Increasing concurrency and memory, as suggested here, works:
https://forum.cloudron.io/topic/8219/backup-hangs/14

Found at Menu > Backups > Configure > Advanced.

Tip to @girish / dev team: adding in a note on the app > backups page would probably be very helpful.

Restore failing? Try [increasing memory and concurrency]

70c14a75-3889-4c9e-abef-799231c38e76-image.png

@nebulon it's up again

If you use something like Cloudflare for DNS then you have to use the raw IP of your server to connect via SSH, since Cloudflare will not proxy other ports besides 443. Cloudron also has rate-limits setup via iptables but those are hard to hit while manually trying to connect and after a short time your local IP will be able to connect again, so I doubt this is what you are hitting.

Also I am not sure what you mean by ping returning 0.0.0.0?

yes, if the backup config has a password set, the UI will expect it to be entered before you can restore.

You can use the top right menu to mark this as resolved.

@roofboard said in Box Crashes.:

After running the apt update / upgrade the problem resolved.

Please be very aware that you are not supposed to apt upgrade on your own (outside the OS upgrade procedure). This may break your installation and might be the reason of trouble you run into.

greherh (2).png

@girish I restored from a backup to get it working..

@girish my bad! You can blame user error for this one 🙂

Hetzner StorageBoxes need a different path when sub-accounts are used, which I forgot to note in my docu and went changing things.

Here's to anyone dealing with change management professionally, ITIL or otherwise, you are a must-have no matter how often peeps complain about 'red tape' 🍻

There's a reason why I keep documenting things but no AI or home assistant has been invented yet to make me actually remember to stick to it 🤣

Notes to self: Before changing critical things like backups(!) make sure to follow the documentation. Do not disable critical features on accounts like CIFS access (d'oh). Also, make sure to test changes thoroughly before committing to them. Also also, enjoy Cloudron and the awesome people making it a super nice place to be 💋

Hm was there some namecheap API change recently?

@humptydumpty said in Issue with 7.3.4:

In Gandi's case, they're not the fastest either according to DNSPerf.

I'd never even seen nor heard of DNSPerf before, but I see they are number 12 on the list which aint too bad 🙂

Good enough for me anyway. I've always found their LiveDNS to be easily fast enough. More on their DNS stuff here: https://www.gandi.net/en/domain/dns

@nebulon It worked when I set both to maximum, many thanks!

@nebulon Thank you, there was an error with Vultr, that's why it didn't work, but their support fixed it.

@fbartels thanks for that. I was getting caught up in a missionary fervor!

@nebulon So throwing off the unused linux images and restarting unbound helped. Consider it solved. Thank you.

@timconsidine said in Auto-update 7.3.4 - all apps down:

Phew, that was a scary platform update

Yes, 7.3 is pretty massive since it updates docker. So, it has to bring down everything and bring it up again.

This will eventually catch up. By design we do not actively track the servers of our users, so that info only comes in as the Cloudron checks for updates.

Thanks a lot @girish !
We're running mattermost under the indicated domain. I have disabled the outbound, since we have an external SMTP server running - it works now!

I didn't know that Cloudron overwrites configs, but well: Lessons learned 🙂

thanks for the report, we will fix that up then for next release.