Cluster Fails Starting cb-datagrid Service
search cancel

Cluster Fails Starting cb-datagrid Service

book

Article ID: 285236

calendar_today

Updated On:

Products

Carbon Black EDR (formerly Cb Response)

Issue/Introduction

  • One or more node's cb-datagrid service fails to startup in the cluster.

Environment

  • EDR Server: All Supported Versions

Cause

Datagrid cannot communicate without Postgres

Resolution

  1. Verify the cb-pgsql service is running on the primary node.
    • service cb-pgsql status
  2. Confirm network communication between minion nodes and primary node, including port 5002, and 5701 are successful.
  3. Ensure the cb-datagrid service isn't stopping on one of the minions.
  4. If networking appears correct, this may be an issue with DNS. An option is to update DatabaseURL in /etc/cb/cb.conf to IP on all nodes and try starting services

Additional Information

  • During cluster startup, the primary node's cb-datagrid service will wait for all other minions to start first
  • Datagrid relies on the primary node's cb-pgsql  to be up and accessible across the cluster
  • All nodes pull information from the cb-pgsql service on the primary node to successfully start cb-datagrid
  • If step 4 corrects the issue, please reach out to your network engineer to check on the DNS servers listed in /etc/resolv.conf