[pgpool-general: 3527] Re: Amazon EC2 Load Balancer health check kills pgPool node

[Tatsuo Ishii]

> Hey,
> 
> We use the same setup on 3.3.1 (AWS with ELB doing TCP health check), but
> we get these error messages:
> 
> 2015-03-16 17:38:05 ERROR: pid 3049: read_startup_packet: incorrect packet
> length (0)
> 2015-03-16 17:38:15 ERROR: pid 3049: read_startup_packet: incorrect packet
> length (0)
> 2015-03-16 17:38:25 ERROR: pid 3049: read_startup_packet: incorrect packet
> length (0)
> 2015-03-16 17:38:35 ERROR: pid 3049: read_startup_packet: incorrect packet
> length (0)
> 
> There is no backend crash I can see.

Interesting.

When I connect to pgpool-II localhost with telnet, I see these:

2015-03-17 08:15:50: pid 6436: ERROR:  failed while reading startup packet
2015-03-17 08:15:50: pid 6436: DETAIL:  incorrect packet length (-720903)

This is git HEAD (almost same as 3.4.1).

Best regards,
--
Tatsuo Ishii
SRA OSS, Inc. Japan
English: http://www.sraoss.co.jp/index_en.php
Japanese:http://www.sraoss.co.jp