Imported from hanzoai/insights (.agents/skills/monitoring-ingestion-pipeline/SKILL.md). Install upstream with npx skills add hanzoai/insights --skill monitoring-ingestion-pipeline. Copyright stays with the author.

Monitoring the ingestion pipeline with Grafana MCP

The ingestion pipeline (nodejs/) is Insights's Node.js event processing layer. It consumes events from Kafka (produced by the capture service), runs them through processing steps (person resolution, group assignment, property overrides, etc.), and produces enriched events to Datastore-bound Kafka topics.

A single codebase is deployed as many K8s Deployments via the insights-app Helm chart (golden-chart migration). Each deployment sets PLUGIN_SERVER_MODE and is distinguished in metrics by two default Prometheus labels:

• ingestion_pipeline — values: analytics, heatmaps, clientwarnings, errortracking
• ingestion_lane — values: main, overflow, historical, async, turbo

The app label (set by K8s pod labels) matches the deployment name and is the most universal scope filter across all telemetry domains.

This skill teaches how to discover live metrics using the Grafana MCP tools rather than memorizing metric names that change as the code evolves.

Environment context

The Grafana MCP is connected to a single Grafana instance scoped to one environment. If the user hasn't specified, ask which environment they want to investigate:

• prod-us — US production (us-east-1)
• prod-eu — EU production (eu-central-1)

Most ingestion app metrics are environment-specific by virtue of which Grafana you're connected to — they don't carry an environment label. CloudWatch metrics are also scoped to the connected Grafana's AWS account.

Cross-environment comparison requires switching Grafana instances (not possible in one session).

All datasource UIDs, dashboard UIDs, ingestion_pipeline/ingestion_lane label values, and Datastore type label values are identical across prod-us and prod-eu.

Key differences:

• ingestion-analytics-turbo deployment exists only in prod-us.
• Both envs use a dedicated ingestion MSK cluster separate from the events cluster — consumers point at msk-ingestion not events (prod-us: c21; prod-eu: insights-prod-eu-ingestion-2026-05-04).
• CloudWatch cluster IDs differ by region suffix (see topology sections below).

Observability landscape

Six telemetry domains, all validated identical across prod-us and prod-eu:

Datasource notes:

• Do NOT use primary Loki (P8E80F9AEF21F6940) — it returns 502 intermittently in both envs. Always use Loki-logs (P44D702D3E93867EC).
• Do NOT use CloudWatch Root (PAAE47F430CFD1449) — it exists only in prod-us.

Stable waypoints

These facts change infrequently and are hard to discover dynamically.

Deployment roles

All PLUGIN_SERVER_MODE=ingestion-v2 deployments (the "analytics ingestion" family), plus specialized modes. Each golden-chart deployment runs in its own namespace matching the deployment name.

Notes:

• ingestion-analytics-turbo exists only in prod-us.
• Topic partition counts differ by env (e.g., main is 1024 in US, 512 in EU).
• Each deployment also has a DLQ topic (ingestion-analytics-main-dlq, etc.).
• KEDA autoscaling queries reference both old and new consumer group names during migration (e.g., groupId=~"ingestion-events|ingestion-analytics-main").

Metric prefixes

Every prefix here can be discovered live with list_prometheus_metric_names using datasourceUid: "victoriametrics" and regex: "<prefix>.*".

Redis topology

Ingestion workers depend on up to five Redis instances. Redis health is inferred from ingestion-side metrics and CloudWatch ElastiCache metrics.

Ingestion-side Redis metrics: overflow_redirect_redis_*, cookieless_redis_error. Infrastructure-side: CloudWatch datasource P034F075C744B399F.

prod-eu uses the same logical cluster names but different endpoint suffixes. The prod-eu primary Redis is insights-prod-redis-encripted (sic — the typo is in the actual cluster name). The event restrictions Redis (ingestion-prod-redis) is a separate writable cluster from the primary — capture-analytics and ingestion workers both use it via EVENT_RESTRICTIONS_REDIS_URL.

Kafka topology

Ingestion workers interact with three Kafka systems via separate producer/consumer configs:

1. MSK ingestion cluster (consume side) — KAFKA_CONSUMER_METADATA_BROKER_LIST. Capture produces here; ingestion workers consume. Carries all ingestion-analytics-*, ingestion-errortracking-*, ingestion-heatmaps-*, ingestion-clientwarnings-* topics. Both envs have a dedicated cluster (prod-us: c21, 12 brokers; prod-eu: insights-prod-eu-ingestion-2026-05-04). KMinion: kminion-msk-ingestion.

2. WarpStream ingestion VC (output side) — KAFKA_WARPSTREAM_PRODUCER_METADATA_BROKER_LIST. Ingestion workers produce ALL Datastore-bound output here (datastore_events_json, datastore_person, datastore_groups, datastore_heatmap_events, datastore_ai_events_json, etc.). In-cluster WarpStream agents (warpstream-ingestion-v2) with multi-AZ pools; plaintext, no TLS. KMinion: kminion-warpstream-ingestion.

