Maximo Open Forum

Hello,

I'm installing MAS (tried 9.1 and 9.2) on a SNO (4.18.43) server in Azure via mas cli and running into an error when the rhn-data-service-0 pod starts which makes the install fail.

Andy,

I have seen something similar previously and in most cases it has to do with the Entitlement Key. The issue sometimes has been when copy/pasting the key for example if using the MASCLI an extra space may get added or I had a scenario when the user provided the key they added a carriage return in the text file versus enabling word wrap. I forget which exact secret you need to look at but in OpenShift Workload --> Secrets look for ibm-entitlement and check that the key has not extra space between the quotes or a carriage return. When viewing use the "Reveal Value" as the YAML is encrypted.

Solution Path A: If Using Automated Helper Utilities (mas-cli)

# Force operational metrics to rely strictly on Azure Disk Block Storage 
--storage-rwo managed-csi \
--storage-rwx azurefile-csi

Solution Path B: Direct azurefile-csi StorageClass Customization

⚠️ Disclaimer: Please validate this modification in a sandbox or staging context before applying it universally to standard multi-node production storage manifests. Bypassing volume permission validations entirely (noperm) can alter security expectations depending on tenant multi-occupancy structures.

# 1. Patch the storage class with required CIFS execution overrides
oc patch storageclass azurefile-csi --type='json' -p='[{"op": "add", "path": "/mountOptions", "value": ["noperm", "mfsymlinks", "actimeo=30"]}]'

# 2. Cycle the crashing metrics collector pod to bind with new mount policies
oc delete pod -l app=rhm-data-service -n ibm-software-central

What this script alters under the hood:

apiVersion: storage.k8s.io/v1
kind: StorageClass
metadata:
  name: azurefile-csi
provisioner: ://azure.com
mountOptions:
  - noperm      # Tells the mount layer to skip remote client chmod/chown verification
  - mfsymlinks  # Retains support for Unix symbolic links on the file share
  - actimeo=30  # Adjusts attribute caching timers for optimized cluster communication

Hi Frederick,

Thanks for the great reply. I tried modifying the storageclass but openshift wouldn't let me. I installed a new SNO and tried with the same results. I then tried deleting the storageclass to then recreate it but it wouldn't let me do that either. Using the interactive install with MAS CLI I chose those storage classes for RWO and RWX and that worked. I had tried both with managed-csi previously but it doesn't like that for the RWX. 

Also, thanks for including those links at the bottom. They were very helpful in learning more about the install process.

# Force operational metrics to rely strictly on Azure Disk Block Storage --storage-rwo managed-csi \--storage-rwx azurefile-csi

# 1. Patch the storage class with required CIFS execution overridesoc patch storageclass azurefile-csi --type='json' -p='[{"op": "add", "path": "/mountOptions", "value": ["noperm", "mfsymlinks", "actimeo=30"]}]'# 2. Cycle the crashing metrics collector pod to bind with new mount policiesoc delete pod -l app=rhm-data-service -n ibm-software-central

apiVersion: storage.k8s.io/v1kind: StorageClassmetadata:  name: azurefile-csiprovisioner: ://azure.commountOptions:  - noperm      # Tells the mount layer to skip remote client chmod/chown verification  - mfsymlinks  # Retains support for Unix symbolic links on the file share  - actimeo=30  # Adjusts attribute caching timers for optimized cluster communication

Hi Fredrick,

That looks very cool. https://ibm-mas.github.io/maf-local-dev-mode/index.html

I don't think I have the time to dive into that right now though I will add it to my list of things to explore. I did get 9.2 up and running. I learned a bunch from troubleshooting in openshift but had more success just starting the install from scratch after errors. It's frustrating that any DB configuration issues don't show up until like an hour into the install. 

Thanks again for the help.

# Force operational metrics to rely strictly on Azure Disk Block Storage --storage-rwo managed-csi \--storage-rwx azurefile-csi

# 1. Patch the storage class with required CIFS execution overridesoc patch storageclass azurefile-csi --type='json' -p='[{"op": "add", "path": "/mountOptions", "value": ["noperm", "mfsymlinks", "actimeo=30"]}]'# 2. Cycle the crashing metrics collector pod to bind with new mount policiesoc delete pod -l app=rhm-data-service -n ibm-software-central

apiVersion: storage.k8s.io/v1kind: StorageClassmetadata:  name: azurefile-csiprovisioner: ://azure.commountOptions:  - noperm      # Tells the mount layer to skip remote client chmod/chown verification  - mfsymlinks  # Retains support for Unix symbolic links on the file share  - actimeo=30  # Adjusts attribute caching timers for optimized cluster communication

If you're going to respond with AI - please fact check before you send people down rabbit holes. Your response makes me think you may have never seen the inside of an Openshift cluster.

This line is my favorite: 

oc delete pod -l app=rhm-data-service -n ibm-software-central

Please use AI responsibly - people are here to learn and ask questions they genuinely don't know the answer to.

Thanks in advance for your understanding 😀

# Force operational metrics to rely strictly on Azure Disk Block Storage --storage-rwo managed-csi \--storage-rwx azurefile-csi

# 1. Patch the storage class with required CIFS execution overridesoc patch storageclass azurefile-csi --type='json' -p='[{"op": "add", "path": "/mountOptions", "value": ["noperm", "mfsymlinks", "actimeo=30"]}]'# 2. Cycle the crashing metrics collector pod to bind with new mount policiesoc delete pod -l app=rhm-data-service -n ibm-software-central

apiVersion: storage.k8s.io/v1kind: StorageClassmetadata:  name: azurefile-csiprovisioner: ://azure.commountOptions:  - noperm      # Tells the mount layer to skip remote client chmod/chown verification  - mfsymlinks  # Retains support for Unix symbolic links on the file share  - actimeo=30  # Adjusts attribute caching timers for optimized cluster communication

Andy,

I'm not an Azure user, and I don't want to distract you from your current path. However, I can tell you that for SNO, you don't need RWX. MAS with Manage on SNO will work with RWO. Using the default storage class, lvms-vg1 (which uses the attached secondary VM disk), is most often sufficient for an SNO deployment.

Hi Brett and Andy,

Thank you for the call-out, Brett; you are 100% correct to flag those errors. I owe the community a sincere apology for the error in my response.

To be completely transparent, I have been building a GenAI A2A Agentic question-and-answer orchestration / AI Maximo Knowledge Base. The product's prototype automatically ingested and answered Andy's question from my local project workspace documentation (which I used to configure a local MAS 8 instance). This question was automatically answered as I was answering a different question. 

The sources of the answers are the IBM MAS CLI Repository and the MAS DevOps Ansible Collection Documentation. The error you called out suggests an alternative Azure solution, an AI hallucination you appropriately identified. 

After an answer was posted without my awareness, I disabled live publishing to all endpoints until I implement a strict manual approval gate in the solution, as soon as I discovered that this question had already been answered.

Appreciate you keeping the forum responsible, Brett. Take a look at my contributions; I am mindful and provide validation resources with my answers on this platform. I only use AI to polish my responses, making them less verbose and more succinct.

Best,