@luckow I run cloudron on a dedicated server with this spec: [image: 1765907216854-screenshot-at-dec-16-21-16-42-resized.png] We run lots of n8n instances. Also gitlab instances. We run a APP and Web development Company. Cloudron is our life saver as it makes keeping our projects easy. Specially deploying and backups. Its fast and a click away. FYI, we are almost close to 600. Thank you for your attention.

@joseph said in Can't see any file in the file manager & SFTP broken: have you tried to restart SFTP in Services -> SFTP already? I hadn't, and I noticed that SFTP service was in recovery mode for some reason (I'm going to try to figure out why, but it's most probably a person issue and not a technical one). Restarting it worked. I should have checked there before asking.

It looks like rsync. @brambuijs if you use the File browser, do you see the rest of your files? The error comes from this line - https://git.cloudron.io/platform/box/-/blob/master/src/backupformat/rsync.js?ref_type=heads#L199 . Can you comment out that line and try a restore again? The file on the server is /home/yellowtent/box/src/backupformat/rsync.js . Just put a "//" at the beginning of the line 199 to comment that error out.

The dialog is also fixed now with https://git.cloudron.io/platform/box/-/commit/612002ec33f54d23ab4e4c38e0e5e0d3853e9db2

It works now

Also note, I am waiting for 9.0.14 to be stable release before I can update and then download a backup. Currently it's not possible to download a backup in 9.0.13 from Cloudflare R2.

Is the question whether this is allowed? If so, yes. As @jdaviescoates said, the Cloudron version needs to match. The OS version can be different. As a precautionary measure take a VPS snapshot of your old 22.04 install before migrating.

@girish Thank you very much for the info. I didn’t manage to reply earlier, but I saved the backup configuration again and the morning backup completed successfully. I’ll also try upgrading to 9.0.15.

Whoa, unexpected. One could also say that it's not a bug, as object stores don't distinguish files from directories, as they're all objects. It's the implementation that does. So it may be more likely that the s3 client/driver/config needs to compensate. ex. Cyberduck works fine.

@dylightful If we can have access to your setup, can you write to me at support@cloudron.io ? Would love to debug and figure out why it's failing . Is this reproducible easily - meaning if you take a backup and restore, it fails?

@nebulon That's great. Apart from that, it seems to be running smoothly; it still takes time to propagate the usage stats, but that doesn't really affect us.

@archos said in Old system backups are not listed: and restoring an older app backup fails with the fsmetadata.json ENOENT error. Does this file exist in the storage?

There is no look-into-it. There is only do-it.

Hello @imc67 Now we can start analysing. Edit the file /home/yellowtent/platformdata/mysql/custom.cnf and add the following lines: [mysqld] general_log = 1 slow_query_log = 1 Restart the MySQL service in the Cloudron Dashboard. The log files are stored at /home/yellowtent/platformdata/mysql/mysql.log and /home/yellowtent/platformdata/mysql/mysql-slow.log. Let it run for a day or more. Then you can download the log files and see what queries run very often causing disk I/O.

@joseph ~~Now I have reseted the router and I still have the same problem. I guess it's my ISPs Cgnat that is the problem. It's strange. It has been working for almost a year with the same ISP and the Cgnat but suddenly it stopped to work. I guess it's the Cgnat that make it impossible to ping the public ip too. My apps with Cloudflare tunnels is working but the tunnel I created to the Cloudron dashboard gets bad gateway and host error from Cloudflare. What am I missing?~~ Now it's working. It was simple when I found it. I had forgot to activate No TLS Verify for the tunnel. Now it's working. Thank you for your help.

Ok, I reached out to Vultr Wednesday about it and they instructed me to contact Ya'll directly. Did a fresh install and it installed 8.0. Glad to see it's updated.

@philkunz said in Backups still not verified or what?: But that points to how important the e2e read back step is. absolutely, that is why we are adding the validation in the first place. Having checksums of what we expect is the first step, next then will be the task (can be quite heavy task, so this won't be automatic probably) to read back and recalculate the hashes.