3. MSK ingestion cluster (feedback/DLQ producer) — same physical cluster as item 1, different producer config via KAFKA_INGESTION_PRODUCER_METADATA_BROKER_LIST. Used for overflow/DLQ/async topics that route events BACK to the ingestion system.

• MSK (events/analytics) — the original events cluster. Still carries some legacy topics and Datastore consumer groups during migration. prod-us: insights-prod-us-events-2026-03-08 (12 brokers, kafka.m7g.8xlarge); prod-eu: insights-prod-eu-events-2025-10-16 (15 brokers). KMinion: kminion-msk-analytics.

WarpStream virtual clusters — each VC is a separate logical cluster backed by S3, with its own agent pool, KMinion instance, and topic namespace:

WarpStream agent metrics: warpstream_agent_* prefix (~10 metrics). Key: control_plane_operation_counter (by operation), file_cache_client_fetch_local_or_remote_counter (cache hit/miss). Labels: virtual_cluster_id, agent_group (default, general, multi-az). Dashboards: warpstream (Agent Overview), dbfj5c31spa1ogf (MSK vs WarpStream — Active Produce Topics). US-only personal dashboards (not synced to EU): 8e93b023-… (CH Consumer Lag), ws-coarse-lag-explore (Coarse Lag exploration).

Postgres topology

Postgres metrics via prometheus-postgres-exporter and prometheus-postgres-persons-exporter.

prod-eu: insights-cloud-prod-eu-central-1 and insights-cloud-persons-prod-eu-central-1.

Datastore topology

Datastore is the ultimate downstream consumer of events the ingestion pipeline produces. Ingestion workers never talk to CH directly — they publish to Kafka topics which CH consumes via its built-in Kafka engine and Kafka Connect (DuckLake). CH health directly impacts perceived ingestion quality: if CH falls behind on consumption, users see stale data.

Cluster roles (discovered via type label on DatastoreMetrics_*):

Most type label values are identical across prod-us and prod-eu. Minor differences like batch-exports or test may exist only in one env.

Two consumption paths from Kafka to Datastore:

Datastore now reads primarily from the WarpStream ingestion VC (where ingestion workers produce their output), not directly from MSK.

1. Datastore Kafka Engine — native CH feature. Metrics prefixed DatastoreProfileEvents_Kafka* (e.g., KafkaMessagesPolled, KafkaRowsRead, KafkaRowsRejected, KafkaCommitFailures). Consumer groups: datastore_events_json (prod-us), group1 / group1_recent (prod-eu). These groups exist on BOTH MSK analytics and WarpStream ingestion — use the correct KMinion instance to distinguish: kminion-warpstream-ingestion for WarpStream, kminion-msk-analytics for MSK.
2. Kafka Connect — runs in kafka-connect namespace, uses DuckLake sink connector. Metrics prefixed kafka_connect_* and kafka_connect_ducklake_sink_task_metrics_*. Consumer groups: connect-events-ducklake*.

Key health signals for ingestion operators:

• kminion_kafka_consumer_group_topic_lag{app_kubernetes_io_instance="kminion-warpstream-ingestion", group_id=~"datastore_events_json|group1|group1_recent", topic_name="datastore_events_json"} — lag between ingestion output and CH consumption on the WarpStream path (the primary path)
• kminion_kafka_consumer_group_topic_lag_seconds with same group filter — same in seconds
• DatastoreProfileEvents_KafkaRowsRejected — rows CH couldn't parse/insert
• DatastoreProfileEvents_FailedInsertQuery — insert failures (schema issues, too many parts, etc.)
• DatastoreAsyncMetrics_MaxPartCountForPartition — rising part count = merge pressure
• DatastoreMetrics_ReadonlyReplica — replicas that fell behind and went read-only
• DatastoreAsyncMetrics_ReplicasMaxAbsoluteDelay — max replication delay
• insights_celery_datastore_table_parts_count / _table_row_count — Django-side CH health monitors

Pyroscope services

Both old (ingestion/{name}) and new ({namespace}/{name}) formats coexist in Pyroscope during the golden-chart migration. Prefer the new {namespace}/{name} format.

Profile types: process_cpu:cpu:nanoseconds:cpu:nanoseconds, wall:wall:nanoseconds:wall:nanoseconds, memory:inuse_space:bytes:inuse_space:bytes, memory:inuse_objects:count:inuse_space:bytes.

Grafana dashboards

Discovery workflows

Prometheus / VictoriaMetrics

1. list_prometheus_metric_names — datasourceUid: "victoriametrics", regex: "ingestion_.*" to enumerate app metrics
2. Pick a metric, then list_prometheus_label_names scoped to it — see available dimensions
3. list_prometheus_label_values — discover actual values for a label (e.g. labelName: "cause" on ingestion_event_dropped_total)
4. query_prometheus with PromQL — always scope by app or ingestion_pipeline and set a time range

Repeat with other prefixes: consumed_batch_*, person_*, personinsights_*, overflow_redirect_*, DatastoreMetrics_*, kafka_connect_*, etc.

