How to collect logs when self hosted

Many of our customers choose to run Cortex application in a self hosted manner. At times backend application logs can be needed to facilitate troubleshooting. To ease collection and packaging of these logs, we offer a script called Brain Dump, which will collect and package the following information:

1. helm-manifest.yaml, a YAML representation of the Kubernetes resources that were generated for the release's chart(s).
2. pods.json, a JSON representation for the list of all Kubernetes pods in the Cortex namespace with respective details.
3. backend-logs.log, recent backend application logs associated with the Cortex deployment.

Current Instructions

1. Navigate to the appropriate directory.
2. Download braindump.sh from Brain Dump.
3. Using your favorite text editor, edit braindump.sh to make the appropriate edits to reflect your environment on the following variables: HELM_MANIFESTand CORTEX_BACKEND_DEPLOYMENT.
4. Make the shell script executable using chmod u+x braindump.sh.
5. Execute the script ./braindump.sh.
6. Upload the tarball file named braindump.tar.gz to your ticket for further analysis.

Pilot Log Collection Utility

Instructions can be found here, which are using troubleshoot.sh Open Source project. This utility will generate a support bundle that can be uploaded to your ticket.