Knowledge Indexing Status is how you watch a knowledge source index and troubleshoot it when a sync stalls or fails.
Scope — Which View Is This?
This page describes a Knowledge Set's Sources view — its knowledge sources grouped by type (Application, Documents, Websites, Data catalog), each shown with a Status and an enable toggle — plus the per-source Overview page with Crawled / Pending / Synced / Failed tiles. On an Advanced agent, sources appear in one Knowledge table (see Agent Knowledge), but each source opens the same Overview page described below.
If you added knowledge directly to a Simple agent, you get a lighter view: the agent editor's Knowledge section lists each source with a status badge (Uploading…, Syncing…, Sync Error, or Synced with a date), and that badge — not this full Sources table — is your indexed-or-not indicator.
Where Do I See a Source's Indexing Status?
In a knowledge set's Configuration tab, sources are grouped by type — Application, Documents, Websites, and Data catalog — each in its own list. Every row shows the source's current Status and Owner, with an enable toggle to turn it on or off. Open a source to see its Overview page, which shows live indexing progress and the source's sync schedule.
Application sources get two lists — Initial sync in progress for sources still finishing their first sync, and All knowledge sources — with live counts of items crawled, items synced, the crawl/change rate, and failures.
What Do the Indexing Statuses Mean?
A source shows one status:
Status | Meaning |
|---|---|
Draft | Added but not yet indexed (also what a source returns to when you disable it). |
Syncing | Currently being fetched and indexed (a queued source also shows as Syncing). |
Completed | Fully indexed and ready for retrieval. |
Scheduled | Fully indexed and set to re-sync on a schedule (a completed source with a sync schedule reads as Scheduled). |
Failed | Indexing hit an error. |
On the source's Overview page you'll also see the sync status, tiles for Crawled, Pending, Synced, and Failed, an Items crawled over time chart, and a Content type breakdown. In the source's file lists, each failed document shows Error Details, so you can see exactly which documents had trouble.
Progress is tracked per document, not just per source: the platform counts how many items it has discovered versus how many it has finished indexing, which is what drives these counters. A "queued" source is real, not cosmetic — syncs are concurrency-limited, so when too many sources sync at once the extras wait for a slot and start automatically (they display as Syncing).
Why a Large Application Source Takes So Long
Application sources sync in two stages: content is first downloaded from the app into a staging area, then indexed into chunks — each with its own progress cursor, so a big source works through its backlog incrementally and resumes where it left off after any interruption. Long-running syncs are therefore usually normal for large sources; watch the Crawled and Synced counts — as long as they advance, it's working.
The platform also runs background checks that detect genuinely stuck syncs and restart them, and it can raise an alert when a sync has been in progress unusually long or when indexing fails. Individual document failures don't block the rest of the source: failed items are retried and reported under Error Details.
Re-syncing a Source
Open a source and use Sync Knowledge to re-index it on demand. The button appears on a source that has finished syncing (Completed) and has Schedule Knowledge Sync turned on. Sources on a schedule re-sync automatically at their Next sync time; you can change the schedule in the source's General settings.
Note: Re-sync after you've added or changed content at the origin so the knowledge set reflects it.
My Knowledge Source Failed to Index — What Do I Do?
What's happening: the source shows Failed, and its detail page lists Error Details for the documents that didn't index.
Why: the most common causes are an expired or revoked connection for an app source, a document the parser couldn't read, or a source that's too large. For app sources you may see a message that "The auth token has expired. Please re-authenticate to continue indexing."
Fix:
Open the source and read its Error Details to see which documents failed and why.
If it's an auth message, re-authenticate the connection, then retry the failed documents.
If specific documents failed, fix or remove them at the source.
Click the retry icon next to each failed document in the source's file list, or wait for the next scheduled sync.
Still stuck? Check the source's indexing settings (parsing method and supported file types) and contact support@unifyapps.com.