Andy,

Based on Joshua Taylor's advice, I added a connection pool for writes to ensure 
that any concurrent writes have different connections.

I use transactions (model.begin()/model.commit()) for write, and using 
SDBFactory.connectStore() to get new connections each time the connection pool 
is empty.

I am using SDB 1.3.4 and Postgres 9.0-801 (based on the mvn artifact versions).

Dave

-----Original Message-----
From: Andy Seaborne [mailto:[email protected]] On Behalf Of Andy 
Seaborne
Sent: Sunday, September 09, 2012 2:16 PM
To: [email protected]
Subject: Re: SDb and concurrent update

Which version of SDB are you using? (And of PostgreSQL?)

This might be JENA-118 [118].

You will need multiple store connections (multiple JDBC connections) if you 
want concurrent transactions.  Locking is not connected internally to 
transactions.

        Andy

Reply via email to