12089]: worker child 5422 exits with status 256
> May 16 16:34:57 pg01 pgpool[12089]: fork a new worker child pid 12125
> 
>   has become a constant error, and we have verified and you got no network
> errors or something. if anyone can help I would be very grateful