Re: Bug with caching SQLTypes in Connection:getSQLType(oid)
От | Antonio Fiol Bonnín |
---|---|
Тема | Re: Bug with caching SQLTypes in Connection:getSQLType(oid) |
Дата | |
Msg-id | 3C1260DC.3030007@w3ping.com обсуждение исходный текст |
Ответ на | Bug with caching SQLTypes in Connection:getSQLType(oid) (Ned Wolpert <ned.wolpert@knowledgenet.com>) |
Список | pgsql-jdbc |
I was the one that started the stated thread. Sorry about not following up, but I really do not know whether it was fixed or not. I cannot remember why, but I know that I compiled a new driver, and it did not get on very well with the backend (I think). Anyway, the development was nearly finished, and when I moved to the production server, not caching the requests was not critical, as the client and the server are on two machines on the same subnet. The development machine was in a different country than the server. So, I "forgot" about the problem. I think the interoperation issues were due to the fact we are on a 7.0.3 server, or something like that. We may be changing our server from a 7.0.3 to a 7.1 or 7.2 by January so I may test that again (new developments using similar techniques will take place). BTW, which one do you recommend, 7.1 or 7.2? The server is a RH6.2 2xPIII (800) 512Mb RAM, upgrading to 1Gb soon. The database size is about 3Gb, and growing every day. Servlet operation against the database is the main performance concern, and the server is most of the time doing many INSERTs on different tables, and precalculating statistics (SELECT - INSERT). So, I recall: I was about to test more, but I could not. I may soon have a second opportunity. Sorry for not clarifying that before. Antonio Fiol Rene Pijlman wrote: >On 07 Dec 2001 15:09:51 -0700, you wrote: > >>In some cases, when get getSQLType(oid) method is called, it doesn't >>cache the result, so it gets called alot. This is seen in Castor in >>a few cases. Anyone seen this before or am I the first? >> > >We've discussed this before: >http://archives2.us.postgresql.org/pgsql-jdbc/2001-09/msg00275.php. > >It's not entirely clear from the rest of the thread if the >problem disappeared when the OP upgraded to the latest >development version, but he did say that the improvement seemed >to be already implemented. > >Regards, >René Pijlman <rene@lab.applinet.nl> > >---------------------------(end of broadcast)--------------------------- >TIP 2: you can get off all lists at once with the unregister command > (send "unregister YourEmailAddressHere" to majordomo@postgresql.org) > >. >
В списке pgsql-jdbc по дате отправления: