Re: executeQuery hangs on busy server - PROPOSAL: socketTimeout parameter

Поиск
Список
Период
Сортировка
От Kris Jurka
Тема Re: executeQuery hangs on busy server - PROPOSAL: socketTimeout parameter
Дата
Msg-id Pine.BSO.4.64.0708100538530.31534@leary.csoft.net
обсуждение исходный текст
Ответ на executeQuery hangs on busy server - PROPOSAL: socketTimeout parameter  (Andrea Spinelli <aspinelli@imteam.it>)
Ответы Re: executeQuery hangs on busy server - PROPOSAL: socketTimeout parameter
Re: executeQuery hangs on busy server - PROPOSAL: socketTimeout parameter
Список pgsql-jdbc

On Fri, 10 Aug 2007, Andrea Spinelli wrote:

> I am using jdbc to access a large database on a busy PostgreSQL server (8.1)
> hosted on Windows, from several web applications.
>
> Occasionally, my executeQuery hang forever. They block on a read operation in
> SocketInputStream.socketRead0 (seen on the debugger) and stay there for
> _days_. They disappear only by stopping the application server (Tomcat).

I believe this is a bug in the 8.1 windows port.  What's the exact version
you are using?  I think this was fixed in 8.1.6.

>       connection.setSoTimeout( 120000 );
>
> Long queries (>2min) fail with a SQLException signalling a communication
> problem with the server - which is what I want.
>
> I think this could be reworked into one more connection parameter
> "socketTimeout", where 0 means no timeout; the value of the parameter would
> of course substitute the hard-coded 120000.
>
> I've seen a feature request on GBorg about stopping long-running queries,
> which could be satisfied by what I'm proposing. (actually, the connection
> parameter works connection-wide, while the feature request deals with single
> queries).

People want the query to be stopped, but they don't want their whole
connection to be killed which is what your change does.

Kris Jurka

В списке pgsql-jdbc по дате отправления:

Предыдущее
От: Andrea Spinelli
Дата:
Сообщение: executeQuery hangs on busy server - PROPOSAL: socketTimeout parameter
Следующее
От: "László Hornyák"
Дата:
Сообщение: Re: statement caching patch from Laszlo Hornyak for review