DNS Failover

[philbrookes]

Is your feature request related to a problem? Please describe. There is no supported manner in Kuadrant to migrate an existing published set of DNS records away from a failed set of infrastructure - as the DNS Operators that created those records are no longer accessible to remove them.

Describe the solution you'd like DNS failover in Kuadrant from an existing primary site(s) to an existing passive site(s). The process is manually triggered, but automatically enacted by the DNS Operator. In the case of a catastrophic failure in the primary site, a sysadmin can do the equivalent of flipping a switch to have the DNS Operator write the passive sites records to the zone and remove all of the primary sites records from the zone, with status and metrics to monitor the progress.

Sprint 1 (2 weeks Nov 03 - 17)

DNS Operator - add group concept

• [x] Kuadrant/dns-operator#626
• [x] Kuadrant/dns-operator#618
• [ ] Kuadrant/dns-operator#627

Figure out cross-owner deletion

• [ ] Kuadrant/dns-operator#620

Add CLI Commands

• [x] Kuadrant/dns-operator#621
• [ ] Kuadrant/dns-operator#622
• [ ] Kuadrant/dns-operator#625
• [x] Kuadrant/dns-operator#624
• [x] Kuadrant/dns-operator#623

Sprint 2 (3 weeks Nov 17 - Dec 08)

• [ ] Kuadrant/dns-operator#637
• [ ] Kuadrant/dns-operator#638
• [x] Kuadrant/dns-operator#641

Sprint 3 (3 weeks Dec 08 - Dec 29)

• [ ] DNS Operator - if active, delete records only owned by inactive groups
• [ ] DNS Operator - if active, delete targets only owned by inactive groups
• [ ] DNS Operator - if active, delete registry entries for inactive groups, once no records or targets are left.

Sprint 4 (3 weeks Dec 29 - Jan 19)

• [ ] Kuadrant/kuadrant-operator#1648

Sprint 5 (3 weeks Jan 19 - Feb 09)

• Buffer.
• [SPIKE] Investigate useful group metrics.