Posts by Soma

Quote from squitx

The question is how to update or recreate my OMV6 config?

You only had to run omv-release-upgrade and the systems would be updated from Debian 10 to Debian 11 and OMV5 to OMV6.

If you still have the previous system on the SDcard, do it that way.

If you destroyed the previous OS and flashed a new IMG to the SDcard, then your SOL.

Quote from squitx

If I recreate the filesystem via web, I suppose it will destroy the disk filesystem to create a new one. Is this correct?

No. Once you go to the Storage--> Filesystems page, click on the arrow and select MOUNT (do NOT select create).

Your drives will be mounted and the config.xml populated with the drives.

Quote from squitx

I'm trying to "import" the omv5 config.xml part of the node conf.system.filesystem.mountpoint to omv6 with the command 'omv-confdbadm update', but I get a dictionary error. I suppose it needs a kind of migration before updating. I've seen the postinst procedure of the openmediavault package does that. Could I use safely this script to update the config file (or parts of it) to from omv5 to omv6 format?

Not possible.

See the size of syslog and kern.log

They are big.

You can get some free space by deleting the .1 versions but you need to figure out what is spamming those files.

Quote from Fleggy

I have done a sync all but it didn't change the log file size

I probably gave you a wrong apt command.

Writing from memory and on phone is not easy.

Well, at least, you have no error this time on the omv-upgrade.

Your /var/log is showing full and a 7.8Gb size.

You need to dig down and see which log is filling it.

Quote from Andifront

Is there a way to use the Nextcloud installation from this guide with no external IPv4 adress (DS-Lite)?

What is DS-Lite?

You can use the guide without the SWAG part and changes and access Nextcloud only on LAN via http://LAN_IP:450

Quote from bgugl

If you can point me to the technical explanation as to why it should not be used

Long story short: NTFS, exFAT or VFAT (Windows FS) will be owned by root:root on Linux. These permissions can never be changed.

They are ok for reading, not for writing.

If you try to use the folders on containers, for eg., the containers need to be ran as root and this is a major security flaw on Linux.

Hence, not recommended.

Just try to take ownership to a regular USER on a file or folder on them (NTFS drive) and it won't change.

Quote from chente

As far as I know what it does is only copy the RAM to the disk.

The way I understand it, sync will release that space on RAM leaving it free again to other data.

Or not,

Quote from jay137512

You mean testing repo?

When i check it and apply, 500-internal server error occurs

Like it's been said, you have a DNS issue for a long time and your OS hasn't been updated since ever.

How is your network confgured? And how is the OMV plugged? Ethernet wire or Wifi?

Just go on CLI, run sudo omv-firstaid and configure the network to only use DHCP IPv4.

After, post each of the outputs inside CODE boxes of:

sudo omv-aptclean repos

ping -c 5 8.8.8.8

ping -c 5 google.com

sudo omv-upgrade

Quote from Fleggy

dpkg: cannot write to log file '/var/log/dpkg.log': No space left on device

This usually means your folder2ram is full, NOT the OS drive.

Search - openmediavault

How much memory do you have on your system?

Specs of the system?

Are you using the flash-memory plugin?

If yes, go to the Services-> Flashmemory and click sync all

Then try on CLI and post the output of:

sudo apt --list-upgradable

sudo omv-upgrade

Quote from khewweifeng

Still have 500 internal server error when I try to compose docker

Please, stick to post errors here.

You need to set your time correct first.

Make sure that the time on the GUI is correct.

As for SMB, see the wiki guides because everything is explained there.

I don't use SMB so, can't help you there.

Quote from khewweifeng

Will sudden power outage will cause this issue?

What do you think? No electronics component like to be cut out of power without a proper shutdown.

A UPS save a lot of problems.

Onward with the issue:

You didn't post the output of lsblk

Nonetheless, your system is borked. The picture you posted of your FS doesn't match the outputs you posted.

According to the showkey, you only have 2 drives referenced + the rootFS (which is the SDcard) but on the picture, you have the 32Gb drive as VFAT which should be a /dev/mmcblk and NOT a /dev/sdd.

Also, your fstab shows 3x drives that also don't match the showkey output.

Fixing this requires too much back and forth posts which I don't have time for it.

Maybe someone else will be willing to.

Honestly, you'll be better off starting with a fresh install with the drives unplugged.

After OMV is installed and running, power down and reconnect the drives.

