Record the exact query and resolver
Capture the requested name, record type, source location, configured resolver, time and returned result. A timeout, a negative response and an unexpected address describe different observations. Preserve the exact result before changing caches or settings.
Determine whether the issue affects one client, one resolver or several locations. Microsoft’s DNS guidance distinguishes client and server investigation; the first useful step is identifying which system answered or failed to answer the request.
Compare the expected answer with observed behavior
Use a fully specified name where appropriate and document any suffix or split-DNS expectation. Compare observations from the affected location with an authorized working comparison.
| Observation | Next question |
|---|---|
| No response | Can the client reach the intended resolver for the required DNS traffic? |
| Negative response | Does the relevant zone contain the expected record? |
| Unexpected address | Is the answer consistent with the intended DNS view and record state? |
| Correct DNS answer | Does the actual application transaction succeed afterward? |
| Different answers by location | Is the difference intended or a configuration discrepancy? |
Illustrative example: one location receives an old destination
An internal application has moved, and one location continues to receive an unexpected address. The workflow records the resolver in use, response, timing and expected record source, then compares a working location.
The engineer uses the evidence to distinguish cache behavior, an unintended resolver or a record discrepancy. The workflow does not assume that every difference is a fault: separate DNS views may be intentional. The example contains no customer data.
Configure the investigation around approved checks
Use NetPrompt’s supported server access or validated connectors for the systems in scope. Review platform-specific commands and permissions before the first run. Network Context should describe the resolvers and expected service names; Troubleshooting Context should define the interpretation and escalation rules.
A DNS query and a DNS configuration change have different operational consequences. Cache clearing, forwarder changes and record edits need the applicable review and approval. This page describes a configurable diagnostic workflow, not a built-in integration for every DNS platform.
Verify resolution and the service separately
After an approved correction, repeat the same query against the relevant resolver and confirm the representative application transaction. Record any expected propagation or cache behavior rather than declaring universal recovery from one response.
Evaluate a healthy name, an intentionally absent record and an unavailable resolver. The output should distinguish their results, name the relevant evidence and identify any checks that could not run.
Common questions
- Does a successful DNS response prove the application is healthy?
- No. It establishes a response to that query. Verify the application connection, authentication and transaction separately as relevant.
- Should the workflow always flush DNS caches?
- No. Preserve evidence and identify the relevant client or server cache first. A cache operation should follow the approved diagnostic procedure.
- Can NetPrompt troubleshoot every managed DNS service?
- Coverage depends on supported host access or validated API operations and permissions. Confirm the required checks for your DNS environment.
Continue exploring
Further technical reading
External references explain general practices; they are not endorsements of NetPrompt AI.