Visibility of Reason for Failure of Connectors

1. Overview

When a connector sync fails, the Knowledge Management (KM) dashboard tells you three things in one place: what failed, why it failed, and what to do next.

Every failure carries a unique error code, is classified into a category and sub-reason, and is accompanied by actions you can take directly from the dashboard — retry the sync, open the execution logs, or reach Leena AI support. Administrators can diagnose and resolve most sync failures without raising a ticket.


2. Where you see it

Failure details appear in two places, and both open the same view:

  • Sync Logs — go to Knowledge Management → Sync logs → Connector sync logs. A failed run shows a Sync failed status chip; hover to see View failure details, and click to open the sync drawer.
  • Connector settings — open the connector under Settings → Integrations and click into the sync run.

Related status chips you may see on the same screen:

ChipMeaning
Sync failedThe run failed. Click to see the error code and reason.
Partially syncedThe run completed, but some articles could not be synced. Each problem is listed with its own code.
Sync stoppedSomeone cancelled the run. No articles from that run are available. This is not a failure and has no error code.
Request timed outThe connector did not respond within the expected window.

3. What the failure card shows

The drawer opens on a failure card containing:

ElementExample
Sub-reason (card title)Authentication failed
What happened (card description)Your ServiceNow connection has expired or no longer has the required permissions.
Error code (chip on the right)SNW-00201
What to do (highlighted bar)Reconnect ServiceNow or update its permissions in Connector settings, then retry the sync.
ActionsRetry sync, View execution logs (or Need help)

The sync progress timeline in the same drawer marks the step that failed, showing Error <code> : <sub-reason> against that step — so you can see whether the run broke at discovery, at content fetch, at permission sync, or at indexing.


4. Actions available on a failure

ActionWhat it doesWhen it appears
Retry syncOn the connector's settings page, re-runs the sync immediately. From Sync Logs, takes you to that connector's settings page, where the sync can be triggered.Always on a failure card. Disabled while a sync is already running for that connector, and on connections that cannot be re-synced from the current environment.
View execution logsOpens Integrations → KM connectors → Execution logs, filtered to the connector's API activity, so you can see the exact calls and responses behind the failure.For Auth / Permission and Source failure categories, and on every article-level failure group.
Need helpOpens the Leena AI support portal so you can raise a ticket.Instead of View execution logs when the category is Internal error — these are failures inside the Leena AI platform, where execution logs will not tell you anything actionable.

5. Error code structure

Every code has the shape E-<PREFIX><SUFFIX> — for example E-SNW00201. In the dashboard the code is displayed as PREFIX-SUFFIX, so E-SNW00201 is shown as SNW-00201.

  • E — Error.
  • PREFIX — a 3-letter code identifying the connector.
  • SUFFIX — a 5-digit code identifying the failure. It is the same across all connectors, so 00201 always means "authentication failed", whichever connector reported it.

Connector prefixes

ConnectorPrefixConnectorPrefix
BoxBOXServiceNowSNW
ConfluenceCNFSFTPSFT
Google DriveGDRSharePoint (Graph)SPG
InteractINTUnilyUNL
SalesforceSFCWorkdayWRK
ZendeskZDKAny other / custom sourceCON

6. Error code reference

Codes are grouped into three categories. The first group of digits indicates the series: 001 source-side, 002 access-related, 003 internal.

Auth / Permission

CodeSub-reasonWhat it means and what to do
00201Authentication failedThe connection has expired or the credentials are no longer valid. Reconnect the source or update its permissions in Connector settings, then retry.
00202Permission deniedThe account Leena AI uses does not have access to the requested content. Grant access at the source, then retry.
00204Request rejected by sourceThe source rejected the request. Check the connector configuration, then retry.

Source failure

CodeSub-reasonWhat it means and what to do
00100Source server errorThe source was unreachable during this sync. Usually temporary — retry once the source is back up.
00101No articles returnedThe source returned no articles, so the sync was stopped to protect your existing content. Verify content exists in the selected scope, then retry.
00102Network errorThe connection to the source dropped. Usually temporary — retry.
00203Rate limit exceededThe source is temporarily limiting requests. Retry after some time.
00205Partial accessPart of the selected scope could not be read. Check access to the affected sites, spaces or folders, then retry.
00206Empty articleThe affected articles have no content at the source. Add content at the source, then retry.
00207Not found at sourceThe affected articles could not be found. They may have been moved or deleted at the source. Retry, and check they still exist if the error repeats.

Internal error

CodeSub-reasonWhat it means and what to do
00300Unknown errorAn unexpected error inside the Leena AI platform. Retry the sync; use Need help if it repeats.
00301Connector errorA failure in the connector service. Retry the sync; use Need help if it repeats.
00302 / 00303Internal Server ErrorA failure while handing content to the Knowledge Management or NLP service. Retry the sync; use Need help if it repeats.
📘

A code that is not recognised is shown as Internal error : Unknown error with the raw code preserved on the card, so it can still be quoted to support.


7. Article-level failures (partially synced runs)

When a run completes but individual articles could not be processed, the drawer lists them grouped by problem — for example empty content, corrupted file, file too large, unsupported file type, permission denied, or not found. Each group shows:

  • the number of affected articles, with the article name and a link to the item at the source,
  • the error code for that group (for example GDR-00206),
  • a short explanation of the problem, and
  • Retry sync and View execution logs.

This makes it possible to fix a handful of articles at the source and re-run, rather than treating the whole sync as failed.


8. Protective stop on an empty source

If a sync discovers zero articles at the source, Leena AI fails the run on purpose rather than removing your existing content. Nothing is deleted from Knowledge Management. Check that content still exists in the selected scope — the folders, spaces, sites or knowledge bases mapped in the connector — and that the account Leena AI uses can still see it, then retry.


Did this page help you?