Subject:
NetKarma Feedback
From:
Luisa Nevers <lnevers@bbn.com>
Date:
10/12/2011 12:28 PM
To:
"Ping, Robert James" <robping@indiana.edu>
CC:
"Jensen, Scott" <scjensen@indiana.edu>, Beth Plale <plale@cs.indiana.edu>, Vicraj Thomas <vthomas@bbn.com>

Hello Robert,

I have been taking a look a the NetKarma Version 2.5 and have some feedback.

1. The NetKarma Version 2.5 Quick Start Guide instruct the user to download
    version 2.0 files rather than version 2.5 :
 "https://pti.iu.edu/sites/default/files/karma-adaptor-2.0.tar.gz" and
 " https://pti.iu.edu/sites/default/files/karma-adaptor_v2.0_user_manual.pdf"

2. The NetKarma Version 2.5  Adaptor requires ANT and JDK to be installed, which
    are not listed as prerequisites in the Quick Start Guide. Also RabbitMQ shoudl
    be listed as a prerequisite.

3. Using the README for the NetKarma Adaptor.  In the section named
   "Edit the Environment Configuration File" the user is instructed to set
    "ADAPTOR_HOME, JAVA_HOME, and ANT_HOME", the first two exist in
   the configuration file, but ANT_HOME does not exist.

4. The Quick Start Guide Steps 4 and 5 instruct to modify karma-adaptor.properties
   and KarmaQueryConfig.txt, but does not provide a path to these files.

5.  The "NetKarma GUSH Adaptor Tool User Manual"  section 2.2 "Service Dependencies"
      instructs me to download  RabbitMQ, but does not specify if I need the server, the client,
     both, or some other  component.

6. The "NetKarma GUSH Adaptor Tool User Manual"  section 3 lists "karma-adaptor.homedir"
     as a karma-adaptor.properties setting, but is  not found in the file.  Is this an optional setting?

7. The "NetKarma GUSH Adaptor Tool User Manual"  section 3 describes the setting
    "messaging.hostname" as "Hostname that hosts RabbitMQ Server" but the configuration
   file karma-adaptor.properties describes the same setting as the
  "<URL where RabbitMQ server is hosted>".  After trying  both seems the the URL description
  is incorrect, properties file needs update.

8. Question, should I be able to process a gush logfile using the ruleset.xml delivered as part of the
    adaptor package?  I tried  parsing a very simple gush log (start and stop gush) with the
   provenance_collection.sh script and   it fails with "Invalid rulefile, use the correct version of the rulefile!"

Thanks,

Luisa