Lake troubleshooting

When the Lake refuses an action or a load fails, it shows a message that names the problem. Find the message in the tables below to see what it means and what to do next.

Read the full error text

Open the table and go to its Sync Log.
Click the Error Stack cell of the failed run. The window is titled Error Stack Trace, or Runner Diagnostics when the run has no stack.
Click Copy.

For a datasource with a red Failed badge, click the badge to open its error logs, then click Copy.

Connections

You see	What it means and what to do
Name is required or Name must be at least 3 characters	The Name on the General step is empty or too short. Type a name of three or more characters.
Code is required, Code must be at least 2 characters or Code must contain only letters and numbers	The Code is empty, too short or contains anything other than letters and numbers. Use two or more letters and numbers.
This name is already in use or This code is already in use	Another datasource has the same name or code. Choose another.
A DataSource with the code already exists. or A DataSource with the name already exists.	The save was refused because the name or code is taken. Choose another.
Host is required, Port is required, Port must be a whole number, Port must be 65535 or lower, Database is required, Service name or SID is required, Username is required, Password is required or Tenant ID is required	A required connection field is empty or invalid. Fill in the field.
Username or client ID is required or Password or client secret is required	The sign-in field is empty for SQL login or for a Microsoft Entra service principal. Fill in the field.
No host found. Paste a JDBC URL, a URI or host:port/database.	The pasted text is not in a supported format. Paste one of these formats, or type the fields one by one.
No host found. Paste a mongodb:// or mongodb+srv:// URI.	The pasted text is not a MongoDB URI with a host the parser can read. Check the URI, or close the paste row and type the connection fields by hand.
Enter only the cluster host name - SRV records supply the port	An SRV (Atlas) connection's host field contains a port or a comma. Enter the cluster host name alone; SRV records supply the hosts and port.
Connection failed, with Error Details: on hover	The connection test from the form footer failed for the current settings. Point at the row to read the Error Details message, correct the settings, then test again.
Connection test failed	The server's reason follows. Check the host, port, firewall and credentials. The test is optional, so you can continue without it.
The host changed. Enter the password again. or The username changed. Enter the password again.	The stored password was cleared because the host or user name changed. Type it again. See Connect a SQL database.
The database URL changed. Enter the service account key again.	A Google connector cannot reuse the stored key after its database URL changes. Upload or paste the key again. See Connector secrets.
Enter the signing secret again: this connection has no stored signing secret to keep. or The sender uses a different scheme now, so the stored secret cannot be reused. Enter the token.	The webhook sender scheme changed, so the stored secret cannot be reused. Enter the new token or signing secret. See Webhook and RabbitMQ.
Enter the secret access key again: the endpoint, region, bucket or access key ID changed	The bucket settings changed, so the stored secret access key no longer applies. Enter it again. See Connect Amazon S3.
Enter the password (or key) again: the server changed	The server address changed, so the stored value was cleared. Enter the password or key again and test the connection.
XML data sources are no longer supported. or The "…" connector type is no longer available; create a Custom connector instead.	The source type was removed from the product. Create a source of a supported type, or build one with Connect Anything.
sourceTimezone must be a valid IANA timezone identifier (e.g. Europe/Berlin)	The Timezone is not a full zone name. Pick a zone from the list; a fixed offset such as an hours-from-UTC value is not accepted. See Timezones and datetime values.
No time zone matches followed by your search	The search in Timezone matches no city, country, zone name or offset. Search again with a nearby big city, the country or the offset, such as `+4`.
Detected: … (source clock …)	A connection test detected the source timezone and the section suggests it. Click Apply to select the detected zone, or pick another zone in the list.
Detected instants in the sample - use … + Convert or Detected instants in the sample - use … + Keep as written	A sample of a Connect Any API source holds datetime instants, so the step suggests a zone and a mode. Click Apply to accept both, or ignore the line.
Already-loaded rows keep the old datetime semantics - the new settings apply to future loads only (full-reload tables self-heal on the next run).	You changed the datetime settings of a saved source; loaded rows keep their old values. Run a Full Reload if the loaded history must use the new settings.
Changing database for PostgreSQL connections is not permitted or The database of a Postgres source cannot be changed after it is created.	You cannot change the database of a saved PostgreSQL connection. Create a new datasource.
Instant Client is not available while SSL/TLS is on. Turn SSL/TLS off under Security to use it. or Not available with the Instant Client driver. Switch Driver to Built-in under Advanced to use SSL/TLS.	The Oracle Instant Client driver cannot use SSL/TLS. Turn SSL/TLS off, or keep Driver on Built-in.
Azure SQL Database encrypts every connection; this cannot be turned off.	Azure SQL Database always encrypts, so the form shows the Always encrypted badge instead of the SSL/TLS switch. No action needed.
Could not read the certificate file: …	The certificate file could not be read. Check that it is a readable PEM (`.pem`, `.crt`), paste the certificate text into the field instead, or choose another file.
Could not read the key file: …	The file chosen for Private key (PEM) cannot be read. Pick a readable `.p8`, `.pem` or `.key` file again.
Account is required, Warehouse is required, Private key is required or Password is required (Snowflake)	A Snowflake field is empty. Fill it in; with Method on Key pair (recommended), upload the key, with Password, enter the password.
Do you wish to Revert all changes? or Do you wish to Reset form?	You clicked Revert with changes in the form. Confirm to put the form back, or cancel to keep your entries.
Unsaved Changes, with You have unsaved changes. Are you sure you want to close? All changes will be lost.	You closed the form or went back to sources with pending changes. Confirm to discard them, or cancel to stay.
No data sources available	The New menu has no source types to offer. Open the picker with Add datasource in the header; if the list stays empty, ask your administrator.
The connection direction of a data source cannot be changed	A save tried to change the datasource's direction, which is fixed at creation. Create a new datasource for the other direction instead.
Failed to create connection, Failed to update connection, Failed to save connection or Failed to duplicate connection	The save failed and the server gave no reason. Check the form for field errors, then try again.
Code and name cannot be updated	A saved datasource keeps its name and code. Create a new datasource instead.
File data sources cannot be duplicated	Create a new file datasource instead.
This code is reserved by a file upload in progress; choose another code	A file upload is using this code. Choose another.
Test the connection and trust the host key before saving	An SFTP source saves only after a successful test and a trusted host key. Test the connection, trust the host key, then save.
Private key authentication is available for SFTP only	FTP and FTPS accept a password only. Use a password, or choose SFTP.
Failed to load source menu	The Lake cannot read the source's structure. Check the connection with Test connection in the datasource form, then select Retry.

