Initializing or Re-initializing of a server/cluster node via cbinit fails with [Errno 20] Not a directory

Products

• Carbon Black EDR (formerly Cb Response)

Issue/Introduction

This is an example of one error seen, data directories and error codes can vary

NotADirectoryError: [Errno 20] Not a directory: '/var/cb/data/solr6/configsets/cbevents_v0/conf/schema.xml
File "/usr/share/cb/virtualenv/lib/python3.8/site-packages/cb/utils/cli_cmd.py", line 165, in decorator
File "/usr/share/cb/virtualenv/lib/python3.8/site-packages/cb/setup/initcb.py", line 694, in main
File "/usr/share/cb/virtualenv/lib/python3.8/site-packages/cb/setup/initcb.py", line 244, in query_for_storage_volume
File "/usr/share/cb/virtualenv/lib/python3.8/site-packages/cb/setup/setup_config.py", line 160, in validate_storage_path
File "/usr/share/cb/virtualenv/lib/python3.8/site-packages/cb/setup/utils.py", line 75, in is_directory_empty_of_files
File "/usr/share/cb/virtualenv/lib/python3.8/site-packages/cb/setup/utils.py", line 75, in is_directory_empty_of_files
File "/usr/share/cb/virtualenv/lib/python3.8/site-packages/cb/setup/utils.py", line 75, in is_directory_empty_of_files
[Previous line repeated 2 more times]
File "/usr/share/cb/virtualenv/lib/python3.8/site-packages/cb/setup/utils.py", line 74, in is_directory_empty_of_files
NotADirectoryError: [Errno 20] Not a directory: '/var/cb/data/solr6/configsets/cbevents_v0/conf/schema.xml

 

Environment

• EDR Server: All supported versions

Cause

Data is present in the directory(s) indicated in the error above causing initialization to fail, data locations and error codes can differ from example

Resolution

CAUTION: Using the new installation procedure on an existing server will likely result in loss of all data, including the configuration and event data collected from sensors.

1. Empty the contents of the directories indicated in the error message 
2. Run initialization process per Server Cluster Management Guide

Additional Information

• Running 'cbinit' will reset a node that has been in production back to factory settings resulting in possible data and configuration settings being lost
• Ensure a complete backup has been performed before removing data and re-initializing the server/cluster node