Skip to main content

Troubleshooting

Last updated: 26 October 2026

Solutions to the problems users hit most often.

Before anything else, two things are worth knowing:

  • The Log panel on the main window shows what the Ingester has been doing. Most problems announce themselves there first.
  • The Open logs button opens the folder with the full log files. Always include these when reporting a problem.

Start here​

SymptomGo to
Files appear in the folder but no jobs are createdNothing happens when files appear
Jobs are created but sit thereJobs pile up but nothing uploads
An upload never gets past 0%Upload is stuck at zero progress
Uploads are much slower than the venue connection should allowUploads are slow
New uploads do not start while a Network Report runsUploads wait for the Network Report
A red ENCODING STALLED bannerENCODING STALLED banner
A job shows failedA job failed after its upload
Access DeniedAccess denied when uploading
Contest code not recognizedContest code not recognized
Please select a partnerPlease select a partner
API Configuration IncompleteAPI Configuration Incomplete
The overlay logo is rejectedOverlay Image URL Error
The program closes on startupFailed to load Ingest Partners
It says another copy is runningAn instance is already running
It asks Close anyway? when you close itClosing while jobs are processing
Registration or licence troubleRegistration and licence problems
A job is stuck and you want to try againRe-running a job

Nothing happens when files appear in the watch folder​

Work through these in order:

  1. Is WATCHER switched on? The button at the top must be blue. Without it, nothing scans the folders.
  2. Are the watch folders correct? Open SETTINGS and check. Folders that do not exist are dropped when you save — if a path is missing from the list, it was mistyped or the drive was not connected.
  3. Is the file extension in the list? A .mov file is ignored unless .mov is in Video file extensions.
  4. Is the API configuration complete? Backend System, API url and API key must all be set. Without them the Ingester will not create jobs at all.
  5. Is the Video CutterV2 still writing? If the companion .json file says FinishedCutting: false, the Ingester deliberately skips the file and writes "Skipping … because FinishedCutting is not yet true" to the log. It will pick it up once cutting completes.
  6. Was the file cancelled before? With Ignore Canceled ticked, a previously cancelled file is never picked up again. Untick it and rescan.

Click Rescan folders to force an immediate scan rather than waiting for the automatic one.

Jobs pile up but nothing uploads​

Check that WORKER is switched on. If it is off while jobs are waiting, the WORKER button blinks amber and the taskbar icon flashes — that is exactly this situation.

If WORKER is on and jobs still do not move, check the Connected indicator in the top-left corner, and the Queued header: if it says waiting for the Network Report, a Network Report is running and uploads start again when it ends.

Jobs in Needs attention never move by themselves — they wait for you. See the Next action column for what each one needs.

Upload is stuck at zero progress​

Usually a slow or unstable connection that cannot cope with the current upload settings.

  1. Open SETTINGS. If File chunk size or Parallel connections is set to a fixed value, set it back to Auto; if both are already on Auto, set File chunk size to 5 MB. Click Save settings. The new values apply to uploads that start after this.
  2. Right-click the stuck job and choose Cancel.
  3. Open the Cancelled group (click its header), find the job — the search box helps — right-click it and choose Re-run "your file".

If the link is shared with a live stream or scoring traffic, also consider setting an Upload rate limit so the Ingester stops competing with them.

If uploads keep getting stuck, run a Network Report and send the saved file to support.

Uploads are slow​

First check the obvious: the Upload speed in the Today strip, whether an Upload rate limit is set, and whether something else at the venue — a live stream, for example — is using the same connection.

Then run a Network Report: open SETTINGS and click Network Report in the Support group. It takes about 6 minutes and measures the connection to the upload storage — packet loss, latency, the route, and upload and download speed. Its summary suggests a Parallel connections and File chunk size for this connection.

Send the saved report to support, and to the venue's internet provider if it shows packet loss. See Network Report for how to read it.

note

The suggested values are not applied. Leave both settings on Auto unless support asks you to change them.

Uploads wait for the Network Report​

While a Network Report is running, uploads that have not started yet wait until it ends, so that they do not disturb the measurement; an upload already running carries on. The Queued header says how many jobs are waiting, for example 4 waiting for the Network Report.

This is expected. Uploads start again by themselves as soon as the report finishes or you cancel it. Encode requests are not affected.

ENCODING STALLED banner​

ENCODING STALLED: jobs are waiting in uploaded and no encode request has succeeded for 10 minutes. Check the workers and the backend.

Files are being uploaded, but the platform has not accepted a single encode request for 10 minutes, so uploaded videos are piling up in Queued as Waiting for encode.

  1. Is WORKER switched on? Encode requests are sent by the worker.
  2. Is the Ingester online? Check the Connected indicator in the top-left corner.
  3. Is the platform reachable? Look in the Log panel for failed encode requests. If the platform is down, there is nothing to fix at the venue: the Ingester keeps trying, and the banner disappears by itself as soon as an encode request succeeds.

If the banner stays for long, contact the Control Room and send the log files.

A job failed after its upload​

A failed job has been uploaded, but the request asking the platform to encode it failed three times in a row — the Ingester had already retried it twice by itself. The job is in Needs attention.

  1. Fix the cause if you know it — typically the platform was unreachable for a while.
  2. Click Retry Encode in the job's Next action column, or right-click it and choose Retry encode "your file". The file is not uploaded again.
  3. If that fails too, Re-run the job to upload it again from the start.

See Retrying the encode request.

Access denied when uploading a file​

An Access Denied message means the API key is missing, wrong, or not authorised for this backend.

  1. Open SETTINGS → API Configuration.
  2. Check the API key — retype it rather than editing it, to rule out stray spaces.
  3. Check that the Backend System matches the key. A JudoManager key will not work against Judobase.
  4. Click Save settings.
note

The change only takes effect after Save settings. Closing the dialog with Cancel discards it.

File upload failed – contest code not recognized​

Every video is linked to a fight by its contest code. For the upload to succeed, that code must exist in the backend the Ingester is pointed at.

The message means the backend was reached, understood the request, and found no such contest. The usual causes, in order of likelihood:

  1. Wrong Backend System. You are uploading a JudoManager contest while configured for Judobase, or vice versa. Open SETTINGS and check the Backend System dropdown — the API url should read https://datav2.judomanager.com for JudoManager and https://datav2.ijf.org for Judobase.
  2. Wrong partner (JudoManager only). The contest code must exist for that partner. Re-open the Contest Data form and select the correct partner.
  3. Typo in the contest code. Check it character by character against the source.
  4. The contest does not exist yet. If the draw or the schedule has not been published in JudoManager, its contests do not exist yet. Cancel the job, wait, and re-run it.

Please select a partner​

Shown when saving the Contest Data form for a JudoManager job without choosing a partner. The partner is mandatory for JudoManager because a contest code is only unique within one partner's events.

Choose the federation or organisation that owns the event from the Partner dropdown. If you are not sure which one, ask the Control Room team.

note

The Partner dropdown does not appear at all when the Backend System is Judobase — Judobase does not use partners.

note

This replaces the older IdPartner setting in settings.json. Partners are now chosen per job in the Contest Data form and are read from IngestPartners.json, so there is nothing to edit by hand.

API Configuration Incomplete​

Please select a Backend System (JM or JB) and provide both the API URL and API Key. All three settings are required for this process to run.

Exactly what it says: one of the three API Configuration settings is empty. The Ingester refuses to save incomplete settings and refuses to create jobs, because a job created without them could never be uploaded.

Choose a Backend System first — the API url fills itself in — then paste the API key.

Overlay Image URL Error​

The overlay image URL … must be a valid HTTP or HTTPS address. Local file paths are not supported.

The overlay image is fetched by the encoder, which runs in the cloud and cannot read your local disk. Upload the logo somewhere reachable over the internet and paste that address instead of a path such as D:\logos\sponsor.png.

Use the Open URL button to confirm the address really shows the image.

Failed to load Ingest Partners, and the program closes​

The file IngestPartners.json is missing or invalid. Application will now close. Contact support for help.

IngestPartners.json lives in the installation folder and lists the JudoManager partners offered in the Contest Data form. Without it, JudoManager jobs could not be given a partner, so the Ingester stops rather than running in a broken state.

Reinstall the Ingester, which restores the file. Do not edit it by hand.

An instance of the application is already running​

Only one Ingester can run at a time, because two copies would compete for the same job database. The first copy is still running — look for it in the taskbar or the notification area.

Closing while jobs are processing​

Some jobs are processing. Uploads in progress go back to new and upload again from the start on the next run; encode requests in progress go back to uploaded. Close anyway?

This is a warning, not an error. Click OK to close anyway — nothing is lost, but the interrupted uploads start again from the beginning the next time the Ingester starts. Click Cancel to keep it running and let the uploads finish. See Closing the Ingester.

Registration and licence problems​

See Registration and licensing for machine key changes, expired tokens and expired licences.

Log files​

Everything the Ingester does is written to disk, in far more detail than the Log panel shows. The Open logs button opens the folder directly:

C:\FairReplay\Ingester\logs
FolderContents
app\The main application log — one file per day. Start here.
thread\The detailed upload and HTTP log — one file per day. Needed for upload problems.
network\The saved Network Reports, one file per report.

Open them with Notepad or any text editor. Files are archived once they pass 100 MB, and the 4 most recent archives are kept, so a busy event does not fill the disk.

When reporting a problem, send the newest file from both folders rather than one of them — an upload failure usually needs the thread log to explain it.

Collecting information for a support request​

  1. The version from the title bar, e.g. 2026.06.01.337-be888ea.
  2. The Id job of the affected job, from the grid.
  3. The payload — right-click the job and choose Show payload for ….
  4. The log files — click Open logs and attach the newest file from both the app and thread folders.
  5. The exact wording of any error message.
  6. For slow or failing uploads, a Network Report — the file it saves in the network log folder.