Connector forms

You see	What it means and what to do
Service account key is required or Invalid JSON format in service account key	The key is missing or not readable JSON. Drop or paste the whole contents of the service account's JSON key file again.
Invalid JSON file	The file chosen with Choose File cannot be read as JSON. Pick a service account's JSON key file, or paste its contents.
Project ID is required, Service account email is required, Service Account Email is required or Invalid email format	Enter the Google Cloud project and the service account address, such as name@project.iam.gserviceaccount.com.
Property ID is required	The Property ID on the Google Analytics form is empty. Enter the Analytics property id, for example 123456789.
Differs from the key's project (…) or Differs from the key's client email (…)	The value is not the project or email the key belongs to. The note is a warning only and does not block the save.
AWS Access Key ID is required, AWS Secret Access Key is required, AWS access key ID is required, AWS secret access key is required or Region is required	The access key pair or region is incomplete. Fill in the fields and submit again.
Output location is required or Use an S3 location such as s3://my-bucket/athena-results/	Enter an S3 address that starts with `s3://` and names a bucket, with no spaces.
Invalid query pattern detected	The load's query ends with a semicolon or contains `INFORMATION_SCHEMA`; either is refused. Rework the query without them and run the load again.
Table not found: …, Cannot find table key for Firestore table: … or Cannot find table key for Realtime Database table: …	The load cannot resolve the table it was set up for. Check that the table still exists in the source, run Sync Schema, then load again.
Folder URL is required or Enter a valid URL (e.g. https://bucket.s3.region.amazonaws.com/prefix/)	Paste the full address of the bucket folder, for example https://my-bucket.s3.eu-central-1.amazonaws.com/exports/, then save again.
This format is no longer supported - choose another one, No longer supported - choose another format, Pick a supported format to see which files are read. or a legacy format marked Kept from the saved settings, no longer offered for new sources	Choose a supported File type and Format on the Format step. The kept legacy format still works for this source but is not offered for new ones.
Changing the format changes the existing table structure. Run Sync Schema after you save.	Save, then run Sync Schema on the datasource. See Run, stop and troubleshoot loads.
Client ID is required, Client Secret is required or Refresh Token is required	An OAuth field is empty. Enter the value from Google Cloud Console, or the refresh token produced by the OAuth authorization flow.
Spreadsheet ID is required or Invalid Spreadsheet ID format. Please use the ID from the Google Sheets URL	Paste the spreadsheet link into Or extract from URL and click Extract ID, or type the bare ID from the link.
Sheet name cannot be empty or Range cannot be empty	A sheet row is incomplete. Give the row a name and a range, then save it.
Sheet name must be unique or Sheet name contains invalid characters	Sheet names must be unique and use plain letters and digits. Rename the row.
Do not include sheet name in range. Sheet name is specified separately, Invalid range format. Range must be in format like A1:B10, B2:F10, or AA1:BC100, Single cells not allowed. Use a range format like A1:A1 for single cell or A1:B10 for multiple cells, Range is too wide. Maximum 1000 columns allowed or Range is too tall. Maximum 100,000 rows allowed	Fix the range: a cell range with two corners, not one cell, within the size limits. The sheet name goes in its own field.
Save at least one valid sheet, Each sheet must have both a name and a range, At least one sheet must have both name and range or Cannot remove the only sheet	A connection needs at least one sheet row with both a name and a range; add one before removing the last.
Database URL is required or Enter a URL such as https://my-project-default-rtdb.firebaseio.com	The Database URL is empty or cannot be read as a URL. Enter or paste the database's address in the form shown.
API Key is required	The connection was saved without an API key. Paste your key into API Key and save again.
Locations are required, Locations cannot be empty or Location is required	The location list is empty or a row has no value. Add at least one row with Add beside Locations and fill in its value.
Location name cannot be empty, Location value cannot be empty or Location name must be unique	A location row is missing its name or value, or reuses a name. Fix the row, then save it.
Data source configuration not found for table: …, API key not found in configuration for table: … or No locations found in configuration options for table: …	The saved Visual Crossing setup is missing for this table's load. Open the source, fix the configuration and save, then load again.
Date must be between 2016-01-01 and current date or Sync Days must be between 1 and 367	The date or the number of days is outside the accepted range. Choose a date from 2016-01-01 to today, or a value from 1 to 367.
Invalid XML file: Unable to parse.	The file could not be parsed as XML, so it was not added to the list. Check that it is a complete, valid XML file, then add it again.
Error reading file.	The file could not be read while it was being added. Add it again; if it keeps failing, check that the file is not damaged.
At least one file is required for upload connection type	Saving an upload source with an empty file list. Add at least one file to the drop zone, or create the source with SFTP as the Connection Type.
Processing method is required for SFTP connection type	The SFTP form was submitted without a processing method. Pick a Processing Type and save again.
Host is required for SFTP connection type, Port is required for SFTP connection type, Username is required for SFTP connection type, Password is required for SFTP connection type or Remote Directory is required for SFTP connection type	A required SFTP field is empty. Fill it in (Port defaults to 22), then save again.
Directory must start with '/' or not end with '/' or Directory must start with '/' and must not end with '/' (e.g. /folder1/folder2)	The Remote Directory must start with / and must not end with /. Enter a path like /folder1/folder2.
Upload processing method is required for upload connection type, Processing method is required for upload connection type or Upload processing type is required when upload processing method is 'together'	Choose how the workbooks are processed (for example each file's sheet as separate table), then save again.
One save uploads at most 300 files; any beyond that were not added. Save, then add the rest.	More than 300 files were added at once, so the extra files were not added to the list. Save the source with the files that were added, then add the rest in another save.
File Too Large, with The uploaded file exceeds the maximum allowed size. Please upload a smaller file.	An upload hit the size limit. Reduce the workbook size (up to 1 GB per workbook is accepted) and upload it again.
AMQP login ✓ · Management API ✗ …	The AMQP login worked but the management API did not answer. Check the management API port and the management plugin, or continue and define the tables manually.
The last test failed. Fix the broker settings and refresh.	Discovery has no fresh test result. Fix the broker fields, run Test connection, then Refresh.
The queue mode or exchange changed. Refresh to list again.	The queue mode or exchange was edited after the last test. Click Refresh.
Management API unreachable: … Define the tables manually below. or RabbitMQ management API is unreachable on port …	The management API did not answer, so discovery cannot list queues or routing keys. Fix the port or the plugin, or define the tables manually.
No queues on this virtual host yet. Create them on the broker, or define the tables manually below. or No queues are bound to "…" yet. Bind them on the broker, or define the tables manually below.	The virtual host or the exchange has no queues. Create or bind queues on the broker, or define the tables manually.
Enter the exchange above, then refresh to list its routing keys.	No exchange is set, so there is nothing to discover in managed-queue mode. Enter the Exchange and click Refresh.
The tables defined manually below replace this list - discovery is skipped.	Manual tables take precedence over discovery. Remove every manual row to go back to discovery.
Table name is required	A table name is missing: the name a manual RabbitMQ table row should use, or the name the webhook import should use (the field starts as `table`). Fill it in.
Table names must be unique, Queue name is required, Column name is required or Pick a column type	A manual table row is incomplete. Give each row a unique table name, fill in the queue it reads from, and give every column a name and a type.
Exchange is required - messages are published into it	A writeback destination has no exchange. Enter the exchange to publish into.
Header name is required - the sender decides which header it uses	A Custom HMAC signature sender has no header name. Enter the header the sender puts its credential in.
Token is required or Signing secret is required	A token or signature sender has no secret. Paste the value or click Generate.
Set a number of days, or keep events forever	Delete after N days is selected but Days is empty. Enter the number of days, or switch to Keep forever.
Copy it now - it is masked after you leave this screen and cannot be shown again. or Copy it now - it is masked after saving.	A generated value is shown once. Copy it into the sender before you leave the screen or save.
A previous secret is still accepted. or The previous secret stops being accepted when you save.	A secret rotation window is open after a change. Click Stop accepting once the sender no longer uses the old value, then save; Undo keeps the old value accepted.
Nothing is verified. Anyone who learns the URL can write events into this connection. Only use it when the sender cannot sign or send a token.	No authentication checks what arrives. Choose a token or signature sender when the sender supports one.
Stripe and Slack are not supported, because they pack several values into one header.	Use Static token in a header or Custom HMAC signature when the system can send one.
The endpoint URL appears once the connection is saved.	The Endpoint URL card has no address before the first save. Save the connection; the URL appears at once.
The secret is never shown here - paste the one set on this connection.	The guide's example carries a placeholder, not the stored secret. Put the real token or signing secret into the sender.
Expired events cannot be replayed. A table on this connection using a truncate-insert sync will lose that history on its next full reload.	Delete after N days removes older buffered events. Choose Keep forever, or accept that the table loses that history on its next full reload.
Fixed once the connection is saved: the importer bakes it into the core table name.	The table name cannot change after saving. Keep it, or add a new connection with the name you want.
Soon	The card is listed, but the connector is not enabled on this platform, so its connection form cannot be opened. Use another enabled source type, or ask your administrator whether this connector can be enabled.
Not available on this platform yet.	The featured Connect Anything or Connect Any API card is not enabled on this platform. Use another source type, or ask your administrator whether this connector can be enabled.

Files and Excel

You see What it means and what to do

Mirror deletes needs the files left on the server; turn it off or leave the files after copying The two file options conflict. Turn off mirror deletes, or leave the files on the server after copying.

The move-to folder must be outside the remote folder Choose a move-to folder that is not inside the remote folder.

Table name must be 1-63 characters: lowercase letters, digits and _ A table name is not valid. Rename it in the Tables step.

The source is fixed once the data source is created. The transport chosen when a file datasource is created cannot be changed later. While uploads are still in the form, remove them to switch; otherwise create a new datasource.

Remove the uploaded files to switch to SFTP / FTP or an S3 bucket. The transport cards are disabled while files are uploaded. Remove the uploaded files on the Source step, then choose another option.

Code is locked to "…" by the uploaded files. Remove the files to change it. The code (and name) cannot change while uploads sit in the form, because the files are stored under the code. Remove the uploaded files, or save the datasource.

Upload completed; the file is being prepared The last part of the upload landed; the file is converted to UTF-8 and checked before it can load. Wait for the file to turn ready.

Some files are being replaced and cannot be deleted until the new version is ready You tried to delete a stored file, or the row beside it, while its replacement is still being prepared. Wait until the new version is ready, then delete whichever row you no longer need.

This upload session belongs to another user The upload session was started by a colleague, and only that person can use its files. Start your own upload, or ask that person to finish or save the datasource.

The upload session expired; upload the files again The upload session (which lasts a day) was gone before the save finished. Upload the files again.

Replace file? or … already exists. Replace it? A file already sits at the same stored path. Choose Replace to swap it once the new version is ready, or Keep current to drop the new copy.

"…" is already uploaded; delete it first to replace it A second upload of the same path was refused because Replace was not confirmed. Confirm Replace in the prompt, or delete the stored file first.

Trust the server host key before continuing Saving an SFTP source needs a trusted host key. Run the connection test, check the fingerprint, then click Trust this host.

The server host key changed. Review it and trust the new key before continuing. The server presents a key other than the trusted one; if nobody re-installed or replaced the server, this can be an attack. Check with your server owner, then trust the new key only if the change is expected.

The server certificate is not issued by a trusted authority. Pin it before continuing. The FTPS server uses a self-signed or private CA certificate, and previews and syncs reject it until it is pinned. Check the fingerprint with your server administrator, then pin the certificate.

The server certificate changed. Review it and pin the new certificate before continuing. The FTPS certificate no longer matches the pinned one; if nobody renewed or replaced it, this can be an attack. Check with your server owner, then pin the new certificate only if the change is expected.

The server changed - run the connection test again The server or its sign-in details changed after the last test. Run Test connection again before continuing.

Upload at least one file, Retry or remove the uploads that failed or Wait until every upload has finished processing The step is not ready to continue. Upload a file, retry or remove the failed uploads, or wait for processing to finish.

Converting to UTF-8 and checking the content. The file is loaded once it is ready. The Processing status badge on a stored file. Wait; the file turns Ready and the next load can read it.

Loaded, but the content is identical to … The Duplicate badge; both stored files hold identical content and are read, so delete one to avoid double rows.

A new version of this file is on its way. Loads keep using the current file until the new one is ready. The Replacing… badge; a replacement upload is being prepared. Keep working; loads use the current file until the new version is ready.

The new version could not be prepared. The current file is still loaded. The Replace failed badge; the replacement could not be stored. Check the file and replace it again, or keep using the current version.

No date found in the file name or path. Date-incremental loads skip this file. The file has no date while a table group expects one, so date-incremental loads skip it. Give the file a name that matches the group's date format, or turn the extraction off on the group.

No table group matches this file, so no table loads it. No table group matches this stored file. Align the file with a group's pattern, or add a group for it on the Tables step.

No files yet. Drop files above to add them. The Files list is empty on an upload source. Add files with the drop area above the list.

No files fetched yet. Fetch now copies the files from the server. The Files list is empty on a server or bucket source and no fetch has run. Click Fetch now to copy the files from the server or bucket.

No files match the search or status filter. The list is empty because a search or a status filter is on. Clear the search, or set the filter back to all statuses.

Delete … file(s)? The confirmation before stored files are removed; the title counts the chosen files. Deleting removes the stored copies; loaded rows stay in the Lake until a run reloads their days, a workbook's sheets go with it, and files still on the server or in the bucket are fetched again by the next sync. Click Delete to confirm.

Files could not be deleted, Reload could not start or Changed days could not be checked The request failed. Read the description and try again.

Files could not be loaded The file list request failed. Read the message and retry with the refresh icon.

