feat: OTEL metrics pipeline for task workers#3061

Open

ericallam wants to merge 25 commits intomainfrom

feat/otel-metrics

Commits on Feb 16, 2026

feat(metrics): add support for otel metrics (WIP)
ericallam
committed
stop hardcoding the triggered_at column
ericallam
committed
better machine ID and don't send metrics in between runs in a warm process
ericallam
committed
filter out all system metrics from dev runs
ericallam
committed
Integrate metrics with the AI query service
ericallam
committed
Type the known columns in attributes so we can filter and aggregate by them
ericallam
committed
time bucket thresholds can now be defined per query schema
ericallam
committed
Add support for prettyFormat
ericallam
committed
keep sending otel metrics in between runs, dev now acts more like prod, decouple flushing metrics with metric bucket intervals
ericallam
committed
Add custom metrics examples and provide otel.metrics
ericallam
committed
Add some nodejs metrics
ericallam
committed
fix typecheck issues
ericallam
committed
always ensure valid values for exportIntervalMillis and exportTimeoutMillis.
ericallam
committed
Make FINAL keyword a table specific thing
ericallam
committed
A bunch of improvements thanks to claudes team of review agents
ericallam
committed
A few legit coderabbit fixes
ericallam
committed
Add changeset
ericallam
committed
Associate logs and spans with the machine id
ericallam
committed
Added the Math.max(0, ...) clamp to match the pattern used in columnFormat.ts
ericallam
committed
Actually collect worker_id and then make it a link to deployments when pretty formatting in Query. Also update the core columns for metrics in Query.
ericallam
committed
Simplified metrics_v1 value columns into a single column, improved how we are filtering out system metrics in dev, removed a few of the event loop lag metrics
ericallam
committed
Fix bun metric collection for missing perf_hooks
ericallam
committed
implement semantic system filesystem and diskio metrics for nodejs and bun
ericallam
committed
Improve Table schema and Examples tabs in the query feature for multiple tables
ericallam
committed

Commits on Feb 17, 2026

value -> metric_value
ericallam
committed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

feat: OTEL metrics pipeline for task workers#3061

feat: OTEL metrics pipeline for task workers#3061
ericallam wants to merge 25 commits intomainfrom
feat/otel-metrics

Commits on Feb 16, 2026

feat(metrics): add support for otel metrics (WIP)

stop hardcoding the triggered_at column

better machine ID and don't send metrics in between runs in a warm process

filter out all system metrics from dev runs

Integrate metrics with the AI query service

Type the known columns in attributes so we can filter and aggregate by them

time bucket thresholds can now be defined per query schema

Add support for prettyFormat

keep sending otel metrics in between runs, dev now acts more like prod, decouple flushing metrics with metric bucket intervals

Add custom metrics examples and provide otel.metrics

Add some nodejs metrics

fix typecheck issues

always ensure valid values for exportIntervalMillis and exportTimeoutMillis.

Make FINAL keyword a table specific thing

A bunch of improvements thanks to claudes team of review agents

A few legit coderabbit fixes

Add changeset

Associate logs and spans with the machine id

Added the Math.max(0, ...) clamp to match the pattern used in columnFormat.ts

Actually collect worker_id and then make it a link to deployments when pretty formatting in Query. Also update the core columns for metrics in Query.

Simplified metrics_v1 value columns into a single column, improved how we are filtering out system metrics in dev, removed a few of the event loop lag metrics

Fix bun metric collection for missing perf_hooks

implement semantic system filesystem and diskio metrics for nodejs and bun

Improve Table schema and Examples tabs in the query feature for multiple tables

Commits on Feb 17, 2026

value -> metric_value