Re: REINDEX deadlock - Postgresql -9.1 - Mailing list pgsql-general

From John R Pierce
Subject Re: REINDEX deadlock - Postgresql -9.1
Date
Msg-id 51188E77.1010304@hogranch.com
Whole thread Raw
In response to Re: REINDEX deadlock - Postgresql -9.1  (Anoop K <anoopk6@gmail.com>)
Responses Re: REINDEX deadlock - Postgresql -9.1
List pgsql-general
On 2/10/2013 9:55 PM, Anoop K wrote:
> We analyzed the application side. It doesn't seem to be create a
> transaction and keep it open. StackTraces indicate that it is BLOCKED
> in JDBC openConnection.
>
> Any JDBC driver issue or other scenarios which can result in <*idle in
> transaction*> ?

JDBC has a wretched habit of autostarting transactions on any query if
the connection is not in autocommit mode.  if you don't want to use
autocommit mode, then you need to issue Commit() calls after each batch
of queries, even if the queries are read only, to release any implied locks.



--
john r pierce                                      37N 122W
somewhere on the middle of the left coast

pgsql-general by date:

Previous
From: Sergey Konoplev
Date:
Subject: Re: Upsert Functionality using CTEs
Next
From: Anoop K
Date:
Subject: Re: REINDEX deadlock - Postgresql -9.1