Files changed outside the sync window Scheduled runs only reload the days inside each table's sync window. Use Reload affected days on the table's row in the banner so the Lake tables match the files.

Reload affected days?, with … of … are reloaded from the current files. Each day is replaced as a whole; days with no files left are emptied. The confirmation before a day reload. Confirm with Reload days to start; closing the dialog cancels before anything runs.

More than 366 days changed - too many for one reload. Run a full reload of this table instead. The changed days for one table exceed the per-reload limit. Run a full reload of that table instead of reloading days.

Some of their days are outside the sync window. Reload them from the banner above the list. The deleted files included days that scheduled runs do not reload. Use Reload affected days on the banner above the list.

Too large to download at once The files are too big to download together. Download them one by one.

Select items to import Nothing is ticked in the source panel. Tick the tables you want to import.

The file is not valid UTF-8 (invalid bytes after byte …); choose its encoding in the format settings The stored copy is not UTF-8 text. Choose the file's Encoding on the Format step and save; the stored files are re-read with the new setting. See Load files.

The uploaded data is missing; upload the file again or Extracting the archive or workbook was interrupted; upload it again The uploaded copy did not survive preparation, or the process was interrupted. Upload the file again.

The archive holds no … files The archive contains no files of the accepted types. Check the archive, then upload it again.

The archive holds … files; split it into archives of at most 10000 One archive holds too many files. Split it into smaller archives and upload them one at a time.

Password-protected archive entries are not supported The archive has encrypted entries. Re-create it without a password, then upload it again.

The archive entry uses compression method …, which is not supported; re-create the zip with standard (deflate) compression The archive was made with an unsupported compression method. Re-create it with standard compression.

Not enough free disk space on the server: … free, … needed for this file; free up space or ask an administrator The upload cannot be stored. Free up space and upload again, or ask your administrator.

Re-reading the uploaded files with the saved format settings failed: … or Re-reading the workbooks with the saved Excel options failed: … The stored files could not be read with the new settings. Fix the setting named in the message and save again.

The uploaded file contains no data rows. The Excel file has a header row but no data. Add rows, then upload it again.

The uploaded file is missing the column(s): … The file lacks columns the table needs. Add them, then upload it again.

First row must contain non-empty column headers Give every column in the first row a name.

Excel exceeds the maximum allowed rows. Maximum: 1000, provided: … The file has more than 1,000 data rows. Split it into smaller files.

At most 50 tables per schema can be imported in one file. … Reduce the selection for the schemas named in the message, or split the file.

… of … row(s) could not be resolved, so nothing will be imported. Some rows could not be matched, so the import did not start. Check the skipped rows for the reason on each.

Import failed: followed by the error The Excel import stopped partway, and some tables may already exist. Upload the file again. Tables already created show as already imported, not duplicated.

The preview looks separated by … The preview sample suggests a delimiter other than the one set. Click Use it to apply the suggestion, or keep your own setting.

Type the delimiter; … is used until then The Other… delimiter field is empty, so the last used delimiter is applied for now. Type the delimiter you want, or switch back to Auto-detect or a preset.

Removed when a file is stored. Files uploaded so far are re-read with this setting after you save., Removed when a file is stored. A change applies to newly added files only; stored files keep their current footer handling., Files are converted to UTF-8 when stored. Files uploaded so far are re-read with this encoding after you save. or Files are converted to UTF-8 when stored. A change applies to newly added files only; stored files keep their current conversion. Footer-row and encoding changes apply to newly added files; in a new source the files already uploaded are read again after you save, and on an existing source stored files keep their current handling.

These options apply when you save. The preview reads the sheets as they are. Excel options do not change the preview; they are applied at save time. In edit mode only the workbooks whose options change are read again.

Matches none of the current workbooks. or Matches …, but … already apply first. The override pattern matches no workbook, or earlier rows already claim those workbooks (the first matching row wins). Change the pattern, or move this row up.

Same options as above, so this row changes nothing. or Same as row …, so this row never applies The override row repeats the default or an earlier row. Change its options, or remove the row.

Enter a workbook file name or a pattern, Match the file name only, without folders or Keep it under 255 characters Fix the override's match text: a workbook name or a pattern with * and ?, the file name only, and under 255 characters.

At most 200 rows. Use a pattern to cover several workbooks at once. The override list is full. Use wildcard patterns to cover several workbooks per row.

Workbook names appear here once files are added. or No other workbooks yet. Add files on the Source step, or type a pattern. The workbook picker has no names to list yet. Add files on the Source step, or type a pattern.

Add files on the Source step to see a preview, Waiting for files to finish processing or No rows to preview The preview has nothing to show: no files yet, files not ready, or no rows in the ready files. Check the format settings and the file itself.

Preview failed The preview request failed and the server sent no message. Click Refresh; if it keeps failing, check the connection and the file.

The delimiter and header are detected from the preview, which has not loaded. Wait for the preview or set them here. Saving was blocked because the preview never loaded and the delimiter or header still depend on detection. Wait for the preview to load, or set Delimiter and Header on the Format step yourself.

Tables of this source are already imported, so the mode cannot change. or Tables of this source are already imported, so the layout cannot change. The Tables step keeps its mode or sheet layout once a table of this source is imported. Choose it before importing tables.

Add at least one group or Choose how the sheets become tables The union mode has no group, or the workbook source has no sheet layout. Add a group, switch to Each file separately, or pick one of the layouts.

No file matches this group or No file matches yet; files may arrive later The union group matches none of the known files; on a server or bucket source this only warns, and the group fills as files arrive. Widen the selector, or add the files on the Source step.

… matching file(s) also match another group A file falls into two groups, so each group loads it into its own table. Narrow one selector so every file matches a single group.

Enter a label or Table name "…" is used by another group The group has no label, or two groups would load into the same table name. Type a label, or rename one of the groups.

Some captured table names produce the same slug A dynamic selector captures values that turn into one table name. Adjust the regex so the captured values stay distinct.

Group "…" has imported tables; delete those tables before removing the group The group's tables are imported, so the group cannot be removed. Delete those tables in the Lake first.

Imported tables keep their name; only the label can change The table name is fixed once its table is imported. Change the label; the stored table name stays.

Enter a regular expression, Lookahead and lookbehind are not supported (RE2), Backreferences are not supported (RE2) or Use exactly one named group (?...) for the table name
Fix the group's Regex selector: a pattern without lookarounds or backreferences, with exactly one (?<table>...) group.
No date pattern was found automatically; enter the format below., Enter a date format, for example yyyyMMdd, Unsupported pattern letter. Use yyyy, yy, MM, dd, HH, mm, ss, Each pattern token may appear only once, The format needs a year (yyyy) or A day (dd) needs a month (MM) Fix the group's date format: use only the listed tokens, once each, with a year, and a month when a day is used.
Every column is String, so no value can be rejected. With all-String columns there is nothing to reject, so the bad-values policy is disabled. Choose Detect types, or a default that types columns, to set a bad-values policy.
… tables match this group; columns and sample rows are shown for the first 10. or One preview shows columns and sample rows for at most 20 tables; this table shows file counts only. The preview samples part of a large group. Narrow the selector to sample a later table; the load still covers all of them.
No preview for this group yet There is nothing to sample for the group. Check the selector and wait for a matching file to become ready.
Excel transform exceeded the …s limit, Excel transform exceeded its memory cap of … MiB, Excel transform failed, most likely at its memory cap of … MiB, Excel transform was killed by signal … (most likely out of memory) or Excel transform failed The conversion ran out of time or memory. Split the workbook into smaller ones, or turn transpose off for tall sheets, then run again; a generic failure carries its reason in the error detail.
Failed to read source files The stored workbooks could not be read. Check the stored files on the source's Source step, then run again.
Transform produced no TSV files The run finished with no converted sheets, usually because no sheet held data. Check the workbooks; a sheet with no rows under its header produces no table.
Configured file "…" is not present among the stored files A workbook the source expects is missing. Add the file back on the Source step, then run again.
Transform cancelled by user The run was stopped before it finished; nothing from the stopped attempt is kept. Run again when ready.
Workbook exceeds the interactive size limit A workbook uploaded in a Mind chat is larger than the 15 MB the chat lane allows. Upload a smaller workbook in the chat, or load large workbooks through an Excel Workbook source in the Lake.

Remote files and servers

When a fetch from an SFTP, FTP or S3 source fails or skips a file, the detail is in the Last sync messages popover; see Stored files.

You see What it means and what to do
Fetch now works only for SFTP, FTP and S3 file sources Fetch now was requested for a source that is not a server or bucket source. Use it on an SFTP, FTP or S3 source; upload sources load what you uploaded.
Files of "…" are already being fetched A fetch is already running for this datasource. Wait for it to finish; the file list refreshes as files arrive.
The remote listing was truncated; counts cover the listed files only The folder holds more files than one listing can return, so counts cover the listed files only. Narrow the remote folder or the folder prefix, or continue: the fetch still copies what it can.
… cannot be used for a file source The server address points at the platform itself, a loopback or a link-local address, so it is refused. Enter the real server's host name or address.
Last sync failed, with Fix the cause and fetch again. Files fetched before the failure stay available. The last fetch for this source failed. Open Last sync messages for the detail, fix the cause, then click Fetch now.
Last sync finished with warnings, with Some files or post-processing steps reported problems. The fetch finished, but something inside it reported a problem. Open Last sync messages and check the affected files.
No sync status yet No fetch has run yet for this source. Click Fetch now, or wait for the next scheduled run.
Sync status not refreshed, with The sync ran longer than the status check. Refresh to see where it stands. The status check gave up before the fetch settled. Click Refresh beside the card.
Fetch could not start The Fetch now request was refused or failed. Read the description, then try again.
Authentication failed for user "…" on … The server refused the sign-in. Check the user name, password or key, then test the connection again. See Load files.
Cannot reach … (…) The server or bucket could not be contacted. Check the address and the network, then try again.
The SFTP connection was closed or The SFTP connection was lost: … The server dropped the connection mid-copy. Fetch again; files copied before the drop stay available.
No trusted host key is stored for this SFTP source; test the connection and trust the host key Saving or fetching needs a trusted server key. Test the connection, check the fingerprint, then trust the key.
Timed out connecting to … The server did not answer in time. Check the address and the network, then retry the fetch.
The bucket "…" is in region …; set the region to … The bucket lives in another region. Set Region to the one named in the message. See Connect Amazon S3.
The access key ID is not known to … The access key does not belong to this store. Check the key ID, and the region or endpoint it is used with.
The secret access key does not match the access key ID The key pair does not match. Enter the secret access key again.
The bucket "…" does not allow anonymous access; enter an access key The bucket is not public. Switch Access from Public bucket (anonymous) to an access key.
Reading … was denied; the access key needs s3:GetObject The key cannot read this object. Give the key read access to the objects, or use another key.
Skipped …: the name contains line breaks or NUL characters The file's name cannot be stored. Rename the file on the server, then fetch again.
Skipped …: "…" is already stored for … A stored file already claims this name or path, so the fetch skipped the duplicate. Delete or rename the stored copy first if you need this one.
The fetch was cancelled before this file was stored The fetch stopped before this file was copied. Fetch again to pick it up.
None of the new or changed files could be fetched: The fetch failed for every candidate file. Read the lines under the message, fix the cause, then fetch again.

