Re: Deadlocks and transactions
| От | David G. Johnston |
|---|---|
| Тема | Re: Deadlocks and transactions |
| Дата | |
| Msg-id | CAKFQuwbDhLKE-kp_MDHZMerXb3uPWFhduPOfmL4B5nBfLLSQVg@mail.gmail.com обсуждение исходный текст |
| Ответ на | Re: Deadlocks and transactions (JORGE MALDONADO <jorgemal1960@gmail.com>) |
| Список | pgsql-novice |
My application is a website.Let´s suppose the following scenario.* User 1 has already loaded a web page and clicks a button that triggers a transaction that includes 2 tables.* User 2 loads the same page and SELECTs data from the 2 tables currently in the transaction generated by User 1. User 2 does not trigger a transaction because he/she only gets data from the DB.
Well, if you want a consistent point-in-time picture of the data in those two tables you should probably perform the selects in a transaction too.
Is there any issue/problem for User 2?
MVCC mechanics eliminate the possibility of deadlocking between update queries and select queries (not select-for-update though). Because of it deadlocking is only possible between two updating transactions.
(Is it correct to reply-to-all when posting back to a question?)
Yes, reply-to-all is preferred; top-posting replies is not. Please inline or bottom-post (and trim quoting) as appropriate.
David J.
В списке pgsql-novice по дате отправления: