Troubleshooting Automations

When a run did not do what you expected, run detail shows the status of each node and the items each node received and produced. An Error Handling rule routes a failed item to the node's Error path or stops it, and Retry Handling starts a new run that resumes at the failed node.

Skipped Run

The trigger produced no items, so no node ran. For a Scheduled, Manual, or Webhook trigger with a folder path, no file matched the folder path and pattern at the time the run started. Check the Folder Path, the Incoming File Pattern, and the Exclude Pattern, and, when the pattern uses date tokens, the Path Time Zone, because a token that resolves to yesterday's date matches yesterday's files. A Skipped run is not a failure and is not retried.

Skipped Node

The node received no items. An If, Switch, or Filter node before it routed every item to another path, or the node before it produced nothing. The item counts on the connections in run detail show how many items each path received. A Skipped node is the expected result when no item was routed to that path, and it is not an error.

Destination Already Exists

Copy File, Move File, Convert Document, Convert Image, Zip, Unzip, Encrypt With GPG, and Decrypt With GPG do not replace a file that already exists at the destination unless the node's overwrite setting is on, and it is off by default. Convert Document, Convert Image, Zip, Unzip, Encrypt With GPG, and Decrypt With GPG fail the item with processing-failure/destination-exists, and Move File fails the move for that file. Copy File skips the file and records it as skipped, so the older file remains at the destination and the run still finishes as Success. Turn on overwriting on that node when replacing the file is what you want, or change the destination template so that files do not collide, for example by adding {{ attributes.name }} or a date token to the path. When you did not expect a file there, check whether another Automation or a Sync writes to the same folder.

Same-Size File at the Destination

With If their names match but sizes are different on, Copy File still skips a file whose name and size both match the file at the destination, so a file that was updated without changing size is not copied. Turn on If both their names and sizes match as well to copy it. We recommend leaving If both their names and sizes match off unless files are updated without changing size, because with it on every run copies the files that have not changed.

not-found Errors

The file the node was going to process no longer existed at its path. Another branch of the same run moved or deleted it before this node read it, or another Automation, a Sync, or a person did. When another branch of the same run moved or deleted it, place the Move File or Delete File node after the nodes that read the original. When another Automation or a Sync moved or deleted it, apply the fixes in Two Automations or Syncs on the Same Files.

Send Email Without Custom SMTP

A graph with a Send Email node is saved only on a site with Custom SMTP configured, because Send Email sends through your own mail server. Configure Custom SMTP, or remove the node.

Transform Node Rejected at Save

Files.com rejects a Transform node when its script does not parse, when the script is known to return a single value rather than an object or a list, or when a destination is set and the script has no output directive. The message gives the position of the failing field in the JSON definition and ends with invalid_fts, invalid_transform_output_shape, or transform_output_required, whose meanings are listed with the other save-time codes. Fix the script in the node and save again.

Incompatible Connection Rejected at Save

A connection that would deliver data to a node that accepts only files, or files to Aggregate, fails at save with incompatible_item_kind, and when you add a node, the Builder shows which node types accept the item kinds the previous node produces. To pass data to a node that accepts only files, write it to a file first with a Transform node that has a destination, or with a Copy File node, which writes each data item as a .json file. To aggregate files, place a Transform node before Aggregate so that each file's contents become data.

Unapproved Agent Compute Command

An Agent Compute node runs only a command that is approved in the Agent's configuration on your server. Add the command to the Agent's approved list, then rerun the failed run from that node.

Remote Server Throttling

Some remote systems limit how many connections or requests they accept in a period, and the limit is set by the remote system, not by Files.com. When a run sends many files to such a server, its operations fail with the following error:

Errored while trying to Connect: 'Activity Limit Reached: The request has been throttled'

Files.com slows down its reconnection attempts and retries the operations later. Check the remote system's connection limits and activity logs to find the threshold, then reduce how many files a run sends with the trigger's File Limit, or schedule runs further apart.

Folder Over the File Limit

A run whose trigger selects files from a folder over the file limits fails, and selection slows down well before the folder reaches them. Organize the folder into subfolders or set File Expiration on it to keep it under the limit.

Zero-Byte Files on a Remote Server

Some remote systems do not accept a file with no contents, so a run that copies a zero-byte file to such a server fails for that file. Check the remote system's documentation for its handling of empty files, and, when the remote system cannot accept them, use an If or Filter node to keep them off the branch that writes to the remote system.

Remote Server Connection Failure

An operation on a Remote Server that times out or cannot connect fails the item with a service-unavailable error. Confirm that the Remote Server connection is active and that the network between Files.com and the remote system allows the connection. For a remote system that is unavailable for a period, turn on Retry Handling, so that the run resumes at the failed node after the interval you set.

Two Automations or Syncs on the Same Files

When two Automations, or an Automation and a Sync, act on the same folder, either can move or delete a file before the other reads it. The signs are not-found errors, processing-failure/destination-exists errors, and files that are skipped although nothing else is wrong. The risk is highest when one of the two moves or deletes files after copying them, and on a Remote Server Mount, where a file can still be written by another system while a run reads it. Give each Automation and Sync its own source and destination folders, use different file patterns so that they never select the same files, or combine the steps in one Automation, whose nodes run in the order you connect them.