Repository navigation
Route NSO activity policy failures to Deliver - #515
Merged
Merged
Conversation
scotwells
marked this pull request as ready for review
September 28, 2026 19:03
ecv
approved these changes
Sep 28, 2026
scotwells
added a commit
to milo-os/activity
that referenced
this pull request
Sep 28, 2026
When an activity policy fails, customers can see gaps in their resource
timelines. Route those failures to the team that owns the policy so the
right team can investigate and restore missing activity entries,
including events already queued for retry.
Set ownership on the ActivityPolicy resource's `metadata.labels`, for
example:
```yaml
metadata:
labels:
meta.datumapis.com/service: network-services
meta.datumapis.com/team: delivery
```
`meta.datumapis.com/service` becomes the alert's `service` label;
`meta.datumapis.com/team` becomes its `team` label. Existing
Alertmanager routes send `service=network-services` to Deliver;
`team=delivery` records the policy's owning team. Setting the team label
alone does not create a route.
Policies without an owner still alert: a missing service stays unset, a
missing team falls back to `sre`, and missing ownership metrics do not
suppress failures. Ownership changes also apply to existing retry
failures.
Validated processor tests and ten alert-rule scenarios covering
ownership changes, missing metadata, replicas, cluster isolation, and
scrape-label collisions; CI is green.
Roll out with datum-cloud/infra#6325 and the
policy labels in
datum-cloud/network-services-operator#515 and
datum-cloud/dns-operator#192. Include this
change in staging's federated-events branch before deployment.
Related to datum-cloud/infra#6019.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
When network activity policies fail, customers can see missing entries in their resource timelines. Assign all nine NSO ActivityPolicies to
network-services/deliveryso these alerts reach the Deliver team responsible for investigating and restoring activity processing.Set these exact labels on all nine NSO policies:
The companion Activity change maps these to alert labels
service=network-servicesandteam=delivery; existing Alertmanager routes select Deliver using the service value.team=deliveryrecords ownership; that label alone does not create a route.This adds ownership metadata; activity content and matching behavior stay the same. Validated ownership on all nine rendered policies; CI is green.
Alert routing requires milo-os/activity#255 and https://github.com/datum-cloud/infra/pull/6325.
Related to https://github.com/datum-cloud/infra/issues/6019.