Cloudron Forum

@girish thank you

@nebulon Hi! Thank you for your quick reply. I believe I've pinpointed the source of those "user verify" logs, which leads to a specific question.

These users all belong to groups using shared mailboxes. For example, MAILBOX1 is owned by GROUP1, which includes USER1, USER2, and USER3. These users access MAILBOX1 in Outlook but authenticate using their personal Cloudron user passwords.

I monitored the logs as departments shut down their PCs to go home, and those users immediately disappeared from the logs. I also tested by temporarily removing a few users from their groups (while they were still working), and the logs confirmed the pattern. Notably, one user from each group was consistently missing from the logs each time (a different one).

This raises my question: When USER1's Outlook initiates a connection, is the password verified successfully—but due to the group/shared mailbox feature, does Cloudron then check against multiple passwords for that mailbox, logging the "incorrect" ones for the other group members?

Hi. My server logs are full of these messages
"Mar 12 16:02:09 box:user verify: USERNAME provided incorrect password"
Multiple times per second.
Where do i even start to look for the source? there is no further info in the logs.

This instance is used as our mail server and the installed apps are: Surfer, Kutt, Rocket Chat, Minio, Jirafeau, SoGo and Wordpress.
There are 65 users on the server but the names i see in the error log are only roughly 10 of these.

@joseph omg! thank you so much for pointing in the right direction. i did upgrade our servers on Netcup a while ago, i kept the old ones in case i needed any data, turns out, the VMs had an autostart set and were running again.... Thank you! Case solved (i hope)

@humptydumpty hi! i checked the headers and everything checks out! To clarify the "clicking the log-link" situation, i actually manually typed /logs.html?taskId=6115 into my browser. But the link in the mail still seems safe after further checking

@jdaviescoates thank you for your reply! I checked the headers and i all checks out, it does come from my cloudron server and uses the address i assigned to the app (wordpress@mydomain.com)
As i have a wordpress instance running (that is live) on the other server (where the failed-backup mails are coming from), i can confirm that wordpress is indeed sending mails to inform about updated plugins. So it is the app sending the info via cloudron's mail feature. Which makes it a weird situation that an uninstalled app is still sending mails lol

Hi, i have 2 Cloudron Servers and i recently started getting mails from each that do not make sense.

The first one is sending me e-mails to inform that Woocommerce plugin could not be updated. This is coming from a Wordpress instance i just recently deleted. The sender email is the one i used with this instance. How is this possible?
Here's the mail (translated from german to english):

"Hello! Some plugins on your website https://redacted could not be updated.

Please check your website now. It is possible that everything is working fine. If updates are available, you should update them.

The following plugins failed to update. If there was a serious error during the update, the previously installed version has been restored.

• WooCommerce (from version 9.4.3 to 9.4.5): https://wordpress.org/plugins/woocommerce/

To manage plugins on your website, visit the plugins page: https://redacted/wp-admin/plugins.php

If you encounter any errors or need assistance, the WordPress.org support forum with its volunteer helpers is available to you:
https://wordpress.org/support/.

The WordPress Team"

This happens 2-3 times a day.

---

The second server is telling me that the backup failed as the bucket does not exist. I recently changed the S3 Provider and the backups are running fine. The time i receive the failed backup mails, gives me a hint that the mail is related to my old backup settings before the S3 provider change. How is this possible?

Here's the mail i receive:

"Dear redacted Admin,

Cloudron failed to create a backup. Please see the logs at https://redacted /logs.html?taskId=6115 for more information.

---

Backup endpoint is not active: Error listing objects. code: NoSuchBucket message: The specified bucket does not exist HTTP: 404

---

Powered by https://cloudron.io

Don't want such mails? Change your notification preferences at https://redacted cloudron/#/notifications

Sent at: Wed, 04 Mar 2026 11:00:02 GMT"

Clicking on the log-link, Cloudron opens the log viewer with the following message: "Mar 04 16:25:10 Logs unavailable. Maybe the logs were logrotated."

This happens multiple times a day.

@james Ok let's see what Cloudron 9 brings

@james Hi james, in the end i attached a second storage to the vm and configured to backups using ext4 and rsync. This works better. I still get regular cp exited with code 1 signal null errors but currently i can live with that. I set my backup schedule to every 2 hours, if every 2nd backup fails with this error, i still have 6 backups per day.

I'm thinking about the possibility that the length of the encrypted filenames might be too much for Wasabi... I'll dig deeper...

Hey joseph, thx for your reply. No this error is when backing up using Cloudrons Backup Module to Wasabi S3. The Synology makes a complete VM Backup in the background, no errors there. I hadn't activated logging in S3, i will change that and run a new backup.

Hi, i have recently setup a 2nd Cloudron instance on Netcup that i only use as a mail server. Currently around 50 Mailboxes with a total of 180GB of data. I do hourly Backups to our on-prem Synology via Tailscale that work fine and fast. Usually between 1 and 5 minutes. Once per night i still want to make a direct backup from Cloudron to Wasabi S3. I already found out that rsync takes a lot of time, so i switched to tar for a few days, and that worked great, but still i would prefer rsync. There's pros and cons for both settings. So i switched back to rsync and the backups work, but during the upload i see a lot of 500 errors.

Here's an example:

Retrying (1) copy of 6sRz+6WKso8LhqISaUkfxZZhA02YwUGbjJPLNwP5YGg/viNdvgKqX2Z0h2Q8PWusD+r432QXN4DIGBmVbmWqqCQ/9EOUhxrSmnxrOk0C1mGNy73UPTrETurLwHZIMWcTaW4/WLjr6ukIaB+olQVkGmjQamtztBwQ0qCzvHqrSTsGUEJWr5EBiWSvHYqYxTOw-O7i/aBD30AGnxxwdei3sZYSXWzbyX2wX7ZCpUkgkj5czFDc/OL8uir0yWaEDftripDEB6aTiwnfGsOK+DCpt3UstR3LUPK4Lq3Vi6Y5W-I4dG6PXKSZaCtohB-DL-mGfDablPYtq-JDXCc3ke9wPI0otcFI. Error: InternalError: We encountered an internal error. Please retry the operation again later. 500

What is this error, how can i fix it and how come the backup still finishes in a positive state?

Hello and thank you for your feedback.
I put this idea on hold for the time being after learning that Okta has had several security breaches in the past. I am reconsidering which solution I want to use at all. Thanks anyway

Hello there, i am currently struggling with setting up Okta external directory.
After some trial and error, i got Cloudron to accept my settings, then when i hit sync, i see under the progress bar the one user i currently have in okta and the task finishes without error. but the user is not created. I guess i have an issue with uid, username or something the like. kinda lost. could anybody help me out here? thx