Folders

You see What it means and what to do
Name is required You tried to create a folder without a name. Enter one.
A folder with this name already exists Enter a name that no other folder uses.
Failed to create folder The create request failed with no server reason. Try again.
Failed to move datasource The folder move failed with no server reason. Try the drag again.
Failed to delete folder The folder delete request failed with no server reason. Try again; if it repeats, check your permissions.
Cannot delete folder, with Please remove or move all data sources from this folder first. The folder still holds datasources. Move them out, then delete the folder.

Importing tables

You see What it means and what to do
Import in progress An import is already running, so Import is disabled. Wait for the run to finish; the panel unlocks itself.
Please wait for the import to complete The source panel cannot be closed while an import runs. Wait for the import to finish, then close the panel.
Disabled in batch select mode View Source is unavailable while batch select is on. Leave batch select, then open the source panel.
Select a datasource, database, or schema or Only available for datasource, database, or schema View Source needs a datasource, database or schema row. Select one of those instead of a table, column or folder.
The schema contains too many tables for processing, maximum number per import is 500, consider splitting One import can take at most 500 tables. Reduce the selection, or split the import into several runs.
No matches here yet No result has loaded so far, but more pages exist. Click Load more to keep paging.
No results found for "…" or term must be longer than or equal to 2 characters Nothing matched; the search needs at least two characters and never lists tables already in the Lake. Check the spelling, type more characters, or confirm the table is not already imported.
… database couldn't be searched or … databases couldn't be searched One or more databases did not answer during the search, so results may be incomplete. Run the search again to retry the skipped databases.
Invalid search cursor The page of results you asked for has expired. Run the search again.
Data source … not found The datasource cannot be read (it was removed, or you have no read permission for it), so its panel and search are refused. Close the panel, refresh the tree and check your access.
All tables already imported or All tables deployed Every table is already in the Lake. Manage the tables in the Lake tree; nothing to import here.
No tables available The source exposes no importable tables. Check the source itself; if it should have tables, test the connection in the datasource form.
Deployment in progress The datasource's tables are being deployed, so the panel is not interactive. Wait for the deployment to finish, then reopen the panel.
Not found or Ambiguous (Excel review) A row names a table the source does not have, or several source tables match once casing is ignored. Correct the name, or copy the table's exact name into the row.
Already imported or Duplicate (Excel review) The table is already in the Lake, or the file lists it more than once; it is skipped, not duplicated. Keep only the first row.
Incomplete (Excel review) Database, Schema and Table are all required; fill in the missing column for that row.
No data returned from the query, cannot validate The custom table's query returned no rows, so it cannot be validated or created. Change the query so it returns rows, then run it again.
Data source with id … (…) does not support SQL queries A custom table was attempted on a datasource type that does not support SQL queries. Use View Source to import its tables instead; custom tables exist only for SQL databases and Snowflake.

Lake screen and tree

You see What it means and what to do
No Data Sources Found, with Connect a data source to see statistics here The dashboard has no datasources to show. Connect one with Add datasource. See Add a data source.
No connections in this folder A folder group on the dashboard holds no datasources. Move datasources into the folder, or add one with the card's Add data source.
… failed (toast) A datasource created moments ago did not become ready; the toast description carries the error. Open the datasource to read the error, then find it on this page.
Click to open error logs The tooltip of the red Failed badge. Click the badge to open Error Details and read or copy the error message.
Error Details - … The error window opened from a red Failed badge. Read the message; Copy copies it. If Unable to copy appears, click Copy again.
Setup incomplete The datasource holds no tables and its setup is not finished. Click Continue beside it to reopen the connection form and finish the setup.
Import tables A datasource with no tables and a finished setup offers this link; it opens the source panel. Click it to choose the tables to import. See Import tables.
Aborts the running sync and restores the data to its state before the sync started - nothing partial is kept. Re-syncing afterwards works normally. What Stop Sync does. Right-click a datasource while it shows Processing and choose Stop Sync; the row shows the muted Stopped chip afterwards.
… - not migrated A column is not created in the Lake table; its row label is struck through. Open the table's column setup and switch the column's migrate flag on, then deploy again. See Define table columns.
… - not synced A column is not loaded on each run; its row label is faded. Switch the column's sync flag on in the table's column setup. See Define table columns.
… items or 99+ The children-count badge beside a row; it stops counting above 99. Point at the badge to read the exact number in its tooltip.
No items to display The tree has no rows to show. Connect a datasource first. See Add a data source.
No items match the current filters A filter hides every row in the tree. Clear the filters. See Search and filter the Lake.
Idle The tree stopped checking a datasource that stayed Processing through its poll budget (about three and a half minutes); this is a backed-off state, not a failure. Use Refresh on the row to check its real state; if it is still working, the checks resume.
Stopped A sync you stopped; it is stored as failed with the user-abort marker and shown as a muted chip, not the red error badge. Start a new sync with Sync Schema when you want to sync again.
Synchronizing…, Stopping… or Refreshing… The menu item's action is running and the item is greyed out. Wait for the action to settle; the row refreshes with its new state.
Processing A load or sync is running on the datasource. Wait; the row is re-checked every 10 seconds and shows Idle if it makes no progress.
Installing A custom connector is being installed. Wait for the install to finish, then finish any pending setup.
Pending A custom connector is waiting for its setup or a confirmation. Open the datasource and complete its setup.
No results for "…" Nothing in the tree matches the text you typed. Check the spelling, or type fewer characters; search matches both the names of items and the labels shown in the tree.
Showing first 50 results The search hit the 50-result cap, so more matches may exist below the cut. Type more characters to narrow the search.
No folders The Folder filter has nothing to offer, because no folder holds a datasource. Move a datasource into a folder, then reopen the filter pane.
No types The Type filter has nothing to offer, because there are no datasources to take source types from. Add a datasource first.
No datasources or No datasources match The scope holds no datasources, or your folder and type choices leave none to list. Widen the choices, or clear them.
No matches The filter pane's search box found no option in this category. Clear the pane's search box, or try another spelling.

Deploy and delete

You see What it means and what to do
Select at least one deployable table Deploy is unavailable because nothing is selected. Select a table that is not deployed yet.
No tables in selection Nothing is selected. Select the tables you want to deploy.
All selected tables are already deployed Every selected table is already in the Lake. Select tables that are not deployed yet.
Selected tables are missing primary key The selected tables have no primary key column. Define one for each table, then deploy.
No deployable tables in selection None of the selected tables can be deployed. Check each table for a primary key and its deployment status.
Deploy Selected Items?, with Tables skipped (…):, … already deployed or … missing primary key The confirmation for the toolbar Deploy action; the dialog counts what will deploy and why tables are skipped. Check the counts, then confirm.
Deploy …: … or This will deploy the … "…" to the Lake The right-click deploy confirmation on a tree node. Confirm with Deploy.
Table is already deployed The table is in the Lake already. Drop it first if you want to deploy it again. See Deploy tables.
Primary key required. Define a primary key column before this table can be deployed. Select Go to Columns, then mark a column as the primary key.
Sync and migrated must be enabled for primary key or partition columns. Keep the sync and migrated settings on for key and partition columns.
Type change refused, with … values of "…" cannot be converted to … (e.g. '…'); truncate the table first and reload after the change The new type would turn stored values into NULL. Truncate the table, change the type, then reload it.
The table with name … already exists or A table with this name already exists in the selected schema A table with this name is already in the Lake. Choose another name. See Import tables.
Table name cannot contain double quotes Remove the double quotes from the name. See Import tables.
All tables deployed successfully or All tables removed successfully No action needed.
Deployment partially succeeded or Deployment failed for all tables Open Failed Deployments next to Deploy. Click Edit to fix a table, or Redeploy to try it again.
Failed Deployments, with … table(s) failed to deploy or … table(s) failed - click to view The failed-items popover behind the badge next to Deploy. Open it, read the error on each row, then Redeploy or Edit. Retries are per table.
Failed to deploy table or Failed to deploy items The deploy failed and the server gave no reason. Check the table, then retry from Failed Deployments.
Removal partially succeeded or Removal failed for all tables Open Failed Deletions next to Delete, then click Retry delete for each table.
Failed Deletions, with … table(s) failed to delete or … table(s) failed - click to view The failed-items popover behind the badge next to Delete. Open it, read each error, then Retry delete or Edit.
Delete Selected Items?, with Items to delete (…): counting … data source(s), … database(s), … schema(s) and … table(s) The confirmation the toolbar Delete opens for the current selection. Check the counts and the dependants list, then confirm; the dialog warns that the action cannot be undone.
This action will permanently remove this … . This operation cannot be undone. The right-click delete confirmation; the placeholder names the item's kind, for example a data source. Confirm with Delete.
Delete anyway The check found dependants, so the confirm button warns; the check itself does not change what gets removed. Read the listed dependants before you confirm.
1 Mind you cannot access, or a count such as 2 presentations you cannot access A dependant sits outside your access, so only the count shows and the row is not clickable. Ask its owner before you change or remove the table.
Cannot remove table "…": ANY API connector "…" reads it through grant "…". Revoke the grant first. or Cannot remove column "…": ANY API connector "…" reads it through grant "…". Revoke the grant first. A data access grant reads this object, so the removal is refused; a broken grant still counts until it is revoked. Revoke the grant on the connector first.
Cannot remove permission group(s) from table "…": it is used in the following Minds: […]. Remove it from those Minds first The permission group is used by Minds linked to this table, so it cannot be removed from the table yet. Remove the table or that group from those Minds, then try again.
Checking what depends on this... The dependency check is running. Wait for it to finish.
Could not load every dependant, so the list may be incomplete. The check partly failed, so some dependent objects may be missing from the list.
Dependants could not be determined or DataPrism dependants could not be determined because the check failed. Acting on the assumption that this table is unused may be unsafe. Part of the check failed, so the list can be missing objects and must not be read as zero. Treat the list as incomplete and check again later.
Failed to load dependants The tab could not load the list of dependants; the reason appears below the heading. Click Retry; if it keeps failing, check again later.
No dependants found, This table is not currently used by any downstream objects. or This static table is not currently used by any downstream objects. The check ran and found nothing that uses the table. No action needed.
Static Not Created, with Please save the static table first to view its dependants The static table is not saved yet. Save it first; the tab lists dependants once it exists.
… are not visible to you. Objects you cannot access still depend on the table. Ask their owners before you delete.
Drop this table from ClickHouse? The drop confirmation; dropping deletes the data but keeps the definition, and anything that depends on it is not dropped and will break until you fix it or redeploy the table. Confirm only when sure; the definition stays, so you can deploy again.
Delete Table, with Are you sure you want to delete this table? This action cannot be undone. The table delete confirmation; Delete anyway appears when something depends on the table. Confirm only when sure; a refusal keeps the dialog open.

Warning
Deleting a table can break dependent objects, including ones you cannot see. Check the list in the delete dialog, and ask the owners of hidden objects, before you confirm.

Loads and syncs

You see What it means and what to do
Unable to synchronize, the table is not deployed The table has not been deployed. Deploy it first.
A message that the table must be added to a workflow before it can be synchronised The table has no workflow. Set Workflow Name on Table > General, then click Synchronize or run its workflow.
The workflow containing this table is currently running. Please wait for the workflow to complete before … The workflow is running. Wait for the run to finish, then try again.
Synchronization started A load or a schema re-read has begun for this datasource or table. Watch the Sync Log and the datasource badge; the message confirms the start, not the finish.
Synchronization blocked by a scheduled run A scheduled run would collide with your sync. Choose Disable Current Table Automatic Synchronization to stop automatic syncs for this table only, or Disable Workflow Synchronization to turn off the schedule for every table in the workflow. Turn each one back on later: the table option on the General tab, the workflow option in Workflows. You can also wait.
Wait for the running synchronization to finish. A table action is disabled because a run is in progress. Wait; the buttons enable when the run ends, or turn off Synchronize on Table > General for the table.
The workflow … is running., A test run of the workflow … is running. or A synchronization in the workflow … is running. A workflow run holds the table, so its actions are disabled. Wait for the workflow run to finish.
Cannot truncate while workflow is scheduled A scheduled workflow includes the table. Turn off Synchronize on Table > General, or disable the workflow schedule from the Workflows panel.
This table is included in scheduled workflow …. Turn off Synchronize on the Table > General tab, then try again. Alternatively, disable the workflow schedule from the Workflows panel. or Workflow … is still on schedule. Disable the workflow schedule from the Workflows panel, then try again. The truncate is blocked by the schedule. Pick either way out, then retry the truncate.
This table has no partitions yet - there is nothing to truncate. The table has no partitions, so there is nothing to do.
You are about to truncate … partitions (… rows) from this table. The confirmation summary of a partition truncate; the action cannot be undone. Confirm only when the count is right.
This table is not partitioned, so truncating clears all of its data. This action cannot be undone. The whole-table truncate confirmation. Confirm to empty the table entirely.
The following partitions do not exist on table "…": … . Refresh the partition list and try again The partition list changed since it loaded. Click Refresh in the dialog and repeat.
Cannot truncate partitions of a lake table with sync type "…": it has no PARTITION BY. Truncate it without partitions to empty it entirely The table's sync type produces no partitions. Truncate the whole table instead.
The table has no partition columns, so it has no partitions to target or Nothing to truncate for the requested target There are no partitions to choose. Choose partitions, or truncate the whole table.
The table has been truncated or The table has been dropped from the database Done. Synchronize or redeploy the table if you need its data again.
Failed to synchronize. Please try again. The sync request failed. Try again. If it fails again, read the Sync Log.
No synchronization logs found The table has no runs yet. Run Synchronize to create the first one.
Sync stopped The datasource sync was stopped; the data is restored to its pre-sync state and you can sync again.
Nothing to synchronize Sync Schema found nothing to add; the Lake already mirrors the source. No action needed.
Refusing to reload day(s) … Re-deploy the table so it is partitioned by its incremental column. The days to reload do not line up with the table's partitions. Redeploy the table so it is partitioned by its incremental column.
Files Loaded, Rows Rescued or Files Skipped (no date) The file-run steps after a file table loads: files read, rows rescued, and files left out with their reason. Review skipped files; no date means the file could not be dated for an incremental load.
Abort Synchronization, with Are you sure you want to abort this running synchronization? The confirm dialog for stopping one run from the Sync Log. Confirm; the run is marked Aborted with the note "Aborted by user".
Delete Log Entry, with Are you sure you want to delete this log entry? This action cannot be undone. The confirm dialog for removing one Sync Log row. Confirm to remove the row.
Load failed or Load finished with warnings The runner panel state above the raw error of a Connect Any API run. Read the error stack, fix the cause and reload.
Data source "…" is not currently syncing No sync is running, so there is nothing to stop.
Data source "…" is still installing The connector is still installing. Wait for the install to finish, then sync.
Data source "…" has code changes that are not installed or Not synchronized: the connector has code changes that are not installed. Reinstall it first. Select Reinstall, then sync.
Data source "…" has unconfirmed changes Confirm the connector with Save and sync, then sync.
Data source "…" is already … - wait for it to finish before retrying The datasource is busy with another operation. Wait for it to finish, then retry.
Not available for this datasource. The source type does not offer the selected sync type. Choose a type the source offers; Amazon S3 sources do not offer Incremental Reload Loop (Truncate Insert).
Full Reload Start Date is required for the Loop sync type when Full Reload When Empty is on or the Full Reload Start Date field marked Required for Loop The loop type saves only with a start date while the empty-table full reload switch is on. Set the date, or turn Full Reload When Empty off.
Sync type "Replication" is never stored on a lake table: an analytical system accumulates history, it does not replicate. Use "Full Reload (Truncate Insert)", "Incremental Reload (Truncate Insert)", "Incremental Reload Loop (Truncate Insert)", "Incremental (Upsert)" or "Onetime". The Replication type is not offered; choose one of the listed types.
Red Failed badge on a datasource The last sync failed. Click the badge to read the error, fix the cause, then run Sync Schema.

Table settings

You see What it means and what to do
Please fill in all required fields A save on the table's General step was refused. Fill in the required fields, then click Apply again. See Configure a table.
Failed to save general settings The General save failed and the server gave no reason. Check the fields, then try again.
Incremental column is required for Incremental sync types. The save was refused because the incremental type has no date column. Choose a Date Column on the General step, then apply again.
Change Date Column?, with You are about to change the incremental date column. and This will drop and recreate the table. All existing data will be permanently lost. A deployed table is changing its incremental column; the change is destructive. Confirm only when the reload afterwards is acceptable; the table loads fresh data on its next run.
Primary Key Change Detected, with Warning: You are modifying a primary key or partition column's definition (key status, data type, or data domain). This will drop and recreate the table and all existing data will be permanently lost. Do you want to proceed? A save would change a primary key or partition column's definition on a deployed table. Confirm to proceed, or cancel to abort the save and keep the table.
The admin group has full access to all tables by default and cannot be removed. The admin group is always included in Permission Groups. Add other groups as usual.
Query not validated The custom table's query has no passing check yet, so Save stays disabled. Click Run Query until it passes.
You have unsaved changes. How would you like to proceed? You are leaving a step, or finishing, with unsaved edits. Choose Save & Continue, or discard the edits.
Please complete all required fields before navigating or Please complete all required fields before finishing Next or Finish is blocked because the open step fails validation. Fill in the fields the step requires, then move on.
Error loading data: … The popup could not load the table; the reason follows the colon. Close the popup and open it again; if it keeps failing, reload the page.
Table ID is required for saving columns. or Failed to save columns The column changes could not be saved. Reopen the table and apply the changes again.
At least one column is required. The table needs at least one column. Keep a column, then apply.
Column name cannot be empty. A column has no name. Give every column a name.
Duplicate column names: … Two columns share a name. The message lists them; rename one.
This column is locked and cannot be edited. The row comes from the source locked. Leave it as it is; edit the open rows only.
Only custom data domain's data type can be edited. The Data Type cell is locked because the column's data domain is not Custom. Choose Custom in Data Domain to pick the type yourself.
Source metadata from the datasource cannot be edited. The Name and Source Type cells mirror source metadata and are read-only. Refresh the columns from the source with Sync Schema when the source changed.
Source now maps to …; the Lake column is … Sync Schema found the source now maps the column to another Lake type; the Lake keeps its current type. Review the suggestion and change the type on a Custom domain if you want it.
No columns defined. The grid is empty; the table reports no columns. Run Sync Schema on the datasource to pick up the source columns.
No permission groups added to table The table has no permission groups yet, so none are offered per column. Use Add in General to add groups first.
All columns already have comments or Primary key is already defined The AI helper is disabled because every target column already has a description, or a primary key exists. Change the columns by hand if needed.
Validation Error, with Please check the columns configuration. A save was blocked by the pre-save check, so nothing was saved. Fix the listed row or column problem, then save again.
Table ID is required for saving RLS settings. or Failed to save RLS settings The row-level security settings could not be saved. Reopen the table's RLS step and save again.
Permission Groups Required, with You must add at least one permission group before you can configure Row Level Security. Add permission groups in the General tab to proceed. The RLS step cannot be set up without a permission group. Add at least one group on General, then return to RLS.
Invalid query settings The query execution settings were refused. Check the values in the dialog, then save again. See Configure a table.
Restored version v…, with Save Settings to apply it. A history restore loaded the version into the dialog but saved nothing. Click Save Settings to apply the restored version.
Identical to the current settings or Same as the current settings The version matches what the dialog already holds, so restore or compare is off. Pick another version.
Could not load query settings history The version list failed to load. Close the History dialog and open it again.
No settings history yet The table has no saved versions of its settings. Save the settings once; the first version is recorded on save.
No settings saved - the default applies. The selected version holds no stored settings, so the defaults were in effect at that point. Pick another version if you expected settings to be there.
This table's source is not MSSQL, so the option is unavailable. Recompile Query Plan applies only to MSSQL and Azure SQL Database sources. Leave the switch off; the rest of the query settings still apply.
Memory limit exceeded A run that goes over max_memory_usage fails with this error. Reduce the query's memory demand (for example with the Eco preset) or raise the limit.
Too many parts An insert into a partition that reaches the parts limit is refused. Let background merges reduce the part count, then retry.
Failed to generate column comments or Failed to suggest primary key The AI helper did not finish and nothing changed. Try again.
Deploy the table to preview lake data The Lake side of the preview is greyed because the table is not deployed yet. Deploy the table, or read the datasource side in Source until then.
Preview Error or Failed to load preview data The lake preview could not load. Read the reason under the title and click Retry; if it keeps failing, check the rows on this page.
Theres no date type column in the table or No column with a Date/DateTime data type was found on this table. An incremental sync type is unavailable because the table has no date column. Add a date column, or choose another sync type. See Choose a sync type.
The recipe declares no datetime cursor for this stream, so it can only be reloaded in full. The stream offers no datetime cursor, so no incremental type can use it. Load the stream in full. See Choose a sync type.
This column is inside the files, so every run reads every file. Use _file_date for file-level incremental loads. A date column inside the files cannot narrow a run. Use _file_date for file-level incremental loads. See Load files.

