---
title: Fix common issues with Scaleway Instances
description: This page helps you fix common issues you may experience when using Scaleway Instances
tags: instances troubleshooting common issues
dates:
  validation: 2025-07-28
  posted: 2025-01-20
---

This page lists the most common issues you may encounter with your Scaleway Instance and provides useful tips for troubleshooting them.

## I experience HTTP errors on Instances

### Symptoms
- You receive `4xx` or `5xx` errors when trying to reach your Instance via HTTP.
- The website or service running on your Instance is slow or unavailable.
- You cannot perform basic HTTP operations like `GET` or `POST` on your Instance endpoint.

### Possible causes
- Incorrect firewall rules: The Instance or security groups might be blocking HTTP/HTTPS ports (`80`, `443`).
- Configuration errors in the web server: Nginx, Apache, or another web server might be misconfigured.
- Application-level errors: Your application code may be crashing or returning incorrect responses.
- DNS misconfiguration: The domain’s DNS settings might be pointing to the wrong IP or not updated.

### Solution
- Check security group and firewall rules:
   - In the [Scaleway console](https://console.scaleway.com/instance/security-groups), confirm that ports 80 (HTTP) and 443 (HTTPS) are allowed inbound for the security group used by your Instance.
   - If you manage your own firewall (e.g., iptables, ufw), verify that HTTP and HTTPS traffic is permitted.
- Verify the web server configuration:
   - SSH into your Instance and inspect your web server’s config (for example `/etc/nginx/nginx.conf` or `/etc/apache2/apache2.conf`).
   - Look for syntax errors or missing directives (e.g., missing `server_name`).
   - Restart the web server to apply changes (`sudo systemctl restart nginx` or `sudo systemctl restart apache2`).
- Check application logs:
   - Check logs in `/var/log/` or wherever your application outputs logs to identify potential errors.
   - Look for stack traces or misconfiguration messages.
- Validate DNS records:
   - Use `dig <your-domain>` or `nslookup <your-domain>` to confirm the IP address matches your Instance’s public IP.
   - If necessary, update the DNS records and wait for propagation.

### Additional resources
- [Scaleway Documentation: How to use security groups](/instances/how-to/use-security-groups/)
- [Nginx Official Documentation](https://nginx.org/en/docs/)
- [Apache HTTP Server Documentation](https://httpd.apache.org/docs/)

## I can't connect to my Instance

### Symptoms
- Your SSH or RDP connection is timing out.
- You see “Connection refused” or “Host unreachable” errors in your terminal.
- Ping (ICMP) requests are dropped.

### Possible causes
- Security group settings: SSH (port `22`) or RDP (port `3389`) might not be open.
- Incorrect IP address: You might be connecting to the wrong public IP.
- User-side firewall or network: Your local network may block outbound SSH or RDP connections.
- Instance not running: The Instance may be shut down or stuck in an error state.

### Solution
- Check your Instance status:
   - In the [Scaleway console](https://console.scaleway.com/instance/servers), confirm the Instance is in a running state.
   - Restart the Instance if it’s not responding.
- Verify the Instance's public IP address:
   - Confirm that the IP in the Scaleway console matches the IP you are using for the connection.
- Review security groups and local firewalls:
   - Make sure the default security groups allow SSH or RDP inbound traffic.
   - Check your local firewall to ensure outbound SSH (port `22`) or RDP (port `3389`) is allowed.
- Use the Scaleway console to connect to your Instance:
   - Try using Scaleway’s web-based terminal in the console to rule out local network issues.

### Additional resources
- [Scaleway Documentation: How to connect to your Instance](/instances/how-to/connect-to-instance/)
- [Scaleway Documentation: Troubleshooting SSH connection issues to Instances](/instances/troubleshooting/cant-connect-ssh/)


## I can't connect to my Instance using an SSH key

### Symptoms
- You see an "Invalid SSH key” or “Permission denied (publickey)” error.
- SSH prompts you for a password instead of accepting your key.
- The connection works with a password but not with the SSH key.

### Possible causes
- Incorrect permissions on the SSH key: The key file might have unsafe permissions or ownership.
- Wrong username: Using `root` versus a custom user.
- Misconfiguration of the SSH daemon: `sshd_config` might disallow key-based authentication.
- Key not installed: The public key might not have been successfully added to the Instance.
- No access to Scaleway's api-metadata: Api-metadata is responsible for loading the SSH keys.

### Solution
- Check file permissions locally:
   - On your local machine, ensure the private key file has restrictive permissions:
     ```bash
     chmod 600 ~/.ssh/id_rsa
     ```
   - Verify that the user and group ownership are correct.
- Verify the correct username:
   - By default, Scaleway Instances use the `root` user. Check if you have created a new user with sudo privileges and if so, use that username.
- Inspect the Instance’s SSH configuration:
   - If you can connect via another method or console, check `/etc/ssh/sshd_config` for lines such as `PasswordAuthentication`, `PubkeyAuthentication`, and `PermitRootLogin`.
   - Make sure `PubkeyAuthentication yes` is set.
- Re-add your SSH Key in the Scaleway console:
   - Go to the [SSH keys](https://console.scaleway.com/project/ssh-keys) tab in the **Project** section of theScaleway console, remove the old SSH key, then add your new, correct public key.
   - Restart your Instance to re-upload the list of allowed SSH keys.
- Make sure that your Instance has access to the api-metadata by running the following command in the serial console:
     ```bash
     # curl http://169.254.42.42/
     ```
     The following output should display:
     ```json
     {"api": "api-metadata", "description": "Metadata API, just query http://169.254.42.42/conf or http://169.254.42.42/conf?format=json to get info about yourself", "version": "12.23.0"}
     ```
     No output means that access to the api-metadata is blocked. Identify the reason for the block and try running the command above again.

### Additional resources
- [Scaleway Documentation: How to generate an SSH key pair](/organizations-and-projects/how-to/create-ssh-key/)
- [OpenSSH Documentation](https://www.openssh.com/manual.html)

## I can't access my Instance

### Symptoms
- You are unable to connect to your Instance via SSH.
- Your SSH key is missing, or you cannot locate the corresponding private key.

### Possible causes
-  The private SSH key associated with the Instance was lost or deleted.
-  The Instance’s SSH configuration or authorized keys file was altered, removing your access.

### Solution
- Check for another SSH key:
   - Verify if another SSH key pair is associated with your [Scaleway account](https://console.scaleway.com/project/ssh-keys).
   - Login using the alternative SSH key

- Use the Scaleway console to add a new SSH key:
   - Upload another SSH key in the [SSH key section](https://console.scaleway.com/project/ssh-keys) of your Instance's Scaleway project.
   - Reboot the Instance from the console to re-upload the new key to the Instance.

### Additional resources
- [Scaleway Documentation: How to generate an SSH key pair](/organizations-and-projects/how-to/create-ssh-key/)
- [Scaleway Documentation: How to power off your Instance](/instances/how-to/power-off-instance/)

## My Instance is locked

### Symptoms
- Attempts to modify or delete the Instance fail with an “Instance is protected” error.
- You see a message in the Scaleway console that the Instance is protected or locked.

### Possible causes
- Security measure: The Instance might be locked due to non-payment or administrative hold.
- User-applied lock: You (or another administrator) enabled resource locking to prevent accidental deletion.
- Ongoing operation: A pending snapshot or backup can temporarily lock the Instance.

### Solution
- Check billing and support tickets:
   - Ensure your Scaleway account is in good standing with no overdue invoices.
   - If there is a lock from Scaleway Support, resolve any open tickets or compliance issues.
- Disable the Protected Instance feature:
   - In the Scaleway console, go to your Instance settings and turn off the lock if you set it manually.
- Wait for ongoing maintenance operations to complete:
   - If an action is running on your Instance, you must wait until it finishes before you can modify the Instance.

### Additional resources
- [Scaleway Billing FAQ](/billing/faq/)
- [Scaleway Documentation: How to use the Protected Instance feature](/instances/how-to/use-protected-instance/)

## How can I open port 25 for my Instances?

### Symptoms
- Your application cannot send emails (SMTP) from the Instance.
- Port 25 connections are blocked or refused.

### Possible causes
- Default SMTP restriction: Access to port 25 is restricted by default to prevent spam.
- Security group settings: The inbound/outbound rules might block port 25.
- Missing or incorrect mail server configuration: Postfix, Exim, or another MTA might be misconfigured.

### Solution
- Configure your security group:
   - In the Scaleway console, ensure outbound connections on port 25 are allowed.
- Set up proper SPF/DKIM/DMARC:
   - To avoid being flagged as spam, configure DNS records for email authentication.
- Use "Authenticated SMTP" (Port 587 or 465):
   - If port 25 is restricted, consider using a mail relay or an authenticated SMTP service on port 587 or 465.

### Additional resources
- [Scaleway Documentation: How to send emails from your Instance](/instances/how-to/send-emails-from-your-instance/)
- [Scaleway Tutorial: Setting up a secure mail server on Ubuntu 22.04 LTS (Jammy Jellyfish)](/tutorials/setup-postfix-ubuntu-bionic/)
- [Scaleway general terms of service](https://www-uploads.scaleway.com/General_Terms_of_Services_v17072024_45d4879c08.pdf)
- [Postfix Documentation](http://www.postfix.org/documentation.html)
- [SPF/DKIM/DMARC Basics](https://dmarc.org/overview/)

## Routed IP errors

### Symptoms
- Your Instance does not respond when using a routed IP.
- Packets might be dropped or misrouted.
- Services are not accessible via the secondary IP.

### Possible causes
- Incorrect network configuration: The routed IP might not be properly configured on your Instance.
- No static routes set: A missing route or gateway could cause traffic to fail.
- security groups or firewall blocking: The traffic from or to the routed IP might be filtered.

