Closed gabyval closed 1 month ago
Hi @durera, we are still blocked with this error, could someone help us?
Discussing in Slack ...
We implemented MAS on HCP with PrivateLink. See no issues during setup. We ensured before suite config, the public accessible routes were in place using LetsEncrypt Certs for MAS Core. There are intermittent errors wrt licensemediator pod while accessing MAS URL Administration option. Shall provide more details on that.
FYI - slack thread https://ibm-ai-apps.slack.com/archives/CNDB5GZ3P/p1714684321087009
@whoiscnu I reattempted the installation on a clean cluster (refer to https://github.ibm.com/sit/cloudpak-storage-test/issues/1110#issuecomment-81431800)
The installation failed at the same spot. "suite verify". I think the reason it failed is the secret arft-sls-cfg
was never creaed and the pod arft-catalogmgr
is failing to mount it...
Hello ive found this issue without using HCPs, the cluster is behind a firewall:
cluster: https://console-openshift-console.apps.qa-hcp-sixpack.apps.blazehub01.rtp.raleigh.ibm.com/dashboards
suite-verify step failed i found this message:
message: 'SLS client registration was unsuccessful: Unable to register SLS client for MAS: An unhandled error was returned from SLS: Unable to register SLS client newqa-c11578da: An unhandled error was returned from SLS: HTTPSConnectionPool(host=''sls.ibm-sls.ibm-sls.apps.qa-hcp-sixpack.apps.blazehub01.rtp.raleigh.ibm.com'', port=443): Max retries exceeded with url: /api/registrations (Caused by SSLError(SSLCertVerificationError(1, ''[SSL: CERTIFICATE_VERIFY_FAILED] certificate verify failed: self signed certificate in certificate chain (_ssl.c:1129)'')))'
For what it's worth, we are successfully installing on HCP. Notably, we inject a ClusterIssuer resource named prod-route53-issuer
in the cert-manager
namespace (via hive syncset) and run with export MAS_CLUSTER_ISSUER=prod-route53-issuer
Resolved this via e-mail & Slack, it was a mis-configuration in the install/environment; there's nothing specific to HCP preventing MAS installation on this flavour of AWS, we have successfully verified MAS install on HCP.
Problem Description: Installation of Maximo Core fails in an HCP(Hosted Control Plane) in phase suite-verify of the pipeline
Message:
Reproducibility: Consistent
Additional info:
Seems there might be a problem from the sls-api-licensing-5594b88bc8-cz5qv container. There are couple errors showing Startup probe failed Also, from inside the container, it is showing the following response:
Trying to use curl with the external hostname:
From the verbose, the following errors are presented: