chore(deps): update dependency prometheus-operator/prometheus-operator to v0.94.0

[scality-renovate]

This PR contains the following updates:

Package	Update	Change	Pending
prometheus-operator/prometheus-operator	minor	v0.93.1 → v0.94.0	v0.94.1

---

Release Notes

prometheus-operator/prometheus-operator (prometheus-operator/prometheus-operator)

v0.94.0: 0.94.0 / 2026-09-09

Compare Source

• [CHANGE] Update the default Alertmanager version to v0.34.0 and discard zero-value duration fields (retention, clusterGossipInterval, clusterPushpullInterval, clusterPeerTimeout) in Alertmanager resources instead of passing them as CLI flags, which Alertmanager rejects and can cause startup failures. Ignored fields are reported via the IgnoredFields status condition. #​8800
• [CHANGE] Add named enum types for Hetzner, Docker Swarm, and OpenStack service discovery fields in the ScrapeConfig CRD (HetznerRole, DockerSwarmRole, OpenStackAvailability). #​8789
• [CHANGE] Reject empty strings in namespaceDiscovery.names, consulSDConfig.services, and consulSDConfig.tags list fields in the ScrapeConfig CRD. #​8786
• [FEATURE] Add retentionPercentage field to Prometheus and PrometheusAgent CRDs for volume-based retention (requires Prometheus >= v3.11.0). Setting only the percentage no longer falls back to the default 24h time-based retention. #​8728
• [FEATURE] Add clusterPeerName field to Alertmanager CRD to override the --cluster.peer-name flag (requires Alertmanager >= v0.30.0). #​8767
• [FEATURE] Expose status conditions as Prometheus metrics for Prometheus, PrometheusAgent, Alertmanager, and ThanosRuler resources (prometheus_operator_<resource>_status_condition). #​8719
• [ENHANCEMENT] Tighten the operator's ClusterRole by replacing wildcard verbs with explicit permissions per resource. #​8752
• [ENHANCEMENT] Support webhook_url_file for Discord receiver in Alertmanager configuration Secret (requires Alertmanager >= v0.28.0). #​8035
• [BUGFIX] Require Thanos >= v0.42.0 for delayed compaction with object storage uploads, raised from v0.41.0. Thanos v0.41.0 mis-validates the delayed-compaction paths and the sidecar crash-loops; deployments pinned to v0.41.0 fall back to disabling local compaction. #​8764
• [BUGFIX] Add URL validation for Mattermost receiver fields in Alertmanager configuration Secret. #​8225
• [BUGFIX] Fix operator panic when AlertmanagerConfig Pushover receivers use userKeyFile or tokenFile instead of secret selectors. #​8775
• [BUGFIX] Fix operator panic when kubelet Service synchronization fails and the service is unavailable. #​8743
• [BUGFIX] Fix resources stuck during deletion when the informer misses the deletion timestamp update. #​8727
• [BUGFIX] Fix TracingConfig fields in Prometheus CRD not being serialized due to invalid JSON struct tags. #​8730
• [BUGFIX] Fix externalId version gating for Alertmanager SigV4 configuration to require version >= v0.34.0 (was incorrectly set to v0.33.0). #​8759

---

Configuration

📅 Schedule: (UTC)

• Branch creation
  • "before 9am on monday"
• Automerge
  • At any time (no schedule defined)

🚦 Automerge: Disabled by config. Please merge this manually once you are satisfied.

♻ Rebasing: Whenever PR becomes conflicted, or you tick the rebase/retry checkbox.

🔕 Ignore: Close this PR and you won't be reminded about this update again.

---

• If you want to rebase/retry this PR, check this box

---

This PR has been generated by Mend Renovate CLI.

[github-actions]

Dependency Bump Evaluation

Version change: prometheus-operator/prometheus-operator v0.93.1 → v0.94.0 (minor)

Changes (v0.94.0):

• Alertmanager default version updated to v0.34.0; zero-value duration fields now discarded
• Named enum types added for Hetzner, Docker Swarm, OpenStack SD fields in ScrapeConfig CRD
• Empty strings now rejected in namespaceDiscovery.names, consulSDConfig.services/tags (ScrapeConfig)
• New CRD fields: retentionPercentage (Prometheus/PrometheusAgent), clusterPeerName (Alertmanager)
• Status conditions exposed as Prometheus metrics for Prometheus/Alertmanager/ThanosRuler resources
• Tighter operator ClusterRole (wildcards replaced with explicit verbs)
• Multiple bugfixes including panic fixes and deletion-stuck fixes

Breaking changes: None affecting this project. The [CHANGE] items target Alertmanager defaults, ScrapeConfig CRD validation, and SD-specific enum types — none of which node-warden-operator uses.

Security concerns: None. No CVEs mentioned. Bugfixes improve operator stability (panic and deletion-stuck fixes).

Impact on codebase: Minimal. The only file changed is test/utils/utils.go — a version constant (prometheusOperatorVersion) used exclusively for e2e test setup (installing the prometheus-operator bundle in a kind cluster so the ServiceMonitor CRD is available). The operator's production code does not import or depend on prometheus-operator Go packages; it only ships a ServiceMonitor manifest via kustomize.

CI status: Checks still in progress (build, lint, test, test-e2e, generate).

Recommendation: SAFE TO MERGE (once CI passes)

Notes: This is a test-infrastructure-only bump. The prometheus-operator binary/bundle is downloaded at e2e time to register the ServiceMonitor CRD; the version change has no effect on the operator's own logic or runtime dependencies.

— Claude Code