[pgpool-hackers: 2809] Re: client_min_messages

Tatsuo Ishii ishii at sraoss.co.jp
Thu May 24 15:37:54 JST 2018


It seems there's something wrong here:

static void
send_message_to_frontend(ErrorData *edata)
	/*
	 * Leave if we are failing on sending the message to frontend.
	 */
	if (++frontend_error_recursion_depth > 2)
		return;

(gdb) p frontend_error_recursion_depth 
$11 = 97

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

> Hi Usama,
> 
> I have question regarding client_min_messages behavior. If my
> understanding is correct, following command will send all messages
> (including debug messages) to frontend.
> 
> pgpool set clint_min_messages to debug5;
> 
> In reality, only some of debug messages are sent to frontend. Even
> higher level logs (for example LOG messages) are lost. The rule for
> choosing these messages among other messages seems to be random. This
> is very different from the behavior of PostgreSQL. Any idea?
> 
> test=# pgpool set client_min_messages to debug5;
> SET
> test=# select 1;
> DEBUG:  pool_write: to frontend: kind:N po:0
> DEBUG:  pool_virtual_master_db_node_id: virtual_master_node_id:1 load_balance_node_id:1 PRIMARY_NODE_ID:0
>  ?column? 
> ----------
>         1
> (1 row)
> 
> Best regards,
> --
> Tatsuo Ishii
> SRA OSS, Inc. Japan
> English: http://www.sraoss.co.jp/index_en.php
> Japanese:http://www.sraoss.co.jp
> _______________________________________________
> pgpool-hackers mailing list
> pgpool-hackers at pgpool.net
> http://www.pgpool.net/mailman/listinfo/pgpool-hackers


More information about the pgpool-hackers mailing list