Azure NSG Troubleshooting: Network Security Group Checklist

Define source and destination

Start by defining exactly what is failing, which server is affected and whether the problem affects one client or multiple systems.

Check NSG association

Check the local configuration that controls the operation. Compare it with a known-good server or client where possible.

Review rule priority

Run a direct test for the dependency. Capture the exact result so you can distinguish configuration, connectivity and application failures.

Use effective security rules

Review the relevant server-side configuration and logs. Look for timestamps that match the reported failure.

Check routes

Validate security, permissions and service state before changing production configuration.

Validate the application port

After the fix, repeat the original test and document the working configuration so the procedure can be reused.

Useful commands

Run these checks from an elevated PowerShell session where appropriate. Replace example names with your environment.

Effective NSG

az network nic list-effective-nsg -g RG01 -n VM01NIC

NSG rules

az network nsg rule list -g RG01 --nsg-name NSG01 -o table

Quick troubleshooting path

Use this sequence to isolate the failing dependency before changing production configuration.

Troubleshooting decision path for azure nsg troubleshooting

What good troubleshooting looks like

Good infrastructure troubleshooting is evidence-driven. Capture the original state, test the dependency that can prove or disprove your hypothesis, make the smallest safe change and repeat the original test.

Example workflow
Symptom → hypothesis → direct test → result → controlled change → validation → documentation

Frequently asked questions

What should I check first?

Start with the exact symptom and validate the dependency closest to the failure, such as DNS, a port, a service, storage or an authentication path.

Should I change production configuration immediately?

No. Capture the current state first, test the suspected dependency and make one controlled change at a time.

How should I document the fix?

Record the symptom, commands used, result, configuration change and validation result so the procedure can be repeated.

Related TechRunbook guides

Operational rule: Capture the original symptom, test one dependency at a time, and validate the fix using the same test that exposed the problem.

Need more infrastructure resources?

Browse free TechRunbook guides, checklists and scripts.

Browse free IT resources →