Skip to content

Jobs, logs, and errors

FlowO is built around job-level observability: every Snakemake job row is tied to status, wildcards, commands, and—when failures occur—structured error payloads and log paths.

Jobs table (run detail → Jobs)

Open a run at /runs/{id}, select the Jobs tab, and sort or scroll by rule and status. Columns typically include rule name, status, runtime, and wildcard key material so you can spot mis-parameterized rules quickly.

Jobs table showing rule names, statuses, and wildcards

From a failed job to the root cause

  1. Status — FAILED jobs surface in the table and in dashboard aggregates.
  2. Error summary — Open the job row or the workflow-level error list; FlowO stores Snakemake’s error message and traceback when provided by the logger.
  3. Shell command — Inspect the resolved command (wildcards expanded) to compare with what you intended to run.
  4. Logs — Open the job log path captured at runtime, or use workflow-level log modals from the Runs list when enabled.

Example error view with rule, message, and log/traceback context

Common statuses

Status Meaning
Queued / pending Waiting on dependencies or scheduler.
Running Currently executing.
Success / completed Finished without Snakemake error.
Failed Snakemake reported job_error or non-zero exit where captured.
Cancelled Run stopped before the job finished.

Exact strings may match your Snakemake version and normalisation in FlowO; use the UI labels as the source of truth.

Troubleshooting tips

  • Wildcards — Wrong expansion is a frequent source of “file not found” or empty inputs; compare table wildcards with Snakefile placeholders.
  • Repeated rule failures — Use the Dashboard “rule errors” style summaries to see hotspots across runs.
  • “Command not found” — Verify conda/env modules on the execution host; FlowO shows what was reported, not your interactive shell PATH.

See also