'flexera-krm-monitor-0' failed to start pod and throws error 'exit status 1'

[vamshijanagama]

Containers:
monitor:
Container ID: cri-o://c077377fee82f22beff46dce1c48b098afea530f29d5c5cf1d9e58dec707012e
Image: fra.ocir.io/flexera/krm:1.6.2
Image ID: fra.ocir.io/flexera/krm@sha256:ffe8a47ba7067c2edfe5f34fbee9be96eec7738a5a65148e8cb4e7994f72bbb7
Port: 8080/TCP
Host Port: 0/TCP
Args:
monitor
--beacon-url
https://beaconURL/ManageSoftRL/
State: Waiting
Reason: CrashLoopBackOff
Last State: Terminated
Reason: Error
Exit Code: 1
Started: Tue, 19 Sep 2023 08:12:30 +0000
Finished: Tue, 19 Sep 2023 08:12:31 +0000
Ready: False
Restart Count: 1
Environment:
POD_NAME: krm-flexera-krm-monitor-0 (v1:metadata.name)
POD_NAMESPACE: flexera (v1:metadata.namespace)
Mounts:
/var/opt/managesoft/tracker from flexera-krm-data (rw)
/var/run/secrets/kubernetes.io/serviceaccount from kube-api-access-lpqn6 (ro)

And pod log shows

krm 2023-09-19T08:06:55.708400859Z INFO -- start: version 1.6.2
krm 2023-09-19T08:06:56.200599523Z INFO -- main: stopped
exit status 1

How to find what could be the issue. To give an overview.

We are on Oracle cloud and Beacon server outside of Oracle cloud which connects through port through 443 to Oracle cloud. Do we need to define port 443 in yaml?

exit status 1

[vamshijanagama]

Any pointers would really help to move further.

[vamshijanagama]

From Debug log I could see

krm 2023-09-22T17:13:54.6913854Z INFO -- start: version 1.5.3
krm 2023-09-22T17:13:54.691557527Z DEBUG -- main: begin
krm 2023-09-22T17:13:54.691596481Z DEBUG -- main: checking directories
krm 2023-09-22T17:13:54.693091489Z DEBUG -- main: creating directory /var/opt/managesoft/uploads/Inventories
krm 2023-09-22T17:13:54.693200277Z DEBUG -- main: configuring the standard inventory agent components
krm 2023-09-22T17:13:54.693214032Z DEBUG -- config::agent: applying bootstrap configuration
krm 2023-09-22T17:13:54.693219743Z DEBUG -- config::agent: base beacon url: https://xx.xx.xx.xx
krm 2023-09-22T17:13:54.728724228Z DEBUG -- config::agent: looking for patches in /tmp/config
krm 2023-09-22T17:13:54.728749946Z DEBUG -- config::agent: no patch directory found
krm 2023-09-22T17:13:54.728755407Z DEBUG -- config::agent: fetching policy updates from the beacon
krm 2023-09-22T17:13:55.039223415Z INFO -- main: stopped
exit status 1

I masked the url with 'xx'. We have correct URL there.

[vamshijanagama]

To add more from the logs if you say that it fails to reach beacon URL below is the snippet of our configuration

We have Kubernetes cluster with two nodes. 

Nodes are able to communicate with external IPs and pods too

root@krm-flexera-krm-node-2bccq:/# nc -zv -w 5 xx.xx.xx.xx 443
Connection to xx.xx.xx.xx 443 port [tcp/*] succeeded!

Our apps are connected through cluster IP->Kubernetes LB(Which is Internal to) -> Cloud Application LB(External).

[JohnSorensenDK]

@vamshijanagama 

I reached out to one of our SME's internally and here's some links to relevant documentation based on the symptoms you're describing above:

• Required SSL certificates for the agent to communicate with the beacon, please refer to SSL issues with the Flexera Kubernetes inventory agent 
• The krm agent running in the monitor pod will call the Flexera agent mgspolicy module to get the policy from the beacon, please refer to Supporting Custom Certificates for HTTPS 
• and it will try to verify the beacons certificate during the TLS handshake and maybe this is failing, please refer  CheckServerCertificate 

Thanks,