Envoy insights

Open SourceVersions: [1.30]64 metrics

OpenTelemetry·

100% trace sampling causes performance degradationwarning

gateway.envoyproxy.io

29d ago▸

EnvoyFilter configuration not applied to proxy

1mo ago▸

Proxy configuration out of sync with control plane

1mo ago▸

Envoy proxy container crashes after EnvoyFilter applied

1mo ago▸

EnvoyFilter present in config but not affecting trafficwarning

1mo ago▸

Debug logging enabled on production Envoy proxywarning

1mo ago▸

Envoy sidecar OOM kill causes connection drops and service outages

envoy.server.memory.allocated envoy.server.memory.heap

1mo ago▸

Memory fragmentation indicated by heap size vs allocated gap

envoy.server.memory.heap envoy.server.memory.allocated

1mo ago▸

Excessive service mesh configuration drives sidecar memory usage by 50-80%

1mo ago▸

High active connection count indicates connection leaks

envoy.http.downstream.connection.active envoy.cluster.upstream.connection.active envoy.http.downstream.connection.total+1 more

1mo ago▸

Large payload buffering drives up Envoy memory consumption

1mo ago▸

High-cardinality metric labels cause stats explosion and memory growth

1mo ago▸

Access logging with high traffic consumes log buffer memory

1mo ago▸

Sidecar memory usage exceeds 85% of limit

1mo ago▸

Predicted OOM kill within 2 hours based on memory growth trend

1mo ago▸

Previous OOM kill detected in container history

1mo ago▸

Excessive worker threads cause unnecessary baseline CPU consumption

envoy.appmesh.cpu.utilization envoy.http.downstream.request.total

1mo ago▸

Large xDS configuration updates cause CPU spikes

envoy.appmesh.cpu.utilization

1mo ago▸

Telemetry collection consumes excessive CPU per request

envoy.appmesh.cpu.utilization envoy.http.downstream.request.total

1mo ago▸

mTLS handshakes are the biggest CPU consumer

envoy.listener.ssl.handshake envoy.appmesh.cpu.utilization envoy.cluster.upstream.connection.total

1mo ago▸

Protocol sniffing adds CPU overhead per new connection

envoy.appmesh.cpu.utilization envoy.cluster.upstream.connection.total

1mo ago▸

Non-mesh traffic unnecessarily consumes sidecar CPU

envoy.appmesh.cpu.utilization envoy.server.total_connections

1mo ago▸

Insufficient CPU limits cause sidecar throttling

envoy.appmesh.cpu.utilization

1mo ago▸

Frequent health checks generate constant background CPU usage

envoy.cluster.health_check.attempt envoy.appmesh.cpu.utilization envoy.cluster.membership.total

1mo ago▸

Complex routing rules with regex are expensive per request

envoy.appmesh.cpu.utilization envoy.http.downstream.request.total

1mo ago▸

HTTP 5xx error rate exceeds 1% threshold causing user-facing failurescritical

envoy.appmesh.target.response.5xx envoy.http.downstream.request.by_status

1mo ago▸

Checkout endpoint P95 latency exceeds 3 seconds causing cart abandonment riskcritical

envoy.http.downstream.request.time envoy.appmesh.target.response_time

1mo ago▸

Service CPU saturation above 85% causes request drops and latency increasecritical

envoy.appmesh.cpu.utilization

1mo ago▸

SLO error budget exhaustion from elevated burn ratewarning

1mo ago▸

Prometheus metrics endpoint causes overhead when using OpenTelemetry exclusivelyinfo

gateway.envoyproxy.io

1mo ago▸