Re: Log level of logical decoding

Поиск
Список
Период
Сортировка
От Francesco Nidito
Тема Re: Log level of logical decoding
Дата
Msg-id VI1PR0902MB2127E26E0BA578F25E274EB3A6AA0@VI1PR0902MB2127.eurprd09.prod.outlook.com
обсуждение исходный текст
Список pgsql-hackers
Hi All,

One week ago I posted the following question in the general list but probably the developers' list is a better place for such question. 

Thanks in advance (again)!

Francesco


From: Francesco Nidito
Sent: 30 November 2018 15:10
To: pgsql-general@lists.postgresql.org
Subject: Log level of logical decoding
 
Hi All,

With reference to an old thread (https://www.postgresql.org/message-id/8307.1419439826%40sss.pgh.pa.us), our application polls the database instance periodically to check if there are new transaction in the WAL waiting to be processed.  As a consequence of this, our log file is full of messages like 

2018-11-30 10:37:29 UTC LOG:  starting logical decoding for slot "test_slot"
2018-11-30 10:37:29 UTC DETAIL:  streaming transactions committing after 4AC/41D5F950, reading WAL from 4AC/41C16868
2018-11-30 10:37:29 UTC STATEMENT:
            SELECT * FROM pg_logical_slot_peek_changes('test_slot', NULL, NULL,
                          'include-timestamp', '1', 'skip-empty-xacts', '1')

As those logs are of dubious utility, and following the comment of Tom in the last email of the old thread, I would like to ask why this message is a LOG and not a DEBUG and if there is any plan to make them DEBUG?

Thanks a lot in advance!

Francesco

В списке pgsql-hackers по дате отправления:

Предыдущее
От: "Matsumura, Ryo"
Дата:
Сообщение: RE: [PROPOSAL]a new data type 'bytea' for ECPG
Следующее
От: John Naylor
Дата:
Сообщение: Re: WIP: Avoid creation of the free space map for small tables