[pgpool-general: 2690] Re: Postgres Sequence numbers different on backends when inserts run concurrently

[Tatsuo Ishii]

> Tatsuo Ishii wrote:
>>> We do see this problem as well, but we neither use JDBC nor parse/bind/exec.
>>> - pgpool 3.3.0
>>> - postgres 9.1.9
>>> - replication, load balancing, two backend nodes
>> 
>> It seems you are using old protocol (version 2) that might cause the
>> problem.
> 
> How do I tell which version of the protocol I'm running?

> 2013-09-23T15:33:59.000+02:00 pgpool[8837]: pool_read_string: read all from pending data. po:67 len:1

If you see this in the pgpool log (debug mode needs to be enabled),
it's old protocol since pool_read_string is called only when the old
protocol is used.

>> Is there any reason for you to use the old protocol?
> 
> It's not a conscious decision, it's just "there is old software".
> (I'm trying to pin down the misbehaving program.)
> 
> 
> Karsten
> _______________________________________________
> pgpool-general mailing list
> pgpool-general at pgpool.net
> http://www.pgpool.net/mailman/listinfo/pgpool-general