Re: Question about DB VACUUM
От | Christopher Browne |
---|---|
Тема | Re: Question about DB VACUUM |
Дата | |
Msg-id | m3zngb7tl7.fsf@wolfe.cbbrowne.com обсуждение исходный текст |
Ответ на | Re: Question about DB VACUUM ("Chris White (cjwhite)" <cjwhite@cisco.com>) |
Ответы |
Re: Question about DB VACUUM
|
Список | pgsql-admin |
In the last exciting episode, cjwhite@cisco.com ("Chris White (cjwhite)") wrote: > BTW, the connection I shutdown, had not read, written or deleted any > large objects. It had read and written to other tables. This is causing > me concern as I am using a thread pool to provide access to the data in > the large object table, and this seems to imply I have to close each > connection after reading/writing or deleting a large object in order for > me to truly reclaim unused space when I issue my periodic vacuum > command. Yup, that sounds like a more-than-vaguely familiar story... The implication may not be _precisely_ correct, but the difference between what you're expecting and reality seems to be difficult to get at. I would expect that if you fired a (perhaps trivial) transaction through each of the connections once in a while, that would "clear things up" too. How to accomplish that may be the challenge... -- wm(X,Y):-write(X),write('@'),write(Y). wm('cbbrowne','ntlug.org'). http://www.ntlug.org/~cbbrowne/postgresql.html "With sufficient thrust, pigs fly just fine. However, this is not necessarily a good idea. It is hard to be sure where they are going to land, and it could be dangerous sitting under them as they fly overhead." -- RFC 1925
В списке pgsql-admin по дате отправления: