[Pgpool-general] pgpool-II 2.2 beta2 released

[Jaume Sabater]

On Wed, Feb 11, 2009 at 12:53 PM, Jaume Sabater <jsabater at gmail.com> wrote:

>>> Feb 11 12:29:13 pgsql1 pgpool: LOG:   pid 26729:
>>> ProcessFrontendResponse: failed to read kind from frontend. frontend
>>> abnormally exited
>>
>> I got different result:
>>
>> $ pgbench -c 10 -t 1000 test3
>> ERROR:  relation "branches" does not exist
>>
>> Am I missing something?
>
> The error I was showing was not the error in the client, but the error
> in the syslog where pgpool is running. It's the error reported by
> pgpool2, not pgbench, nor PostgreSQL.
>
> I am going to try it again, though.
>
> --
> Jaume Sabater
> http://linuxsilo.net/
>
> "Ubi sapientas ibi libertas"
>

Yep, again. This is what happened on the client side:

$ dropdb -h postgresql -p 9999 -U pgpool2 bench_replication
$ createdb -h postgresql -p 9999 -U pgpool2 bench_replication
$ /usr/lib/postgresql/8.3/bin/pgbench -h postgresql -p 9999 -U pgpool2
-d bench_replication -c 2 -t 100
pghost: postgresql pgport: 9999 nclients: 2 nxacts: 100 dbName:
bench_replication
ERROR:  relation "branches" does not exist

And this the syslog of the server running pgpool2:

Feb 11 13:00:55 pgsql1 pgpool: LOG:   pid 26817:
ProcessFrontendResponse: failed to read kind from frontend. frontend
abnormally exited

And this is the log from PostgreSQL on the master node:

2009-02-11 13:00:55 CET ERROR:  relation "branches" does not exist

Should the error in the syslog (the one generated by pgpool2) appear?
Or should pgpool2 "not care" about those errors, as they are not its
fault?

-- 
Jaume Sabater
http://linuxsilo.net/

"Ubi sapientas ibi libertas"