e-volv Observer

Every event captured, searchable for months

e-volv Observer captures logs and traces from every project in a workspace hub. Events land through OTLP/HTTP or the @e-volv/logs and e-volv-logs SDKs, are redacted at the client and again at the server backstop, and are written to TimescaleDB hypertables with compression and tiered retention. Search with a deliberately small query language (level filters, service and attribute matches, full-text phrases), tail live streams over SSE, and dashboards read per-minute continuous aggregates instead of scanning raw rows.

• JS/TS and Python SDKs with batching, backoff and drop-oldest under quota
• OTLP/HTTP JSON ingest from any OpenTelemetry collector
• Query language: level>=error service:api "connection timeout"
• Live tail over SSE, in the app and from the evolve-logs CLI
• Dashboards from continuous aggregates: volume, error rate, p95 latency
• Error groups with fingerprinting, merge/split and regression alerts

30days hot retention

1. Incidents and the failure graphErrors are extracted as they land, fingerprinted into groups, and when they burst the incident builder walks failing traces and the call graph to produce a failure graph an agent then explains.