Introduction
This document describes the general procedure for Redundancy Configuration Manager (RCM) Health check.
Procedure
These are the mentioned components In RCM:
Controller
Checkpointmgr
Bfdmgr
OPScenter
Configmgr
In response to the issue, identify problematic components for health checks:
rcm show-statistics bfdmgr
------------------------- to check bfdmgr status
rcm show-statistics configmgr
------------------------- to check how many UPs are registered for config push
rcm show-statistics checkpointmgr
------------------------ to check how many UPs are registered for checkpoint sync with their no of instances
rcm show-statistics checkpointmgr-endpoint upfAddr ipv4_address
------------------------ to check how checkpoint status for one particular UP
rcm show-statistics checkpointmgr-endpointstats
------------------------ to check how checkpoint status for overall all the UPs registered with their instance
rcm show-statistics checkpointmgr-session upfAddr ipv4_address
---------------------- to check statistics of checkpoint sync per UP based on its IP address
rcm show-statistics controller
--------------------- to check no of UPs and its states registered with controller
rcm show-statistics switchover
rcm show-statistics switchover-verbose
--------------------- above 2 command to check the UP switchover details
rcm support-summary --------- whole RCM components details would be captured
show running-config --------- To get whole config present in opscenter
Note: Execute all these commands from Opscenter in RCM.
Here is the process to access the Opscenter:
kubectl get svc -o wide -A
By this command you will get all the pods with their IPs and name
ssh -p 2024 admin@<ops-center pod IP>
Here port is 2024 and user is admin for opscenter and IP we can get from svc command
cat /etc/smi/rcm-image-version
To check RCM version which can be directly executed on RCM but verify the file name : “rcm-image-version” once in the same path as file name can be different for different customers
Health Check Commands for Pods
Use these commands for pod-level checks when necessary:
systemctl status kubelet -------------- kubelet statistics
systemctl status host_reboot.service ------------- to check status of Active RCM
kubectl get pods -A ------------- list of pods created in the node
kubectl describe pod <pod name> -n rcm ------------- shows details of each container in this pod
kubectl get svc -n rcm ------------- To get service IP and port of container
ssh -p 2024 admin@<pod IP> ------------ Get into that Pod
kubectl get cm –n rcm ------------ To check different config maps
kubectl logs <pod name> -n rcm ------------- to check pod level logs