Domain health guide
A practical domain monitoring checklist
Domain reliability depends on several public systems changing together. A practical monitoring routine gives DNS, web, SSL, and email changes an owner before customers notice them.
Track the public dependencies
Start with the domain names that matter to customers, including primary websites, key marketing sites, and domains used for sending mail. Record the owner for DNS, web delivery, certificate renewal, and email configuration.
Do not assume every service uses the same provider or change process. A DNS update, CDN migration, certificate renewal, and email provider change can have different owners and different failure modes.
Review the signals that fail together
DNS resolution is foundational: if it fails, website and certificate checks may be blocked. HTTPS availability and certificate validation help indicate whether visitors can reach a secure endpoint.
For email, review SPF, DKIM, and DMARC together. A record can exist without being the intended policy, so use the public signals as prompts for a controlled configuration review.
- List customer-facing domains and their operational owners.
- Check public DNS resolution after DNS, CDN, hosting, or registrar changes.
- Check HTTPS and certificate validation after certificate or routing changes.
- Review SPF, DKIM, and DMARC after email-provider or DNS changes.
- Assign a response owner and record the change that triggered the check.
- Use continuous monitoring where manual checks leave long gaps between changes.
Turn findings into an operating routine
A finding is most useful when it has a clear next owner. Record whether the issue belongs to a DNS provider, website platform, email provider, or internal team, then verify the public result after remediation.
CloudSpex prioritizes available public findings and can retain monitoring history for managed domains. A one-time public scan shows the current state and should be repeated after changes.
Avoid false certainty
Public monitoring cannot see private infrastructure, internal approval workflows, or every application dependency. A clean result should not be read as a complete security assessment.
Use the checklist to improve repeatability, then combine it with your change management and incident process.
Why domains fail without anyone changing them
Most domain incidents are not caused by an attack or even by a mistake made that day. They come from things that expire or drift: a certificate reaches its end date, a registration lapses because the card on file was replaced, a provider changes the addresses behind an SPF include, a site is deleted and its DNS record is left behind. Nothing is edited on the day of the failure, so nobody is looking.
A monitoring checklist therefore has two jobs. It has to notice change, because unexpected change is the earliest sign of trouble, and it has to notice the approach of a date, because expiry is predictable and still the most common cause of outages.
Registration and nameservers
Everything else depends on the registration. If the domain expires or its nameservers are changed, the website, email and every subdomain stop together.
- The registration expiry date, with warnings at sixty, thirty and seven days.
- Whether auto-renew is on and the payment method is current.
- The registrar lock status, which prevents unauthorised transfers.
- The list of nameservers. Any change that you did not make is an incident until proven otherwise.
- The contact email on the registration, which is where renewal and transfer notices go.
DNS records
Record monitoring is about comparing the present with the last known good state. Keep the previous value of every record you watch, because the question during an incident is always what changed and when.
- A and AAAA records for the website and important subdomains.
- MX records, which decide where mail is delivered.
- NS records, compared between the registrar and the zone.
- TXT records that carry SPF, DMARC and domain verification values.
- CNAME records, including whether their targets still resolve.
- CAA records, which control who may issue certificates.
- DNSSEC status, if the domain is signed: the DS record and signature validity.
Certificates and HTTPS
Check certificates from outside, on every hostname that serves one, and check more than the date.
- Days until expiry of the certificate actually served.
- Whether the hostname is covered by the certificate.
- Whether the full chain is sent.
- Whether HTTP redirects to HTTPS and the final page loads.
- A change of issuer, which may be planned or may be a sign that something else is now answering for the host.
Email authentication
Email problems are the slowest to be noticed, because the sender sees nothing. Messages are filtered or rejected at the recipient's side and the first report comes from a customer who did not get an invoice.
- SPF: exactly one record, valid syntax, and a lookup count within the limit of ten.
- DKIM: a valid key at each selector your senders use.
- DMARC: one record, the current policy, and a working report address.
- DMARC aggregate reports: new sending sources and rising failure counts.
