Re: Very long time to commit or close connections
От | Chris Smith |
---|---|
Тема | Re: Very long time to commit or close connections |
Дата | |
Msg-id | 02bf01c3de9e$ad7e4b00$6f00000a@KYA обсуждение исходный текст |
Ответ на | Very long time to commit or close connections ("Chris Smith" <cdsmith@twu.net>) |
Список | pgsql-general |
Oops, meant to copy the list on this... ----- Original Message ----- From: "Chris Smith" <cdsmith@twu.net> To: "Tom Lane" <tgl@sss.pgh.pa.us> Sent: Monday, January 19, 2004 8:01 AM Subject: Re: [GENERAL] Very long time to commit or close connections > > There are not that many things happening during connection close. Does > > your app use temp tables? LISTEN/NOTIFY? > > Nope. It's actually a very simple application. It uses plain select, insert, > update, and delete statements with nothing particularly unusual beyond that. > > > Nope ... how about looking at the contents of the pg_locks system view > > when this happens? What external conditions correspond to the instant > > when the hangs are released? (I do not believe that it just happens by > > magic --- look for background cron jobs starting or finishing at that > > time, eg a cron job that issues VACUUMs.) > > Hmm... I will definitely look. I pretty much know that it's not a vacuum. I > suppose I could have our system monitor insert information from the 'ps' > utility when there are outstanding connections. > > > What PG version is this, anyway? > > 7.3.2. We could upgrade within the 7.3 series, but have not qualified the > application on 7.4 at this point. (there are complications, mostly because of > a patch we apply to the JDBC drivers, which change for 7.4). > > -- > www.designacourse.com > The Easiest Way to Train Anyone... Anywhere. > > Chris Smith - Lead Software Developer/Technical Trainer > MindIQ Corporation >
В списке pgsql-general по дате отправления: