Re: elog() patch
От | Bruce Momjian |
---|---|
Тема | Re: elog() patch |
Дата | |
Msg-id | 200203011850.g21Io3K21315@candle.pha.pa.us обсуждение исходный текст |
Ответ на | Re: elog() patch ("Zeugswetter Andreas SB SD" <ZeugswetterA@spardat.at>) |
Список | pgsql-hackers |
> > > I also like LOG, since I don't like the current NOTICES in the log. > > > > Good, that was one of my goals. > > > > > Imho INFO and WARNING would be nothing for the log per default. > > > LOG would be things that are only of concern to the DBA. > > > My preferred client level would prbbly be WARNING (no INFO). > > > > Well, that is interesting. Currently we would send WARNING/NOTICE to > > the logs because it is an exceptional condition, though not as serious > > as error. > > Well, string truncation is imho not for the log, might interest the app > programmer but probably not the dba ? And if your point was to get rid > of the notices in the log (as is mine) you would have to not log Warning, > no ? OK, now things could get complicated. If we don't want to see NOTICE/WARNING in the log, you could say you don't want to see ERROR in the log, particularly syntax errors. Where do we go from here? You could set your server_min_messages to FATAL, but then you don't see LOG messages about server startup. I have considered allowing individual message types to be specified, particularly for the server logs, but the user API for that, particilarly as SET from psql, becomes very complicated. Maybe I have the ordering wrong for server_min_messages. Perhaps it should be: DEBUG5-1, INFO, NOTICE/WARNING, ERROR, LOG, FATAL, PANIC Nothing prevents us from doing that. Well, anyway, not sure how much I like it but I throw it out as an idea. -- Bruce Momjian | http://candle.pha.pa.us pgman@candle.pha.pa.us | (610) 853-3000+ If your life is a hard drive, | 830 Blythe Avenue + Christ can be your backup. | Drexel Hill, Pennsylvania19026
В списке pgsql-hackers по дате отправления: