Sync runs: statuses, retries and logs
What Queued, Running, Succeeded, Failed and Cancelled actually mean.
Updated Aug 3, 2026
Every sync — scheduled or manual — is recorded as a run with a status, item count, duration and a full log.
Statuses
- Queued — waiting for a worker slot (each workspace has a fair-use concurrency cap so one tenant can't starve others).
- Running — in progress; the card shows live percentage, items seen and ETA.
- Succeeded — completed; item counts and new-signal counts are recorded.
- Failed — stopped on an error; the message is on the run and the details are in its log.
- Cancelled — a human pressed Stop. Partial results up to that point are kept.
The log drawer
Click any run to open its log: timestamped entries for each stage (authenticate → index → fetch → parse → embed). Logs are retained per your workspace retention policy.
Retries
Transient failures (network, rate limits) retry automatically with exponential backoff — up to three attempts before a run is marked Failed. Failed scheduled runs don't pile up: the next scheduled run supersedes them.
Re-sync from scratch
Connector menu → Re-sync from scratch rebuilds the connector's signals from the full backfill window. Existing signals are deduplicated by source id, so processes and opportunities built on them survive. Use it after big scope changes.