Get Instant AI Troubleshooting Support: Your Quick Roadmap
Getting iMaintain AI up and running can feel like juggling spanners on a moving line. If you hit a snag, you’re not alone. In this guide, we unpack the top installation hiccups and offer clear fixes. Whether it’s a missing image pull or a secret that won’t create, we’ll point you in the right direction. For full AI troubleshooting support, tap into the expertise built into iMaintain. Access AI troubleshooting support with iMaintain
We’ll cover everything from cluster resource checks to RBAC policies. You’ll also find best practice tips so your team avoids repeat errors. Ready to stabilise your maintenance intelligence platform? Let’s dive in.
Common Installation Issues and How to Diagnose Them
Setting up iMaintain’s maintenance intelligence platform is usually smooth. But even small hiccups can stall your project. Below, we walk through the most frequent problems and how to troubleshoot them.
1. Operator Image Retrieval Failure
Problem: During setup, the AI operator image doesn’t pull from the registry.
Diagnosis:
– Look for “Failure to pull from quay” or similar in the cluster events.
– Confirm network access and registry availability.
– Check that your cluster nodes are running and healthy.
Resolution:
– Restart the image registry service or fix network routes.
– If you still see pull errors, gather logs with the must-gather tool and reach out to support.
2. Insufficient Cluster Resources
Problem: Installation stops with a “prerequisites not met” message.
Diagnosis:
– In Red Hat OpenShift Cluster Manager, open your cluster’s Add-ons tab.
– Click Configure Red Hat OpenShift AI, then check the Prerequisites tab.
– Note any warnings about low CPU, memory or machine pool size.
Resolution:
– Add nodes or scale up your existing machines.
– Coordinate with your infrastructure team to allocate extra resources.
– Once capacity is restored, rerun the install.
Need more detail on cluster sizing? Learn how the platform works
3. Unsupported Infrastructure
Problem: The operator detects an environment it doesn’t recognise.
Diagnosis:
– Switch to Administrator in the OpenShift console.
– Filter pods by errors in All Projects or redhat-ods-operator.
– Check logs for “ERROR: Deploying on $infrastructure, which is not supported.”
Resolution:
– Verify your environment against the supported configurations guide.
– Migrate to a supported platform or consult your hosting provider for upgrades.
4. Custom Resource (CR) Creation Failures
Problem: The AI or Notebooks CR never appears.
Diagnosis:
– In the Workloads → Pods view, find the rhods-operator pod with errors.
– Open its Logs and search for:
– “ERROR: Attempt to create the ODH CR failed.”
– “ERROR: Attempt to create the RHODS Notebooks CR failed.”
Resolution:
– If it’s a transient glitch, restarting the operator pod may help.
– Persistent failures? Capture logs and share them with the support team.
At this point, it’s wise to pause and ensure your underlying platform is solid. Discover AI troubleshooting support powered by iMaintain
Dashboard Access and RBAC Glitches
Once CRs are in place, you should see the AI dashboard. If it’s unreachable or items aren’t loading, follow these steps.
5. Dashboard Not Accessible
Problem: The redhat-ods-applications, redhat-ods-monitoring or redhat-ods-operator namespaces are Active, but the dashboard errors out.
Diagnosis:
– In All Projects, filter pods by non-Running/Completed statuses.
– Click on the problematic pod’s Status link or open its Logs.
Resolution:
– Identify the specific error in the logs.
– Restart the failing pod or check for missing dependencies.
– If the dashboard still fails, consult the must-gather output and raise a support ticket.
6. RBAC Policy Creation Issues
Problem: The dedicated-admins group policy won’t create.
Diagnosis:
– Similar to CR checks, inspect rhods-operator pod logs for:
– “ERROR: Attempt to create the RBAC policy for dedicated admins group in $target_project failed.”
Resolution:
– Verify your cluster’s role-based access control settings.
– Ensure you have cluster-admin privileges during installation.
– Adjust your security policies or ask your OpenShift admin to grant the right roles.
Ghost Secrets: Automation Fails to Create Keys
Under the hood, iMaintain’s automation creates secrets for monitoring and alerts. If those vanish, installation can break.
7. Dead Man’s Snitch, PagerDuty, SMTP and Parameter Secrets
Problem: Secrets for the Dead Man’s Snitch operator, PagerDuty, SMTP or ODH parameters don’t appear.
Diagnosis:
– Switch to Administrator and watch for errors like:
– “ERROR: Dead Man Snitch secret does not exist.”
– “ERROR: Pagerduty secret does not exist.”
– “ERROR: SMTP secret does not exist.”
– “ERROR: Addon managed odh parameter secret does not exist.”
Resolution:
– Confirm your Managed Tenants SRE automation is configured correctly.
– Manually create the missing secret via oc create secret commands if urgent.
– Otherwise, update your SRE process and rerun the reconciliation.
Installing secrets smoothly sets up health checks and alerts. Once you’ve nailed these steps, your AI operator should stay healthy.
Best Practices to Avoid Repeat Failures
Nothing beats prevention. Here are some quick wins to keep your setup robust:
- Document every change.
- Label clusters and nodes consistently.
- Automate must-gather snapshots on failures.
- Train your maintenance team on operator basics.
Keeping these tips in your playbook means fewer calls for AI troubleshooting support.
Feeling ready to level up your maintenance workflow? Schedule a demo with our team and see how iMaintain captures your engineering know-how, prevents repeat faults and builds a shared intelligence layer.
Wrapping Up
Installing a sophisticated platform like iMaintain AI needn’t be a headache. By walking through image pulls, resources checks, CR creation, dashboard access, RBAC policies and secret generation, you’ve got a clear action plan. This isn’t just about fixing errors—it’s about empowering your engineers with a reliable intelligence engine.
Remember, solid foundations lead to predictable outcomes. With these steps in hand, you can minimise downtime, retain critical knowledge and truly move from reactive fixes to proactive maintenance.
For any lingering questions or tailored advice, don’t hesitate to reach out. Talk to a maintenance expert
What People Are Saying
“We struggled with image pull errors for days. The step-by-step checks in this guide helped us pinpoint a network firewall issue. Now our iMaintain AI operator is rock solid.”
— Emily S., Maintenance Manager
“The CR creation diagnostics spared us from guessing. We fixed a permissions mix-up and got our notebooks running in under an hour.”
— Raj P., Reliability Lead
“Thanks to the secret-creation tips, our monitoring pipelines came online without a hitch. The platform’s human-centred AI is spot on.”
— Lisa T., Operations Supervisor