[Mifos-developer] session timeout

[Adam Monsen]

In a recent patch
( http://article.gmane.org/gmane.comp.finance.mifos.devel/3362 ) I moved
session timeout configuration from the database to the application-wide
configuration properties file,
applicationConfiguration.default.properties...

SessionTimeout=60

But I just realized this value isn't read from anywhere in the Java
code, and both src/org/mifos/META-INF/web.xml and test/WEB-INF/web.xml
have a session timeout defined as follows:

<session-config>
<session-timeout>30</session-timeout>
</session-config>

So having session timeout in the application-wide configuration
properties file seems redundant, and not at all helpful.

Thoughts?

--
Adam Monsen

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/

[Van Mittal-Henkle]

Nice catch.

Can you verify (or maybe you already have) that changing the web.xml
value has the expected effect and that changing the other config value
has no effect?

Once we know that for sure, then we can send this issue to the
functional mailing list and ask if the web.xml configuration is good
enough or if having this configuration along side other configuration
values is important.

--Van

> In a recent patch
> ( http://article.gmane.org/gmane.comp.finance.mifos.devel/3362 ) I
> moved session timeout configuration from the database to the
> application-wide configuration properties file,
> applicationConfiguration.default.properties...
>
> SessionTimeout=60
>
> But I just realized this value isn't read from anywhere in the Java
> code, and both src/org/mifos/META-INF/web.xml and
> test/WEB-INF/web.xml have a session timeout defined as follows:
>
> <session-config>
> <session-timeout>30</session-timeout>
> </session-config>
>
> So having session timeout in the application-wide configuration
> properties file seems redundant, and not at all helpful.
>
> Thoughts?

[Adam Monsen]

On Thu, 2007-12-20 at 14:13 -0500, Van Mittal-Henkle wrote:
> Nice catch.
>
> Can you verify (or maybe you already have) that changing the web.xml
> value has the expected effect and that changing the other config value
> has no effect?

Yes, correct. Only the value in the web.xml seems to have any effect on
raising or lowering the actual session timeout.

And lowering session timeout is an easy way to repro issue #1475, btw.
Note added to issue.

> Once we know that for sure, then we can send this issue to the
> functional mailing list and ask if the web.xml configuration is good
> enough or if having this configuration along side other configuration
> values is important.

[...]

Ok, I can care of that after I eat some lunch. Thank you!
-Adam

--
Adam Monsen