Skip to content

fix(charts): explicit lenient probe tolerances (sse-proxy, otel wrappers) + Quantity schema typing #32

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
braghettos merged 1 commit into from
Aug 7, 2026
Merged

fix(charts): explicit lenient probe tolerances (sse-proxy, otel wrappers) + Quantity schema typing #32

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
13 changes: 13 additions & 0 deletions charts/krateo-sse-proxy/templates/deployment.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -47,18 +47,31 @@ spec:
optional: {{ .Values.clickhouse.passwordSecret.optional }}
- name: LISTEN_ADDR
value: {{ .Values.listenAddr | quote }}
# Lenient liveness tolerance (org standard): the Kubernetes defaults
# (timeoutSeconds: 1, failureThreshold: 3) kill HEALTHY pods under CPU
# contention — a slow-but-alive /health reply counts as a failure and the
# resulting restart storm worsens the contention. Only a truly dead process
# should die: 10s period x 6 failures = 60s of continuous failure before
# a restart.
livenessProbe:
httpGet:
path: /health
port: 8080
initialDelaySeconds: 5
periodSeconds: 10
timeoutSeconds: 5
failureThreshold: 6
# Readiness stays prompt (5s period, 3 failures) so a genuinely unhealthy
# endpoint is removed from the Service quickly, but timeoutSeconds: 5 keeps
# load alone from flapping the endpoint in and out.
readinessProbe:
httpGet:
path: /health
port: 8080
initialDelaySeconds: 3
periodSeconds: 5
timeoutSeconds: 5
failureThreshold: 3
resources:
{{- toYaml .Values.resources | nindent 12 }}
securityContext:
Expand Down
42 changes: 42 additions & 0 deletions charts/otel-collector-daemonset/values.schema.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -528,6 +528,48 @@
}
}
}
},
"livenessProbe": {
"type": "object",
"additionalProperties": true,
"properties": {
"initialDelaySeconds": {
"type": "integer"
},
"periodSeconds": {
"type": "integer"
},
"timeoutSeconds": {
"type": "integer"
},
"failureThreshold": {
"type": "integer"
},
"terminationGracePeriodSeconds": {
"type": "integer"
}
}
},
"readinessProbe": {
"type": "object",
"additionalProperties": true,
"properties": {
"initialDelaySeconds": {
"type": "integer"
},
"periodSeconds": {
"type": "integer"
},
"timeoutSeconds": {
"type": "integer"
},
"successThreshold": {
"type": "integer"
},
"failureThreshold": {
"type": "integer"
}
}
}
},
"description": "Values forwarded to the upstream opentelemetry-collector subchart."
Expand Down
15 changes: 15 additions & 0 deletions charts/otel-collector-daemonset/values.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -176,3 +176,18 @@ opentelemetry-collector:
requests:
cpu: 100m
memory: 256Mi

# Probe tolerances, passed through to the upstream opentelemetry-collector chart's
# templates/_pod.tpl (org standard). The Kubernetes defaults (timeoutSeconds: 1,
# failureThreshold: 3) kill HEALTHY pods under CPU contention — a slow-but-alive
# health reply counts as a failure and the resulting restart storm worsens the
# contention. Liveness is lenient so only a truly dead process dies: implicit
# periodSeconds default 10 x 6 failures = 60s of continuous failure before restart.
livenessProbe:
timeoutSeconds: 5
failureThreshold: 6
# Readiness keeps prompt endpoint removal (failureThreshold: 3, implicit 10s period)
# but timeoutSeconds: 5 keeps load alone from flapping the endpoint in and out.
readinessProbe:
timeoutSeconds: 5
failureThreshold: 3
42 changes: 42 additions & 0 deletions charts/otel-collector-deployment/values.schema.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -515,6 +515,48 @@
}
}
}
},
"livenessProbe": {
"type": "object",
"additionalProperties": true,
"properties": {
"initialDelaySeconds": {
"type": "integer"
},
"periodSeconds": {
"type": "integer"
},
"timeoutSeconds": {
"type": "integer"
},
"failureThreshold": {
"type": "integer"
},
"terminationGracePeriodSeconds": {
"type": "integer"
}
}
},
"readinessProbe": {
"type": "object",
"additionalProperties": true,
"properties": {
"initialDelaySeconds": {
"type": "integer"
},
"periodSeconds": {
"type": "integer"
},
"timeoutSeconds": {
"type": "integer"
},
"successThreshold": {
"type": "integer"
},
"failureThreshold": {
"type": "integer"
}
}
}
},
"description": "Values forwarded to the upstream opentelemetry-collector subchart."
Expand Down
15 changes: 15 additions & 0 deletions charts/otel-collector-deployment/values.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -263,3 +263,18 @@ opentelemetry-collector:
requests:
cpu: 100m
memory: 256Mi

# Probe tolerances, passed through to the upstream opentelemetry-collector chart's
# templates/_pod.tpl (org standard). The Kubernetes defaults (timeoutSeconds: 1,
# failureThreshold: 3) kill HEALTHY pods under CPU contention — a slow-but-alive
# health reply counts as a failure and the resulting restart storm worsens the
# contention. Liveness is lenient so only a truly dead process dies: implicit
# periodSeconds default 10 x 6 failures = 60s of continuous failure before restart.
livenessProbe:
timeoutSeconds: 5
failureThreshold: 6
# Readiness keeps prompt endpoint removal (failureThreshold: 3, implicit 10s period)
# but timeoutSeconds: 5 keeps load alone from flapping the endpoint in and out.
readinessProbe:
timeoutSeconds: 5
failureThreshold: 3
Loading