### Solution
- Verify the IP configuration in the Scaleway console:
   - Confirm that the routed IP is associated with the correct Instance.
- Update your network configuration files:
   - On your Instance, add the secondary IP to the appropriate network interface (e.g., `eth0:0` or via netplan).
   - Make sure the netmask and gateway are set according to the instructions.
- Check your routing table:
   - Use `ip route` or `route -n` to confirm the route for the additional IP is properly set.
- Test connectivity:
   - Ping the secondary IP from an external source. If it fails, look at local logs (`dmesg` or `/var/log/syslog`) for network errors.
- Adjust your security group:
   - Ensure inbound and outbound traffic is allowed for both IP addresses.

### Additional resources
- [Scaleway Documentation: How to use flexible IP addresses with Instances](/instances/how-to/use-flexips/)
- [Scaleway Troubleshooting: Fix unreachable Ubuntu Focal Instance when it has two public IP addresses](/instances/troubleshooting/fix-unreachable-focal-with-two-public-ips/)
- [Scaleway Troubleshooting: Fix lost IPv6 connectivity on old Debian Buster images when migrating to routed IP](/instances/troubleshooting/fix-lost-ip-connectivity-on-debian-buster/)
- [Scaleway Troubleshooting: Fix unreachable IPv6 RHEL-based Instances after transitioning to routed IP](/instances/troubleshooting/fix-unreachable-ipv6-rhel-based-instance/)
- [Netplan documentation](https://netplan.readthedocs.io/en/stable/)

## Disk space errors with Instances

### Symptoms
- “No space left on device” errors when installing packages or writing files.
- The system becomes unresponsive, or services crash due to insufficient disk space.
- You cannot create or expand files on the Instance.

### Possible causes
- Full root partition: Logs or application data have consumed all available disk space.
- Improper partition setup: Your primary partition might be too small.
- Uncleaned temporary files: `/tmp`, `/var/tmp`, or application cache directories might be full.

### Solution
- Check your disk usage:
   - Use `df -h` to check available space on all mounted volumes.
   - Use `du -sh /var/log/*` or other directories to find large files.
- Clean up unnecessary files:
   - Remove old logs from `/var/log/` or archived backups you no longer need.
   - Clear temporary files in `/tmp` or application cache directories.
- Resize or add a volume:
   - If you consistently run out of space, consider resizing your volume or attaching an additional volume via the Scaleway console.
- Implement log rotation:
   - Configure `logrotate` or other log management solutions to prevent logs from filling the disk.

### Additional resources
- [Scaleway Documentation: How to attach a Block Storage volume to an Instance](/block-storage/how-to/attach-a-volume/)
- [Scaleway Documentation: How to increase the size of your Block Storage volumes](/block-storage/how-to/increase-block-volume/)
- [Linux Log Rotation Guide](https://linux.die.net/man/8/logrotate)

## Packet loss issues with Instances

### Symptoms
- High latency or intermittent connectivity to your Instance.
- Pings show dropped packets.
- Applications that rely on constant connectivity (VoIP, real-time apps) experience disruptions.

### Possible causes
- Network congestion: On either the Instance side or your local network.
- Incorrect MTU settings: The maximum transmission unit might cause fragmentation issues.
- Firewall or rate limits: Overly strict firewall rules can drop packets.
- Hardware or routing issues: Upstream or local network hardware problems.

### Solution
- Run basic network diagnostics:
   - Use `ping` and `traceroute` from both your local machine and another external server.
   - Check if packet loss is local or global.
- Check the Instance's load:
   - High CPU or network usage on your Instance can cause dropped packets. Monitor system resources with `top` or `htop`.
- Optimize the MTU
   - Try lowering the MTU (Maximum transmission unit) to 1400 or 1450 and see if packet loss improves, especially in VPN or container setups.
- Adjust firewall/security groups:
   - Confirm that your firewall is not inadvertently dropping packets or rate-limiting ICMP.

### Additional resources
- [Scaleway Documentation: How to use security groups](/instances/how-to/use-security-groups/)
- [Scaleway Documentation: Preventing outgoing DDoS](/instances/reference-content/preventing-outgoing-ddos/)
- [Scaleway Tutorial: Configuring Firewalls for Instances](/tutorials/installation-uncomplicated-firewall/)

## I can't delete my Instance

### Symptoms
- Attempts to delete the Instance fail with an “Instance is protected” error.
- You see a message in the Scaleway console that the Instance is protected or locked.

### Possible causes
- User-applied lock: You (or another administrator) enabled resource locking to prevent accidental deletion.

### Solution
- Disable the Protected Instance feature:
   - In the Scaleway console, go to your Instance settings and turn off the lock if you set it manually.

### Additional resources
- [Scaleway Documentation: How to use the Protected Instance feature](/instances/how-to/use-protected-instance/)

## Next steps

- Always consult the official [Scaleway Instances Documentation](/instances/) for detailed, up-to-date information to troubleshoot your Instance.
- If you are still experiencing issues after following these tips, open a [support ticket](https://console.scaleway.com/support) in your Scaleway console, providing detailed descriptions of your issue, logs, screenshots, or error messages. This will help the Support team diagnose and resolve your issue more quickly.