Issue Summary

SERVFAIL/network error responses from Custom Domain Email Notification and Verifying a domain to manage accounts leads to validation failures, and are likely due to issues/timeouts when a DNS lookup is performed on a customer’s name servers against Atlassian's DNS.
Reasons can range from unknown change in the customer’s configuration to issues with their domain name servers themselves.

Suggested solution

Retry on SERVFAIL/network error response.

A follow-up would be to include a secondary, fallback, DNS resolution on Atlassian's side for intermittent customer DNS server responses.

Why this is important

For Email domain verification: Customer are put in circles figuring out why their once working DMARC emails are failing validations. It should limit impact internally as well as make customer lives easier

Domain verification for Managed accounts: We notice that the Domain verification does not succeed even after 72 hours of DNS change and even though we can verify via "dig" or "nslookup" that the correct txt records with token has been updated. This is because there have not been any retries since the first validation check.

What customers can do for now:

For Email domain verification:

• Reach out to their domain provider to check if there have been any changes on the domain provider's end as well as validate to customer's own configuration
• Rerun "Run DNS record check" in the UI in the DNS record modal to re-check/validate (this is manual retry, in a sense)
• Confirm customer's are not blocking Atlassian's DNS recursors on their nameservers: (52.202.16.180 or 52.203.230.183)

Domain verification for managed accounts:
Navigate to the Org, click on Settings > Domains. For the domain wherein verification shows up as failed, click on "Verify domain" again. This will trigger another validation and the domain should be verified.