Pulled new container without reading UPGRADE readme and now it's down

[jlhjlh]

Problem:

I recently restarted my Stalwart mail container, which was set to pull the latest version. I didn't realize there were breaking changes between point releases. Unfortunately, I didn’t check the upgrade instructions before restarting, and now I’m encountering several issues.

I don't know the exact version I was running previously, but the system was working fine about a month ago.

Current Issues:

1. Web Admin Interface Unreachable: After the restart, Stalwart is blocking requests from my Nginx reverse proxy, preventing me from logging into the web admin interface. Here are the log messages I'm seeing:

Log

2024-10-10T13:44:35Z INFO Blocked IP address (security.ip-blocked) listenerId = "http", localPort = 8080, remoteIp = 172.18.0.3, remotePort = 60162 2024-10-10T13:44:36Z INFO Blocked IP address (security.ip-blocked) listenerId = "http", localPort = 8080, remoteIp = 172.18.0.3, remotePort = 60136 ... 2024-10-10T13:46:49Z INFO Blocked IP address (security.ip-blocked) listenerId = "http", localPort = 8080, remoteIp = 172.18.0.3, remotePort = 40908

2. It looks like Stalwart is blocking HTTP requests from the internal Docker network IP used by Nginx (172.18.0.3), preventing me from accessing the web admin panel.

3. Upgrade Process Stalled: After reading the upgrade instructions post-restart, I realized I needed to perform an upgrade via the web interface, but I can’t log in due to the blocking.

4. Potential Database Migration Issue: Since the new version of Stalwart was pulled, I suspect the container might have run database migrations automatically. I’m now worried I can’t safely revert to an older version because of potential schema changes.

5. Failed Attempts to Revert: I tried reverting to version v0.9.0 from around two months ago (the last stable version for me), but when I attempt to pull any of the v0.9.x images, I get 404 errors.

Additional Information:

• Volumes: My Stalwart container volumes are mapped to local folders for persistent data storage.
• Docker Setup: Both Nginx and Stalwart are running as Docker containers, and the IP being blocked (172.18.0.3) is the internal Docker network IP for Nginx.

What I've Tried So Far:

1. Reverting to Older Versions: I tried pulling older images like v0.9.x, but I keep getting 404 errors.
2. Web Admin Upgrade: I read that I need to complete the upgrade via the web interface, but I can't log in due to the IP blocking.

What I Need Help With:

1. Disabling or Relaxing IP Blocking Temporarily: How can I disable or relax the IP blocking in Stalwart, at least temporarily, so I can access the web admin interface and complete the upgrade? Are there command-line options or configuration file changes I can make to bypass this?

2. Reverting to Older Versions: If database migrations have occurred, what are my options for safely reverting to an older version (if at all)? Are there alternative ways to roll back the changes if I can't pull the older images?

3. 404 Errors on Older Versions: Why am I encountering 404 errors when trying to pull older versions like v0.9.0? Have these images been deprecated or removed?

Any help or suggestions would be greatly appreciated! Thanks in advance.

[jlhjlh]

As an update, I was able to update the web admin using: root@bfc47fb5ef63:/opt/stalwart-mail# curl -k -u admin:asdfasdf http://172.18.0.2:8080/api/update/webadmin

But the same errors persist. I'm not sure what else to try here.

[mdecimus]

You need to find out why the internal IP addresses were banned. Check the logs for details.
The most likely reason is that you haven't enabled the proxy protocol and Stalwart is seeing all requests coming from the internal IP. Then someone on the internet tried to attack or abuse your server and Stalwart ended up blocking the internal IP instead of the real one.

[Cannerd-Staff-Admin]

How do we unblock ips?

[mdecimus]

From the webadmin is the easiest way.