bug: Default PrometheusRule alert incorrectly references "Reth" instead of Abstract external node

[Sertug17]

Description

The default PrometheusRule alert in prometheusrules.yaml is named RethNodeDown with summary/description referencing "Reth Node". Abstract is a zkSync-based L2 running matterlabs/external-node it has no relation to Reth (Rust Ethereum execution client). This appears to be a copy paste artifact from a Reth-based chart template.

Location

charts/abstract-node/templates/prometheusrules.yaml (L26-33):

{{- if .Values.metrics.prometheusRule.default }}
    - name: {{ include "common.names.fullname" $ }}-default
      rules:
        - alert: RethNodeDown
          expr: up{job="{{ include "common.names.fullname" . }}"} == 0
          for: 5m
          labels:
            severity: critical
          annotations:
            summary: Reth Node {{ printf "{{ $labels.instance }}" }} down
            description: Reth Node {{ printf "{{ $labels.instance }}" }} is down
{{- end }}

Impact

• Alert routing confusion: Teams using label-based routing (e.g., PagerDuty/Opsgenie rules matching alert names) may not route RethNodeDown alerts correctly for their Abstract infrastructure.
• Incident response: On-call engineers receiving a RethNodeDown alert with "Reth Node is down" will look for a Reth process/container that doesn't exist, delaying root cause identification.
• Dashboard/runbook misalignment: Any runbook or dashboard linked to this alert name will reference the wrong component.

Suggested Fix

        - alert: AbstractExternalNodeDown
          expr: up{job="{{ include "common.names.fullname" . }}"} == 0
          for: 5m
          labels:
            severity: critical
          annotations:
            summary: Abstract External Node {{ printf "{{ $labels.instance }}" }} down
            description: Abstract External Node {{ printf "{{ $labels.instance }}" }} is down

Affected Files

• charts/abstract-node/templates/prometheusrules.yaml