- MTA-STS and TLS-RPT, where published: the policy file is reachable and matches the MX records.
dig example.com TXT +short | grep spf1
dig _dmarc.example.com TXT +short
dig s1._domainkey.example.com TXT +shortSubdomains and leftovers
Subdomains are created for projects and outlive them. A record that still points at a deleted cloud resource can be claimed by someone else, who then serves content on your name. Review the full list of records on a schedule, and remove any whose target no longer exists or belongs to you. Pay particular attention to CNAME records that point to hosting platforms, storage services and marketing tools.
How often to check
Frequency should follow how quickly a problem causes damage and how quickly the thing can change.
- Website availability and HTTPS: every few minutes if the site earns money, otherwise hourly.
- DNS records and nameservers: at least daily. Changes are rare, and each one matters.
- Certificates: daily. Expiry is measured in days, and a daily check gives a clear countdown.
- Email authentication records: daily.
- DMARC aggregate reports: as they arrive, reviewed weekly.
- Registration expiry: daily, with warnings well ahead.
- A full review of all records and subdomains: quarterly.
Deciding what deserves an alert
A monitoring setup that sends too many alerts is ignored within a month. Separate the findings that need a person now from those that belong in a weekly summary.
Alert immediately on a change to nameservers or MX records, a certificate with fewer than fourteen days left, a DMARC or SPF record that has become invalid or disappeared, and a site that is down. Put in the summary things that are worth knowing and not urgent: a policy still at p=none, a missing security header, a certificate that is thirty days from expiry and renews automatically.
Alert once when a condition starts and once when it clears. Repeating the same alert every hour teaches people to mute the channel.
Who owns each finding
A finding with no owner is not fixed. For each domain, record who controls the registrar account, who controls DNS, who manages the mail provider and who runs the website hosting. In small organisations these are often four different people or companies, and some of them are former suppliers.
Store the access details somewhere the organisation controls, not in one person's mailbox. A large share of recovery time in real incidents is spent finding out who can log in to the registrar.
A weekly and a quarterly routine
Continuous checks catch failures. A short routine catches the slow drift that no single alert reports.
- Weekly: read the summary, clear or assign every open finding, and look at the DMARC report for new senders.
- Monthly: confirm that alerts still reach a person by checking who received the last one.
- Quarterly: export all DNS records and remove what is no longer used.
- Quarterly: confirm registrar access, auto-renew and the payment method.
- After any change of supplier: review nameservers, MX, SPF includes and DKIM selectors, and remove the old supplier's records.
What public monitoring covers and what it does not
Everything on this checklist can be observed from the public internet, which is why it can be monitored without access to your systems. That is also its limit. External monitoring does not see internal services, application errors behind a login, the state of backups or whether a server is patched. It tells you that the public face of the domain is intact and warns you when it changes. CloudSpex runs these public checks on a schedule, keeps the history of each result and reports a check as not assessed when it could not be measured, rather than counting it as a pass or a failure.
Frequently asked questions
How often should I check a domain?
Check after relevant changes and use continuous monitoring when important domains can change between manual reviews.
Does a healthy score mean nothing can fail?
No. A public scan is a limited view of the signals it can safely assess at that time.
Can CloudSpex monitor every subdomain?
The public scan has limited subdomain-risk coverage. Managed monitoring scope should be confirmed in the product dashboard.
What is the single most important thing to monitor?
The registration expiry and the nameservers. If either fails, the website and email fail with it, and recovery is the slowest.
Do I need monitoring if my hosting provider handles everything?
The provider monitors its own platform. It does not watch your registrar, your mail authentication records or records you created elsewhere, and it will not tell you when a DNS record points at something that no longer exists.
How many domains make monitoring worth automating?
Even one, if email or revenue depends on it. Manual checks are done carefully for the first month and then forgotten.
Use public checks as a starting point
CloudSpex reads limited public-facing signals. It does not change DNS, website, certificate, or email settings. Confirm the responsible provider and test relevant changes before applying a remediation.
Editorial details
Author and reviewer: CloudSpex. Sources: public technical standards and provider documentation relevant to the topic. Last reviewed: November 10, 2026. We update guides when the product behavior or the underlying guidance changes.