Batch edit

You see What it means and what to do
Select a ready datasource to batch edit The Batch edit toolbar button is inactive because the marked item is not a ready datasource. Select a datasource that shows as ready, then open Batch edit.
Failed to load tables The grid could not load the tables; the reason follows the message. Close the popup and open Batch edit again.
No tables found or This data source has no tables to configure. The datasource holds no tables, so there is nothing to batch edit. Import tables from the source first.
Error: No data source selected The popup opened without a datasource. Close it and open Batch edit from a ready datasource.
Partially saved, with … of … table(s) updated. Check the error bar for details. Some tables saved; failed rows stay in the grid with their changes. Open Failed Updates and fix or retry the remaining rows.
Save failed, with No tables were updated. Check the error bar for details. Nothing saved and all changes were kept. Fix the failures and save again.
No changes, with There are no changes to save. The save found nothing to do. Edit a cell first.
Failed Updates, with … table(s) failed to save or … table(s) failed - click to view The failed-items popover behind the toolbar count. Open it, then Retry, Edit or Copy error per row.
Workflow name is required when synchronize is enabled, after a table name in Batch edit The table has Synchronize on but no workflow. Turn Synchronize off for it, or set its Workflow Name on Table > General. See Batch edit and organise.
Incremental column is required, Minus day must be set, Loop increment must be at least 1 or Full reload start date is required for the Loop sync type when Full Reload is enabled A batch edit cannot be saved. Each problem is listed with its table name; fix the listed rows, then click Save. See Batch edit and organise.
Deploy table? The deploy confirmation opened from the grid's Status column. Check the table name, then confirm with Deploy.
This table is deployed and cannot be redeployed from here. The Deployed badge in the grid is locked. Manage the deployed table from its own popup instead.
Row count is read-only. or Catalog location is read-only. The cell is locked; the value comes from the table itself. Read the number, you cannot change it here.
Available only for incremental sync types., Only available for the Incremental Reload Loop sync type or Only available for the Incremental Reload and Loop sync types with Full Reload enabled The cell does not apply to this row's sync type. Switch to the named sync type to use it, or edit a row that uses one.
Change mode in Query Settings The Settings Mode cell is read-only in the grid. Change it from the Query Settings dialog.
No columns available for incremental sync The greyed-out incremental sync type needs a date or datetime column the table has none of. Pick a type that is not greyed out; to use incremental types, import a date column from the source first.
Change DDL Type from … to …?, with This will drop and recreate the table. All existing data will be permanently lost. The sync type change moves a deployed table to another DDL family, so the table must be dropped and rebuilt. Continue only when losing the data is acceptable.

Workflows

You see What it means and what to do
Workflow name is required Enter a name for the workflow.
No workflow container found for this datasource The datasource has no place for workflows. Ask your administrator to check it.
No datasources ready, with Deploy tables from the lake menu first. Only datasources with at least one table appear here. The view found no datasource with at least one table. Import and deploy a table into the datasource first.
No workflows yet, with Create a workflow with the New button or No workflows found for …. Create one with the New button. The selected datasource has no Lake workflows to list. Create one with the panel's Add Workflow button.
Can't run yet, with None of this workflow's tables are deployed, so it has nothing to load into. Deploy at least one table, then run it. Run now is blocked because none of the workflow's tables is deployed. Deploy at least one of its tables, or use Show datasource to open the datasource.
Cannot add workflow, with No workflow container exists for the selected datasource. The datasource has no workflow container yet, so the registration form cannot open. Check the datasource's state, then try again.
No deployed tables in this workflow. The selected workflow has no deployed tables, so the source list is empty. Deploy a table of the workflow first.
No other workflows with the same type on this datasource. No destination workflow of the same type is offered, including because none is editable by you. Create another workflow of the same type, or check your modify permissions.
No tables match your search. The transfer search matched nothing in the list. Clear or change the search.
No PK The table has no primary key, so it cannot be transferred. Add a primary key on the table's Columns step.
Transfer failed The move request failed; nothing moved. Read the reason and try again; the ticked tables stay selected.
You do not have permission to modify any destination workflow here. The transfer button is disabled because no destination workflow is modifiable by you. Ask for Modify Lake Workflows on a destination workflow.
Delete Metadata, with Are you sure you want to delete this metadata? The workflow delete confirmation, opened from the card's Delete. Confirm to delete; if it refuses because tables remain, move them to another workflow first.
Add Workflow, with Select a datasource first The panel's create button needs a selected datasource and the Add Lake Workflows permission. Select a datasource first; if the button is missing, ask for the permission.
Tables, Workflows, Deployed, Pending Deployment, Missing PK or Workflows with no tables The stat chips on a datasource row; Missing PK counts tables without a primary key. Point at a chip to read its name.
No transformations match. The filter hides every group, or there is nothing to list. Change or clear the filter in Filter tables, source systems, transformations…; if it stays empty without a filter, there are no transformations to move.
Failed to move transformations The move failed and the server sent no explanation of its own. Try again; if it repeats, check that you hold Add/Modify Warehouse Objects.
Unassigned The transformation's chip shows no Warehouse workflow, so it is not in one yet. Tick its group, choose a target workflow and move the group to place it in one.
Finalized The transformation is finalized; finalized transformations of one table and source system must move together. Keep the group whole when you move it.
Cannot move the transformations into "…": "…" is still filled by "…" in …. All transformations writing the same table for the same source system must belong to the same workflow The move would leave finalized transformations of one table and source system behind, which the load rules forbid. Tick every finalized transformation of that table and source system and move them together.
There is a job running on metadata: …. Please wait until it finishes. A run of the source or the target workflow is still going, so the workflows cannot take the move. Wait for the run to finish and move again.
No nodes found matching "…" Nothing the view currently shows matches the text. Clear the box, show a hidden level, or move the timeline; hidden levels and timeline-hidden steps are never searched.
Unknown A group cannot take a name because the run's data gives it no semantic source, table or source system. Open a step's details to see what it belongs to.
Paused A group whose work is done at the selected moment reads Paused while that moment is earlier than the run's end. Move the timeline to the end, or click the reset button (Reset to latest).
No Workflow Data, with Workflows will be displayed after at least one run. The workflow has no runs yet, so the diagram window and its Summary view have nothing to show. Run the workflow once. See Run a workflow by hand.

