dnsrecord condition only propagate owned records status #597
Conversation
philbrookes
force-pushed
the
dns-policy-status-propagation
branch
4 times, most recently
from
d98bd7d
to
d45b53f
Compare
philbrookes
force-pushed
the
dns-policy-status-propagation
branch
from
d45b53f
to
e3a358b
Compare
| @@ -133,7 +133,7 @@ type DNSPolicyStatus struct { | |||
| HealthCheck *HealthCheckStatus `json:"healthCheck,omitempty"` | |||
|
|
|||
| // +optional | |||
| ProbeConditions map[string][]metav1.Condition `json:"probeConditions,omitempty"` | |||
| RecordConditions map[string][]metav1.Condition `json:"recordConditions,omitempty"` | |||
There was a problem hiding this comment.
Makes more sense than probeConditions. I do wonder how we intend to do this for other policies though, i had expected a more policy generic way of reporting status about policy sub resources.
| if reference.Controller != nil && *reference.Controller && reference.Name == dnsPolicy.Name && reference.UID == dnsPolicy.UID { | ||
| controlledRecords.Items = append(controlledRecords.Items, record) | ||
| } | ||
| } |
There was a problem hiding this comment.
What's the significance of it being a controller reference or not? This also differs from how we find related sub resources elsewhere, wondering which approach is more desirable, not sure if we use the owner ref in other polices or not.
There was a problem hiding this comment.
I just copied this logic from here, happy to change it, if there's a better pattern, this does seem to work for me though.
There was a problem hiding this comment.
Yeah, i think we should come back to this. We can go with as is for now.
All listed records were propagating to all dns policies.
This PR does 2 things: