= Setting up and Running the IPP Data Distribution Client =

(back to [wiki:IPP_for_PS1 IPP for PS1]; back to [wiki:GPC1_DataDistribution GPC1 Data Distribution])

This page shows how to set up and run the IPP Data Distribution Client.  In this example, a working simtest distribution set may be downloaded from the IfA IPP installation.  

 * Build a current ipp and insure that it works by running simtest.  See [wiki:IPP_Installation] and [wiki:SimtestNew].

 * Create an empty database in mysql with some name.  This example uses {{{rc_test}}}: 
{{{ 
dbadmin project somehost someaccount rc_test 
}}}

 * Populate the tables: {{{pxadmin -dbname rc_test -create-mirror}}}

 * Add the source (a reference to our data store) to the database
{{{
receivetool -dbname rc_test -definesource    \
   -source  http://datastore.ipp.ifa.hawaii.edu/ \
   -product ps1-md \
   -workdir $WORKDIR \
   -comment 'simtest data source'
}}}

This works for ifaps1:

{{{
receivetool -dbname ifaps1_0 -definesource    \
   -source  http://datastore.ipp.ifa.hawaii.edu/  \
   -product ps1-md \
   -workdir /data/ifaps1.0/ps1data/MD/ \
   -comment 'ifa md repository'
}}}

 * workdir is the local directory to download the files to
 * dbname is the name of the database you chose (ifaps1 uses ifaps1_0)




Where {{{$WORKDIR}}} is the top level directory for the data to be received

 * Start up pantasks and issue the following comands

{{{
  module pantasks.pro
  module receive.pro
  add.database rc_test
  controller host add somehost
  run
}}}

replacing {{{somehost}}} with the name of the hosts to run the pantasks jobs.

Note: These hosts must have internet access to make http requests to the data store and ssh must be set up to allow ssh login without supplying a password (see simtest notes).

Multiple hosts may be specified as usual.


Summary information such as faults and copy/extract times can be found in the receiveResult table of the mySQL database, file names and byte size (and other information) in receiveFile:
{{{
  select * from receiveResult where fault>0;
  select * from receiveResult;
}}}

Example list of files successfully downloaded (fault=0), size, download time, and extraction time:
{{{
SELECT receiveResult.file_id,receiveFile.file,receiveResult.dtime_copy,receiveResult.dtime_extract,receiveFile.bytes 
FROM receiveResult,receiveFile 
WHERE receiveResult.fault=0 and receiveResult.file_id=receiveFile.file_id;
}}}

Is/will there be functionality in ippMonitor to access/monitor this information?

When faults happen, how can the download be redone? This seems to work (is it the proper method?)
{{{
receivetool -dbname rc_test -revert -fault 1
}}}

What if files were added to a -product (is this possible?), how would an update be made?  {{{ -revert -product dist0 }}} ?