Dynamic DNS updates are not retried in case of SERVFAIL
Describe the bug When testing Dynamic DNS setups, Kea DHCP doesn't retry SERVFAIL or other failure modes relating to UPDATE after a lease is created even if a new DHCPREQUEST/ACK handshack is forced on the client. Only be deleting the lease database and then rerunning a client got the update filed.
To Reproduce Steps to reproduce the behavior:
- Configure Kea to use DDNS updates via TSIG to BIND
- Cause a SERVFAIL or other update failure condition (i.e., wrong key)
- Attach client that is NOT in the lease table to get a lease
- Confirm that Kea tried and failed to do DDNS
- Fix the SERVFAIL condition
- Renew the lease on the client, DDNS is not updated.
Expected behavior DDNS is retried on updates with client renewal should the SERVFAIL
Environment:
- Kea version: 2.2.0, compiled from tarball on site
- Ubuntu 22.04.1
Additional Information This was done as part of a livestream learning how to use Kea documenting the failure in real time.Llinks and timestamps available on request.
Contacting you GitLab is fine, can provide more ways if needed.