CA Spectrum Scheduled Discovery fails to run and is showing a java.io.Filenotfoundexception exception in the history

Document ID : KB000031380
Last Modified Date : 14/02/2018
Show Technical Document Details

Issue:  

When implementing a scheduled discovery that uses an import file to populate the ips/hostnames, the scheduled discovery does not launch, and the following is seen in the history of the discovery:

Failed to read file: /usr/Spectrum/IPlist.txt

java.io.FileNotFoundException: /usr/Spectrum/IPlist.txt (No such file or directory)

Cause:  

This is due to having multiple OneClick servers.  When you create the scheduled discovery that uses the import file, and you specify that the import file be pulled from the webserver, the configuration is set on the AdiscConfig model within the SpectroSERVER database.  When the time comes for the discovery to launch, the SpectroSERVER initiates the discovery, and the first OneClick to process and lock the discovery thread is the one that will look locally for the file.  If you have the file on only one webserver, and a different webserver locks the thread, the error shows since the file does not exist on the other webserver.

Solution:  

The solution is to either put the import file on all OneClick Webservers (in the same directory that was specified in the autodiscovery config) or use a mount/shared drive so that all OneClick webservers can access it.  This way, no matter which OneClick locks the discovery thread, the import file will be used.