Web Server Monitoring Dashboard: Requests Average Duration Query error

[tamir-alltrue-ai]

Description

In the default web dashboard, the underlying query fails with:

External error: Execution error: Cannot convert Duration(Microsecond) to epoch

I looked at the underlying query - it looks like valid postgres syntax to me but it is not running.

WITH dataset AS (
	SELECT
	 (attributes->>'http.method') || ' ' || COALESCE(attributes->>'http.route', attributes->>'http.target') as endpoint,
	 extract(epoch from end_timestamp - start_timestamp) * 1000 as duration_ms
	FROM records
	WHERE attributes ? 'http.method' and (attributes ? 'http.route' or attributes ? 'http.target')
)
SELECT endpoint, avg(duration_ms) as average_duration_ms
FROM dataset
GROUP BY endpoint
ORDER BY average_duration_ms desc
LIMIT 7;

Important line is extract(epoch from end_timestamp - start_timestamp) * 1000 as duration_ms

Python, Logfire & OS Versions, related packages (not required)

logfire="0.53.0"
platform="macOS-14.4-arm64-arm-64bit"
python="3.11.6 (v3.11.6:8b6ee5ba3b, Oct  2 2023, 11:18:21) [Clang 13.0.0 (clang-1300.0.29.30)]"
[related_packages]
requests="2.31.0"
pydantic="2.7.1"
fastapi="0.110.0"
openai="1.14.3"
protobuf="4.25.3"
rich="13.7.1"
executing="2.0.1"
opentelemetry-api="1.27.0"
opentelemetry-exporter-otlp-proto-common="1.27.0"
opentelemetry-exporter-otlp-proto-http="1.27.0"
opentelemetry-instrumentation="0.48b0"
opentelemetry-instrumentation-aiohttp-client="0.48b0"
opentelemetry-instrumentation-asgi="0.48b0"
opentelemetry-instrumentation-dbapi="0.48b0"
opentelemetry-instrumentation-fastapi="0.48b0"
opentelemetry-instrumentation-grpc="0.48b0"
opentelemetry-instrumentation-httpx="0.48b0"
opentelemetry-instrumentation-jinja2="0.48b0"
opentelemetry-instrumentation-psycopg2="0.48b0"
opentelemetry-instrumentation-redis="0.48b0"
opentelemetry-instrumentation-requests="0.48b0"
opentelemetry-instrumentation-sqlalchemy="0.48b0"
opentelemetry-instrumentation-sqlite3="0.48b0"
opentelemetry-instrumentation-system-metrics="0.48b0"
opentelemetry-instrumentation-tornado="0.48b0"
opentelemetry-instrumentation-urllib="0.48b0"
opentelemetry-instrumentation-urllib3="0.48b0"
opentelemetry-proto="1.27.0"
opentelemetry-sdk="1.27.0"
opentelemetry-semantic-conventions="0.48b0"
opentelemetry-util-http="0.48b0"

[alexmojaki]

This is specifically an issue with the new FF engine, see #408

The underlying problem is fixed in apache/datafusion#12514 but that code isn't being used yet on our platform.

Change epoch to second and the dashboard should work in both TS and FF.

[alexmojaki]

This was fixed shortly after being opened.