qeda-logo

DocsOperate your services

Viewing logs

Every service card opens onto the same Observability tab — build, deploy, and runtime logs, searchable and filterable, whether a deploy is healthy or stuck.

Click any card on a project's canvas — app or companion — to open its detail panel. It opens on the Observability tab, which is where a deploy's status, its pipeline, and its logs all live.

When a deploy is still going or hits a problem, the top of the tab shows a 5-step pipeline (Initialization, Build image, Push manifest, Deploy to cluster, Live) with a check or a cross on each step, plus a one-line reason when one fails. From there, "Fix & redeploy from the latest commit" re-runs the pipeline without leaving the panel — useful after pushing a fix, since most config changes don't need a full rebuild.

The Observability tab showing a failed deployment pipeline with a "Fix & redeploy" button
Each step of the pipeline reports its own state — this one got through the build and the push, then failed to come up healthy on the cluster.

Build, Deploy, and Runtime logs

Scrolling down past the metrics, the Logs panel splits into three tabs, each with its own line count: Build Logs (the buildpack or Docker build output), Deploy Logs (what happened pushing the image and manifest to the cluster), and Runtime Logs (what the running container itself prints to stdout/stderr). A "live" indicator on Runtime Logs means it's tailing in real time, not a static snapshot.

The Logs panel showing the Build Logs tab, with timestamped monospace build output
Every line is timestamped and tagged with its stage — BUILD here, matching the tab that's open.

Searching and filtering

The search box filters whichever tab is open to matching lines only — useful once a log passes a few hundred lines. Next to it, All / Warnings / Errors narrows by level without typing anything; the line count on the right updates to reflect the active filter. The copy and download icons grab the currently-filtered view, not always the full log.

Note. Runtime Logs only exist once a container has actually started — a build that fails before that point will only ever have Build (and maybe Deploy) logs to show.