DX Platform - Check the "Entropy" settings

Products

• DX Application Performance Management
• DX Operational Intelligence
• CA App Experience Analytics

Issue/Introduction

What are the available options to check and validate the entropy settings in a k8s or openshift setup?

Environment

DX Platform 2x

Resolution

Option 1: Using dxi-entropy-check.sh

https://techdocs.broadcom.com/us/en/ca-enterprise-software/it-operations-management/dx-platform-on-premise/23-3/installing/installation-scenarios-2330/Install-as-a-Namespace-Administrator/Kubernetes---Pre-installation-Tasks-Namespace-Admin.html#concept.dita_8a4a4f5e-a97f-4ec2-854f-5081590e0aac_EntropyCheck 

 

Option 2: Without using dxi-entropy-check.sh

If you are unable to use <dx-platform>/tools/dxi-entropy-check.sh script due to Kubernetes or Openshift user restrictions, alternately, you can use:

cat /proc/sys/kernel/random/entropy_avail

it returns a number that indicates how much "entropy" is available to the kernel

If it returns anything less than 1000, Then we know there is low entropy in the node.

 

IMPORTANT:

If entropy is insufficient pods may stuck in init state or crash (apmservices and acc pods are more sensitive to insufficient entropy)

- to see currently available entropy on the host: cat /proc/sys/kernel/random/entropy_avail

- to make sure that pods will get sufficient amount of entropy it is better to use command like in dxi-entropy-check script: timeout -s ALRM 20 dd if=/dev/random of=/dev/null bs=1024 count=1024 iflag=fullblock;

-  normal output likeo    # timeout -s ALRM 20 dd if=/dev/random of=/dev/null bs=1024 count=1024 iflag=fullblock;o    sh-4.4

$ timeout -s ALRM 20 dd if=/dev/random of=/dev/null bs=1024 count=1024 iflag=fullblock;o    1024+0 records ino    1024+0 records outo    1048576 bytes (1.0 MB, 1.0 MiB) copied, 0.828665 s, 1.3 MB/s

Additional Information

https://knowledge.broadcom.com/external/article/190815/aiops-troubleshooting-common-issues-and.html