# Common Issues

## Connection issues

<details>

<summary>404 Not Found when accessing the CSV URL</summary>

This is the most common issue with the CSV destination. The URL exists but returns a 404 error because the data flow has never completed a successful run.

**Fix:** Go back to your data flow in Coupler.io and click **Save and Run**. Wait for the run to complete successfully. Once at least one run has finished, the URL will return data.

{% hint style="danger" %}
Do not share or use the CSV URL in any downstream tool until the first successful run is complete. The 404 will persist until data has been written to the endpoint.
{% endhint %}

</details>

<details>

<summary>URL stops working or returns 404 after previously working</summary>

If the URL was working and then stopped, the access token may have been regenerated. This happens when someone clicks the **Update** button in the destination settings, which creates a new URL and invalidates the old one.

**Fix:** Open the data flow in Coupler.io, go to the destination settings, and copy the current URL. Update any tools or scripts that were using the old link.

</details>

## Data issues

<details>

<summary>Special characters appear garbled when opening in Excel</summary>

If accented characters, currency symbols, or non-Latin text appear as garbled characters when you open the CSV file in Microsoft Excel, the Excel-friendly mode (BOM) may be turned off.

**Fix:** Open your data flow's destination settings and make sure **Excel-friendly mode** is enabled. This adds a byte order mark (BOM) to the file, which tells Excel to interpret the file as UTF-8. Re-run the data flow after enabling it.

{% hint style="info" %}
This setting is enabled by default. If you previously disabled it and are opening the file in Excel, turn it back on.
{% endhint %}

</details>

<details>

<summary>Column headers are missing from the CSV file</summary>

If your CSV file starts directly with data rows and has no header row, the **Include headers** setting may be turned off.

**Fix:** Open your data flow's destination settings and make sure **Include headers** is enabled. Re-run the data flow. The first row of the CSV file will contain column names.

</details>

<details>

<summary>Fields containing commas or line breaks break the CSV structure</summary>

If field values contain commas, double quotes, or line breaks, they should be properly escaped in the CSV output. If a downstream tool is not parsing the file correctly, the issue is likely in how that tool handles CSV quoting.

**Fix:** Verify that your consuming tool supports RFC 4180-compliant CSV parsing. Most modern tools (Excel, Google Sheets, Python's csv module, pandas) handle quoted fields correctly. If you are using a custom parser, make sure it respects double-quote escaping.

</details>

## Permission errors

<details>

<summary>Access token is invalid or expired</summary>

The CSV URL includes an access token as a query parameter. If this token has been invalidated (by clicking **Update** in destination settings), anyone using the old URL will receive an authentication error.

**Fix:** Retrieve the new URL from your data flow's destination settings and distribute it to any users or tools that need access. Treat the access token like a password — avoid sharing it publicly.

{% hint style="warning" %}
The access token in the CSV URL grants read access to your exported data. Do not post the full URL in public forums or issue trackers.
{% endhint %}

</details>