Re: [Catalina] Confirmed: Classreloading Session Persistence doesn'twork with Turbine

["Craig R. McClanahan" <Cr...@eng.sun.com>]

Jon Stevens wrote:

> Yea, I have proof AND a way to test it in 10 easy steps! :-)
>

Not as good as the 7-stepper last time, but it will have to do :-).

>
>     (but no idea on how to fix it :-()
>
> What I did to confirm and test it was this:
> (sorry Craig, it is going to require installing MySQL and loading the Scarab
> schema...it isn't hard and directions are provided below):
>

Not a problem ... I want this puppy to work.

One quick question (that I'm sure will be answered when I look into this in
detail) -- the log files show that it did the "Unload persistent sessions"
stuff, but no actual sessions or attributes were listed.  Was that because there
were none, or because you did not set the debugging detail level on the session
manager up?

    <Context path="/scarab">
        ...
        <Manager debug="99"/>
      ...
    </Context>

If this was not done in your tests, could you try it as well?  Doing this will
cause each individual session being unloaded or loaded (along with all of the
attributes) to be dumped to the log files as well.

Craig

Re: [Catalina] Confirmed: Classreloading Session Persistence doesn'twork with Turbine

[Jon Stevens <jo...@latchkey.com>]

on 2/22/01 2:42 PM, "Craig R. McClanahan" <Cr...@eng.sun.com>
wrote:

> Not a problem ... I want this puppy to work.

So do I. :-)

> One quick question (that I'm sure will be answered when I look into this in
> detail) -- the log files show that it did the "Unload persistent sessions"
> stuff, but no actual sessions or attributes were listed.  Was that because
> there
> were none, or because you did not set the debugging detail level on the
> session
> manager up?
> 
> <Context path="/scarab">
> ...
> <Manager debug="99"/>
> ...
> </Context>
> 
> If this was not done in your tests, could you try it as well?  Doing this will
> cause each individual session being unloaded or loaded (along with all of the
> attributes) to be dumped to the log files as well.
> 
> Craig

The log file was copy/pasted asis.

Here is the relevant section of the server.xml (as checked into Scarab's CVS
tree):

        <Context path="/scarab" docBase="scarab" reloadable="true"
debug="0">
           <Logger className="org.apache.catalina.logger.FileLogger"
                     prefix="scarab_log." suffix=".txt"
                  timestamp="true"/>
           <Loader checkInterval="3"
                className="org.apache.catalina.loader.StandardLoader"/>
           <Manager debug="99"/>
        </Context>

Maybe the Context debug=0 caused it to not print out the right information?

I was also going to try with the latest nightly of Tomcat, but I want to
confirm that the /lib/ directory is right before I go messing around with my
Scarab CVS tree to do the upgrade (before upgrading was replacing one .jar
file...now it is more work). :-)

thanks,

-jon

-- 
If you come from a Perl or PHP background, JSP is a way to take
your pain to new levels. --Anonymous
<http://jakarta.apache.org/velocity/> && <http://java.apache.org/turbine/>