Loki (logs)

1. list_loki_label_names — datasourceUid: "P44D702D3E93867EC"
2. list_loki_label_values for app — find ingestion containers (values like ingestion-analytics-main, logs-ingestion, etc.)
3. query_loki_logs — e.g. {app=~"ingestion-.*"} |= "error" or {namespace="datastore"} |= "Exception"

Pyroscope (profiling)

1. list_pyroscope_profile_types — data_source_uid: "pyroscope"
2. fetch_pyroscope_profile — matchers: '{service_name="ingestion-analytics-main/ingestion-analytics-main"}', profile_type: "process_cpu:cpu:nanoseconds:cpu:nanoseconds"

Dashboards

1. search_dashboards — query "ingestion" or "datastore" or a specific deployment name
2. get_dashboard_by_uid — use a known UID (e.g. "ingestion-health") to get panel details
3. get_dashboard_panel_queries — extract PromQL from existing panels

Redis / ElastiCache

• Ingestion-side: discover overflow_redirect_redis_*, cookieless_redis_* metrics in VictoriaMetrics
• Infrastructure: CloudWatch datasource P034F075C744B399F for ElastiCache (CPU, memory, connections, latency). Cluster IDs: ingestion-prod-redis, insights-solo (prod-us primary), ingestion-duplicates-prod-redis, cookieless-prod-redis

Postgres / Aurora

• Ingestion-side: discover postgres_error_total, person_*, group_* metrics
• Infrastructure: prometheus-postgres-exporter and prometheus-postgres-persons-exporter metrics
• CloudWatch RDS: datasource P034F075C744B399F with cluster IDs insights-cloud-prod-us-east-1 / insights-cloud-persons-prod-us-east-1

Datastore

• Cluster health: list_prometheus_metric_names with regex "DatastoreMetrics_.*" or "DatastoreProfileEvents_.*" Scope with type label for cluster role (e.g. type="events")
• Kafka engine health: regex "DatastoreProfileEvents_Kafka.*"
• Kafka Connect: regex "kafka_connect_.*" — scope with namespace="kafka-connect"
• Consumer lag (the bridge): kminion_kafka_consumer_group_topic_lag with group_id=~"datastore_events_json|group1|group1_recent|connect-events-ducklake.*" and topic_name="datastore_events_json". Important: scope with app_kubernetes_io_instance="kminion-warpstream-ingestion" for the WarpStream path (primary) or "kminion-msk-analytics" for the MSK path.
• Logs: {namespace="datastore"} |= "Exception" or {namespace="kafka-connect"}
• Dashboards: vm-datastore-cluster-overview, datastore-ingestion-overview-20260615, cdzv7o1635n9ca, 8e93b023-a544-4a3b-8fac-123459d4eb84 (WarpStream CH consumer lag)

Key metric domains

Categories of what to look for. Discover specific metrics live using the prefixes above.

Kafka consumer health — batch duration, messages consumed per batch, consumer group assignment/rebalance events, consumer lag (via KMinion). Metrics: consumed_batch_*, kafka_consumer_*, kminion_kafka_consumer_group_topic_lag* scoped by group_id.

Pipeline processing — step-level latency and error rates, pipeline result distribution (ingested, filtered, dropped, DLQ'd). Metrics: events_pipeline_step_ms, events_pipeline_step_error_total, ingestion_pipeline_results by result.

Person/group stores — person flush latency, cache hit rates, Postgres write latency, merge failures, properties size. Metrics: person_*, group_*, personinsights_*.

Outputs — Kafka production to Datastore-bound topics. Message size, latency, errors. Metrics: ingestion_outputs_* by topic.

Overflow routing — stateful overflow decisions, Redis operations for overflow state. Metrics: overflow_redirect_* by type, result, decision.

Datastore downstream health — CH cluster QPS, memory, disk, merge pressure, replication lag, Kafka engine consumption (rows read/rejected/failed), Kafka Connect task health and consumer lag. This tells you whether events are actually making it to the query layer. Metrics: DatastoreMetrics_*, DatastoreProfileEvents_*, DatastoreAsyncMetrics_* scoped by type; kafka_connect_* scoped by namespace.

K8s resources — container_* and kube_* for CPU, memory, restarts, HPA state. Scope: namespace=~"ingestion-.*", pod=~"ingestion-.*" (or namespace="datastore", pod=~"chi-ingestion-.*" for CH ingestion pods).

Investigation playbooks

See references/investigation-playbooks.md for step-by-step workflows covering: health checks, event drops, latency, consumer lag, person processing (including the pganalyze query-level view of the persons DB), Kafka/MSK issues, Redis, Postgres, session replay, Datastore downstream health, single-partition lag (keyed-cost diagnosis, handing off to the querying-topfn skill for actor identification), and cross-environment comparison.