Connectors and webhooks

You see What it means and what to do
Recipe validation failed: … The recipe has issues. Fix the ones listed, then select Save and sync.
Fix the recipe issues listed below, then continue. Fix the listed issues before you go to the next step.
Save the connection first, then fill the Connection settings or Save the connection first, then test it The connector's connection does not exist on the server yet. Save the connector first, then enter the settings or run Test connection.
The connector is created - Name and Code are locked. Name and Code are fixed once the connector exists; only Folder and later steps stay editable. Continue the wizard; create a new connector if the name or code must change.
Build the recipe first: describe your API to the assistant, paste a manifest or import a zip., Build the recipe first - there is nothing to sync yet. or Nothing to sync - the streams match the saved recipe. The recipe is empty, or the connector already matches the saved recipe. Build or change the recipe, then use Save and sync.
The saved recipe has not been validated since it last changed. Save and sync (or Sync Schema in the tree) validates it and discovers streams., Data source "…" has unconfirmed recipe changes - run Save and sync before this action or Native API recipe of data source "…" has never been validated - run Save and sync first The recipe changed after its last validation, or was never validated. Run Save and sync (or Sync Schema) to validate it, then repeat the action.
Cannot revert past the initial recipe version or No changes since the last checkpoint The oldest recorded recipe version cannot be reverted past, and a revert with nothing recorded after the checkpoint has nothing to do. Save a new version after editing.
Secrets are not copied when duplicating. Enter them again. or Secrets are never inside a recipe zip - re-enter them in the form after importing. A duplicated or imported connector never carries stored secrets. Enter the secrets in the connection form again, then Save and sync.
Secret values are never accepted in chat or tool arguments. The assistant refuses secret values sent through chat; it can only report which secret keys are set or missing. Enter the value in the connector form's Secrets section, then continue.
Approve the pending hosts below - the recipe can only call approved hosts. or Approve the recipe hosts A host the recipe calls is not approved, so the step or the sync stays blocked. Approve or revoke every pending host under Hosts, then continue.
You need execute permission on this connector to test it. Ask your administrator for Execute/Abort Lake Objects, or save without testing.
The connector config references secret(s) that are not set: …. Set them in the Secrets panel, then sync again. The sync cannot start until every referenced secret has a stored value. Set them in the Secrets panel, then sync. See Connector secrets.
Waiting for user to store secret(s) in the Secrets panel. Requested: … The connector agent is waiting for you to store the requested credential values, and the turn is suspended. Store the values in the connector's Secrets panel, then send the next message.
"…" is the placeholder the UI shows instead of a stored secret, not a real value. A placeholder or redacted marker was treated as a credential value. Only store a value the user actually provided; markers and placeholders are not values.
Clear this secret? Clearing a stored vault secret cannot be undone. Confirm only when the value is no longer needed; while the configuration still references the name, Clear stays disabled.
The key is not valid JSON, This is not a service account key ("type" must be "service_account") or The key is missing … The service account key is not valid JSON, is not a service-account key, or misses named fields. Export the key again, or fix the named fields.
Vault is empty or Nothing is stored in this connector's Secrets vault yet. No secrets are stored for this connector yet. Add the values the configuration references; nothing else is affected.
Only .zip files are accepted. Upload a .zip file.
The zip must be 20MB or smaller (this one is …MB). Upload a smaller zip.
The file must be 48 KB or smaller. Upload a smaller file for this setting.
The connector install failed - retry it on the Build step before syncing. Retry the install on the Build step, then sync.
Add the configuration before continuing. In the create flow the configuration is empty, so Next stops. Write the configuration, then continue.
Wait for the install to finish, Fix the install error and retry or Reinstall the connector to continue Next on the Build step is disabled; the hint says which build state blocks it. Wait for the install, retry a failed install, or reinstall after code changes.
The connector is still installing - syncing waits for it. The connector is not ready, so syncing waits for the install. Click Go to Build and finish the install.
Install failed The install did not finish; the error stays on the step. Read the step detail, fix the cause and click Retry.
Reinstall needed, Code changed since the last install or Code changed since the last install. Reinstall to apply it - new dependencies from pyproject.toml are installed too. The connector's code changed since the last install. Click Reinstall on the Build step or the banner; new dependencies are installed with the code.
Live progress is unavailable - checking the status every few seconds. The live install progress stream was lost, so the card falls back to polling. The status keeps updating; if the install stops being tracked, retry it from the card.
Delete this connector?, with This permanently deletes the connector you just created and its code. This cannot be undone. The confirmation for deleting a connector from the install card, offered only in the create flow after creation. Confirm to delete it, or cancel to keep it.
Last sync failed The connector's last data load failed; the error card sits on the Connection step. Read the error card, fix the cause, and sync again.
AI Configuration Suggestion Fill with AI produced a configuration, shown as an unsaved suggestion. Review and Accept or Reject; masked inline secret values must be re-entered or moved to the vault before testing.
Malformed secret reference at: …. Use the exact form {{secret:NAME}}. The YAML holds a {{…}} value that is not a whole-value secret reference (a typo, bad characters, or text around it). Replace it with {{secret:NAME}} standing alone as the whole value, or remove it.
Store a value for … in Secrets, or remove the reference before continuing. Next was clicked while a referenced secret has no stored value. Store the named values in Manage Secrets, or remove the references.
… references secret …, which is not stored. The configuration references a vault name with no stored value. Store a value under that name, or fix or remove the reference.
… holds an inline value. Move it to the Secrets panel and reference it as {{secret:…}}? The saved configuration holds a plaintext value where a secret belongs; the suggestion never blocks. Click Move to Secrets, or dismiss the suggestion.
Fill with AI hid … inline secret value(s). Type the real value or move it to Secrets before testing. Fill with AI masked an inline secret value, so the test is held back. Type the real value in the editor, or move it to the vault.
Available once the connector is installed and up to date. Test connection is unavailable until the install has finished and no reinstall is pending. Finish or redo the install on the Build step, then retry.
file must contain a 'config:' section at the root level The configuration's shape is wrong, so the check cannot run. Keep a config: section at the root of the YAML and test again.
Failed to save secrets Staged vault values could not be saved, so the save stopped. Fix the secret value or name and save again, or use Revert to drop the staged values.
<redacted - re-enter this value> or <inline secret - move to the vault> A sensitive value was held back from the file. Enter the real value, or store it in Secrets and reference it, before saving or testing.
Save the connection first to browse and edit tap files. The connector is not ready for file editing yet. Create the connector first (click Create & continue on General in the create flow), then open the editor again.
This connector is not available. The step or editor could not load the connector. Close it and reopen it from the connector's Build or Recipe step; if it persists, check the connector's state in the Lake tree.
Failed to save file Saving the open file failed. Click Save again; copy important changes out of the editor before closing it.
Failed to paste, Failed to move …., Failed to read …. or Failed to create …. A file-tree action failed. Retry it; if it keeps failing, refresh the tree and check the connector's install state.
Could not protect the file or Please refresh the tree and try again. Protecting the file failed; a stale tree causes it, and without Lake modify access the request is refused. Refresh the tree and try Protect again; check your Lake permission.
Merged AI changes The connector agent's edits were merged into your unsaved changes. Review the file and click Save.
Your edits conflict with the AI's changes The agent changed lines you also edited, so the editor cannot merge them silently. Use View diff, then choose Keep mine or Take AI's.
File deleted while you were away A file you had unsaved edits for no longer exists. Choose Restore as new file to recreate it from your draft, or Discard draft to keep it deleted.
Could not restore … or Could not restore unsaved edits The editor could not re-apply a draft or your unsaved edits. Reopen the files and check their content before saving.
No versions yet or No checkpoints found for this tap The version history is empty. Ask the connector agent to change a file; versions appear in History as the files change.
Checkpoint not found: … The version you asked for does not exist. Reopen History and pick a version from the list.
Cannot revert - this is the earliest checkpoint The first version cannot be reverted; there is nothing before it. Pick a later version, or continue from the current state.
Could not load the version history. The version history failed to load. Close History and open it again.
Couldn't load earlier versions - retry Loading older configuration versions failed; the versions already on screen stay. Click the retry action in the list.
This version has no configuration or Identical to the editor content The selected configuration version holds no YAML, or matches what is already in the editor, so restore and compare are unavailable. Pick another version.
Version loaded into the editor Restore this version put the selected version into the configuration editor; nothing is saved yet. Click Save to apply it.
You have … unsaved file(s) open. Applying the zip discards those edits. The confirm dialog found unsaved files in the editor. Cancel and save the files first if you want to keep those edits.
… is protected - the uploaded copy was ignored and the server copy kept The zip tried to change a protected file; protected files are never overwritten. Edit it in the editor instead; protection is permanent.
This zip is empty. The picked archive holds no files. Recreate the zip from the connector folder and upload it again.
The zip must have pyproject.toml at the connector root (next to meltano.yml). Zip the files inside the folder, not the folder itself. The archive wraps the connector in an extra folder, so the required files are not at the root. Zip the contents of the connector folder, not the folder itself.
Upload failed The zip was refused and nothing was applied. Fix the zip and upload it again.
Download failed Downloading the connector zip failed. Try the download again.
Invalid regex: … A search pattern is not a valid regular expression. Fix the pattern and search again.
Unsaved changes or Unsaved changes - Save (Ctrl+S) to create a version The manifest has edits that are not saved; a save records a version. Press Ctrl+S to save.
Valid · … streams · hosts: … The manifest passed its check; the line lists the stream count and the hosts, with " (unapproved)" on hosts still pending. Approve any unapproved host under Hosts if the connector should call it.
… errors - … The manifest check found issues; the line summarises the first one. Click the line to jump to the first issue, fix it, and validate again.
Schema hints are unavailable right now (the manifest schema could not be loaded). Validation on save still runs on the server. The editor's inline hints are off for now; the server still checks the recipe. Keep editing and save or validate as usual.
The assistant changed manifest.yaml, with Reload to see its version (your unsaved edits are lost) or keep editing and save to overwrite it. The assistant's edits and your unsaved edits overlap, so they were not merged. Click Reload to take the assistant's version, or Keep mine to keep yours; saving is blocked until you choose.
The manifest declares no connection settings yet. Add spec.connection_specification in manifest.yaml or ask the assistant. The recipe has no settings section, so the form is empty. Add the spec in the manifest editor, or ask the assistant to add it.
This setting uses a shape the form cannot render yet. Edit it as JSON or ask the assistant. A setting's shape has no form control. Edit that value as JSON, or ask the assistant.
Make "…" a plain setting?, with Its stored value is deleted and the version history that mentioned it is scrubbed. Enter the value again afterwards. Declassify removes the stored secret value and its history. Proceed only if you can enter the value again as a plain setting; Keep secret cancels.
Disable certificate verification?, with Credentials for this source will be sent without verifying the server's certificate. Only do this for a trusted internal host. Turning Verify TLS certificate off needs a confirmation. Confirm only for a trusted internal host; otherwise keep verification on.
Import this recipe zip?, with manifest.yaml and the public connection settings in the archive replace the current recipe. The import replaces the manifest and the public settings, keeps the replaced recipe as a version, and leaves the connector unconfirmed until you Save and sync; nothing is loaded into the Lake, and unsaved edits in manifest.yaml are discarded. Confirm to apply the archive, or cancel.
Recipe imported - … settings applied or Some parts need your attention before the connector can run. The import was applied; the result dialog lists Secrets to enter in the form and Warnings. Enter any secrets, read the warnings, then Save and sync.
Import failed The import did not complete. Check the zip against the import rules and try again.
Could not save the recipe The save was refused or failed (for example a conflict banner is up). Resolve the conflict or the reported issue, then save again.
Could not export the recipe (…). The export failed. Read the error and try the export again.
Locked while tables are imported from this connection. Delete those tables first to change it. The RabbitMQ connection cannot change while it has imported tables. Delete those tables, then edit the connection.
Webhook authentication failed: … (the sender gets 401) The sender's signature, token, header or timestamp does not match Sender & security on the connection; the reason names the failed check (missing auth header, token mismatch, signature mismatch, timestamp outside tolerance, or connection auth misconfigured). Compare the sender with the Request guide.
Unknown webhook endpoint (404) The sender posts to the wrong address. Copy the Endpoint URL again and update the sender.
Empty or unreadable request body (400) The request has no body, or the body cannot be read. Send the event in the body.

