db-derby-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Knut Anders Hatlen <Knut.Hat...@Sun.COM>
Subject Re: Embedded Derby and .dat files
Date Tue, 23 Sep 2008 12:41:43 GMT
Fuzzo <mcuccato@gmail.com> writes:

> Thanks for the answer.
>
> I've read these instructions in documentation but the Derby instance i use
> is embedded in ActiveMQ, not in my own application, and i don't wont modify
> the source code! :)

You don't need to modify the source code to try it. Just start the ij
command line tool, connect to the database and invoke the
statements. That would give a better indication of what the problem is.

> My questions are:
> - can i override any ActiveMQ Derby properties to force default checkpoint
> (after 10MB of logs)?

The default settings are:

derby.storage.logSwitchInterval=1048576
derby.storage.checkpointInterval=10485760

That is, switch to another log file when the current log file exceeds 1
MB, and checkpoint when you have 10 MB of log.

> - why ActiveMQ shoud enable log archiving mode?

No idea, it's just a guess. Checking if disabling log archiving helps,
would be a quick way to find out if this is the case.

Another possibility, also mentioned by others, is that there is a
long-running transaction or a transaction that the application has
forgotten to commit or abort. The checkpoint will keep all the log back
to the start of the oldest active transaction, so that could explain the
size of the log. In that case, most of the log should be deleted when
the database is rebooted. Does that happen?

-- 
Knut Anders

Mime
View raw message