Data access grants

You see What it means and what to do
No data access yet. A grant lets the recipe drive its requests from the rows of one Lake or WH object. PII columns are sent as tokens. The connector has no grants. Click Add data access (admin role) to create one.
Start with a lowercase letter, then lowercase letters, digits or underscores (48 max). The grant name breaks the naming rule. Rename it per the rule.
"…" is already used. or A data access grant named "…" already exists The grant name is already taken on this connector. Choose another name.
The recipe reads the rows as the stream grant_…. How the recipe references the grant; use that stream name in the recipe.
Could not load the objects. The object list failed to load. Retry; if it fails again, reload the connector.
No objects match. Only objects with a PII admin view can be granted. The object search found nothing grantable. Check the Source and the search text; a table that is not deployed yet cannot be granted.
Pick an object., Pick at least one column. or Enter a whole number greater than 0. The grant form is incomplete, or the row cap is not a positive whole number. Fill the missing part and continue.
The object has no date or datetime column. No column can bound the load window. Leave Date column (optional) on None; every run reads the whole object.
Incremental loads read only the rows whose date falls inside the run window. Without one, every run reads the whole object. What the optional date column changes. Set one when the object has a date column and loads should be incremental.
A run that would read more rows than this fails before the first API request. The row cap is enforced before any request is sent. Raise the cap or narrow the object.
PII columns are sent as tokens, never as raw values. Granted PII columns leave as tokens. Tokens are what the API receives.
Each time the recipe runs, the values of these columns are sent to every host below. The grant confirmation restates what is sent where. Check Columns and Hosts before clicking Grant access.
Type the object name to confirm Grant access stays disabled until the typed name matches the object exactly. Type the object's name as it appears.
No approved hosts yet. The grant is pinned to the hosts approved later; each new host needs an admin confirmation. No host is approved at confirmation time, so the grant waits for later hosts; every new host needs a fresh admin confirmation. Approve hosts later.
The approved hosts changed while this dialog was open. Review the new host list below, then confirm again. The host list changed mid-confirmation. Review the new list and confirm again.
Revoke data access "…"?, with Loads that still reference the stream grant_… fail from the next run. The grant stays in the audit log. Revoking stops loads that still reference the grant. Revoke only when nothing needs the grant any more.
Data access "…" granted, Data access revoked, Could not grant the data access or Could not revoke the data access The result of a grant action. On failure, retry; grant changes are admin actions.
A connector may hold at most 8 active or broken data access grants The connector reached the grant cap. Revoke a grant that is no longer needed before adding another.
Grant "…" was revoked and cannot be changed A revoked grant is final. Create a new grant if access is needed again.
Column "…" is PII-tagged and cannot be the date column - tokens cannot bound the run window or date column "…" is PII-tagged: tokens cannot bound the run window - pick another date column A PII-tagged column cannot bound an incremental window. Pick another date column; if the grant is already confirmed, confirm it again after the change.
Stream "…" is a data access grant; a grant feeds a parent stream and is never loaded into the Lake. A load was requested for the grant stream itself. Import the parent stream, not the grant.
Grant "…" of connector "…" was revoked or does not exist The stream references a grant that is revoked or gone. Set the access up again or remove the reference, then run again.
Grant "…": … Shown on a Broken grant, with a reason such as a column that no longer exists or changed type, a computed column change, an object definition change, or a definition that needs an admin to re-confirm. Fix what the reason names, then let the next run re-check the grant.
Enter a hostname or IP, optionally with :port. The Add host value is not a valid host. Enter a hostname or IP address, with an optional port.
No hosts yet. The assistant adds the API's host when it writes the recipe; you can pre-approve one here. The hosts list is empty. Add a host yourself, or wait for the assistant's recipe.
Host approved, Host revoked, Failed to approve host or Failed to revoke host The result of an approve or revoke action on a host. On failure, retry; the recipe can call only approved hosts.
Also send granted data to these hosts?, with This connector has data access grants. Approving the hosts pins every active or broken grant to them as well. Approving hosts on a connector with live grants also pins those grants to the new hosts. Review the listed grants and confirm with Approve and send, or cancel.
Grant …: values of … from … will also be sent to … The per-grant line of the hosts confirmation, listing what will be sent where. Confirm only when each listed grant should reach the new hosts.

Diagrams and lineage

You see What it means and what to do
Table already exists in this layout The dragged table is already on this layout, so the diagram highlights the existing card instead of adding a second one. No action needed.
Failed to add table The table could not be added to the layout. Try the drag again; if it keeps failing, reload the page and try once more.
No related tables found or All related tables are already on the diagram The selected table has no relationships, or every linked table is already on the canvas. Create a relationship if one is missing. See Create a relationship.
To be deployed or Missing PK The table is not deployed yet, or has no primary key. Deploy the table, or open it and define a primary key in its Columns tab, then deploy. See Configure a table.
Are you sure you want to reorder the diagram? A reorder that runs without a chosen preset asks for confirmation first. Click Reorder to arrange the tables, or Cancel to keep the current arrangement.
No tables or columns found matching followed by your search text The diagram search found no table or column for what you typed. Check the spelling; search matches only the text you type.
Failed to create reference, Failed to update relationship or Reference remove failed The server refused a relationship change (create, cardinality change or removal). Read the reason under the message; for a create or removal, check that you have Add/Modify Lake Objects, then try again.
Relationships can only be created between fields of the same type The two fields you picked have different data types, so Create stays disabled. Pick two fields of the same type; the field lists show each type in brackets.
Duplicate layout name, with A layout with this name already exists. Please choose a different name. or Duplicate tab name, with A tab with this name already exists. Please choose a different name. Creation or renaming was refused because another layout or tab uses the name (spaces are trimmed and case is ignored). Choose a different name.
Invalid name or Layout name cannot be empty. A layout must have a name. Type a name in New Layout Name; the confirm button stays unavailable while the box is empty.
Failed to create layout, Failed to rename layout, Failed to duplicate layout or Failed to delete layout The layout request failed; the reason is shown under the message. Read the reason and retry; check your Lake permissions if it repeats.
Remove Table from Layout, with Are you sure you want to remove this table from the layout? Confirmation shown when removing a table from a custom layout; the table leaves this layout only and stays in the Lake. Click Remove to take it off the layout, or Cancel to keep it.
Remove Relationship, with Are you sure you want to remove this relationship? The removal confirmation for a selected line; Remove deletes the link, Cancel keeps it. Click Remove only when the link can go.
Are you sure?, with This action will permanently delete this custom diagram layout. This action cannot be undone. The delete confirmation for a custom layout; Delete removes it for good. After a delete the tab before it becomes active, and if none is left the built-in diagram returns.
Last Cached … or Refreshing… The panel header shows the age of the diagram's data copy; it reads Last Cached - when nothing is cached. Click the refresh control to load the latest data.
Built-in The badge on the Lake Diagram tab; it marks the standard view, which cannot be renamed, duplicated or deleted. Create a New Layout for your own arrangement.
Refresh failed or Failed to clear local diagram cache The refresh could not complete, or one refresh step failed while the others still ran. Run Refresh from Database again; if it repeats, check the datasource behind the failing step. The browser copy is retried on the next refresh.
No schema available or Column metadata has not been loaded for this table yet. The details panel cannot list columns yet. Open the item again once its column metadata has loaded; the rest of the panel still works.
Lineage preview unavailable. The details panel has no small lineage preview for this item. Click Open in Focus under Related items instead.

Access

You see What it means and what to do
Access denied. Required permission(s): … You lack the permission named in the message. Ask your administrator for it. Execute/Abort Lake Objects or Execute/Abort WriteBack Objects both allow Synchronize and Stop Sync. See Who can do what in the Lake.

Questions, answered

Where do I find the full error text?

Open the table's Sync Log and click the Error Stack cell, then click Copy. For a datasource with a red Failed badge, click the badge and use Copy there.

What do I do when a deploy says "Deployment partially succeeded"?

Open Failed Deployments next to Deploy. Read each table's error, then click Edit to fix the table or Redeploy to try it again.

How do I retry a failed delete?

Open Failed Deletions next to Delete, then click Retry delete for each table that failed.

Why does the delete dialog say the list of dependants may be incomplete?

The check that finds dependent objects partly failed, so some may be missing from the list. Check them before you delete.

Why does the delete dialog say some dependants are not visible to me?

Objects you cannot access still depend on the table. The dialog shows how many there are, but not what they are. Ask their owners before you delete.

Why does my webhook sender get 401?

The sender's signature, token, header or timestamp does not match the connection's Sender & security. Compare the sender with the Request guide. See Webhook and RabbitMQ.

Why does the source panel say "Failed to load source menu"?

The Lake cannot read the source's structure. Use Test connection in the datasource form to check the connection, then select Retry. See Connect a database.

Why is Truncate blocked?

A scheduled workflow includes the table. Turn off Synchronize on Table > General, or disable the workflow schedule from the Workflows panel. See Workflows in the Lake.

Why is a type change refused?

The new type would turn stored values into NULL. Truncate the table, change the type, then reload the table. See Configure a table.

Why was a file skipped during a fetch?

Not every file is copied. A fetch skips a file whose name cannot be stored, one whose stored name is already taken by another file, or one whose copy was cancelled. The reason is listed per file in Last sync messages. See Stored files.

Why is my stored file refused as "not valid UTF-8"?

The stored copy is not UTF-8 text. Choose the file's Encoding on the Format step and save; the stored files are re-read with the new setting. See Load files.

Why was my archive refused?

An archive must be a .zip of at most 10,000 accepted files, without password protection and with standard compression. Split a bigger archive, or re-create it without a password or a special compression method, then upload it again. See Load files.