All it's needed is to go to the File Systems page and mount them (NOT create)

Quote from khewweifeng

Error message too long cannot post in here,

----------
          ID: quota_check_no_quotas_96cf22db-14dd-40ae-b917-60e1110dc32e
    Function: cmd.run
        Name: quotacheck --user --group --create-files --no-remount --verbose /dev/disk/by-uuid/96cf22db-14dd-40ae-b917-60e1110dc32e
      Result: False
     Comment: Command "quotacheck --user --group --create-files --no-remount --verbose /dev/disk/by-uuid/96cf22db-14dd-40ae-b917-60e1110dc32e" run
     Started: 07:22:41.427532
    Duration: 50.058 ms
     Changes:
              ----------
              pid:
                  13734
              retcode:
                  1
              stderr:
                  quotacheck: Your kernel probably supports ext4 quota feature but you are using external quota files. Please switch your filesystem to use ext4 quota feature as external quota files on ext4 are deprecated.
                  quotacheck: Scanning /dev/sdb1 [/srv/dev-disk-by-uuid-96cf22db-14dd-40ae-b917-60e1110dc32e] quotacheck: Checked 16 directories and 43 files
                  quotacheck: Cannot create new quotafile /srv/dev-disk-by-uuid-96cf22db-14dd-40ae-b917-60e1110dc32e/aquota.user.new: Read-only file system
                  quotacheck: Cannot initialize IO on new quotafile: Read-only file system
                  quotacheck: Cannot create new quotafile /srv/dev-disk-by-uuid-96cf22db-14dd-40ae-b917-60e1110dc32e/aquota.group.new: Read-only file system
                  quotacheck: Cannot initialize IO on new quotafile: Read-only file system

Your drive is Read-Only.

Post each of the outputs inside CODE boxes (symbol </> on the BLUE banner)

cat /etc/fstab

sudo omv-showkey fstab

blkid

lsblk

sudo fdisk -l | grep "Disk "

Quote from xiazi64

everything was working and config.xml was not in a mess, which was very strange

Out of curiosity, the error you have, has some hits on duckduckgo:

Fatal error 9: Input is not proper UTF-8 at DuckDuckGo

Do you have any fancy names on the shared folders? Awkward character's on the names?

Just wondering.

The config.xml has a UTF-8 encoding on the header and the error is stating a bad character NON-UTF8

pi@raspberrypi:~ $ sudo cat /etc/openmediavault/config.xml | more

<?xml version="1.0" encoding="UTF-8"?>
<config>
 <!-- Configuration version -->
 <version>0.1</version>
 <!-- System configuration - Do not add any configuration here -->
 <system>

Quote from xiazi64

I backed up the entire PI system (burning it to an iso file)

Rergarding this, although it worked (I don't know which program you use to make the ISO), it's advised to clone the SDcard to an IMG file.

Use Win32diskImager (Windows) or USBimager (Windows, Mac, Linux).

Flashing it back to the SDcard can be done with the above tools or RaspberryPi Imager, Balena Etcher, etc.

Quote from xiazi64

returned to the previously backed up system, everything was working and config.xml was not in a mess, which was very strange.

Yes, it's strange but still possible that the SDcard is going bad.

If you want, you can try to make a proper IMG of the SDcard with the OS running good (see above), flash the IMG to a brand new SDcard, check if all is OK and then replicate the steps you did to see if it's the same.

Before that, just make a copy of config.xml just in case it corrupts:

sudo cp -a /etc/openmediavault/config.xml /etc/openmediavault/config.xml.bak

No more thoughts I can say to you.

Quote from fbeye

Doesn’t it (SWAG) NEED to be 443, not modified 444? Cause https [is] 443 from WAN side. If you manually type 444 you bypassing default procedure, which is no no. You’ll need to reconfigure SWAG for 443, even if it means a macvlan or inside a VM.

No.

The guides here specificly state to use WAN 443 portforwarded to a different port on the LAN.

YML then reverses it from<different port>:443 again.

Quote from Waxywo76

ok i change

- 443:443 now is ok! Thanx

This means you made your router portforward WAN:LAN 443:443 and not

443:444.

Quote from Waxywo76

https.//subdomain.duckdns.org

You need to type a sub.subdomain (as is mentioned on the guide):

See the www. before the YOUR-SUBDOMAIN???

Quote from xiazi64

I have installed OpenMediaVault on my Raspberry Pi

How? Which guide did you follow?

On what media are you running OMV? SDcard? USB stick?

Quote from xiazi64

I seem to know the problem, it is located in the/etc/openmediavault/config. The XML files mysteriously appeared the messy code.

Have you been editing it either on CLI or via Windows or something else?

Or even worse, did you made a Shared Folder from /

Quote from xiazi64

When I restore to the previous backup system, the backup system reconfig.xml is not messy, and everything works fine.

How did you made the backup?

A bad config.xml can mean a bad media, corrupt OS or wrong editing.

You can force a filecheck just in case but it won't magically "fix" bad characters on files.

sudo touch /forcefsck

sudo reboot

Quote from jpmhh

I wanted to try fsck, but how?

sudo touch /forcefsck

sudo reboot

Quote from jpmhh

Randomly broken card?

Almost certain. Or maybe a knock-off in the middle of all the good one's.

If you see the Raspberry install guide for OMV, you will find there hints on how to test the SDcard:

omv6:raspberry_pi_install [omv-extras.org]

Quote from gderf

There was some change made that voided those credentials. But I don't know how to use it.

New Release from linuxserver/qbittorrent (don't know if it comes from upstream qbittorrent) now creates a random Password that is mentioned in the log:

The web UI is at `<your-ip>:8080` and a temporary password for the `admin` user will be printed to the container log on startup.

You must then change username/password in the web UI section of settings. If you do not change the password a new one will be generated every time the container starts.

Quote from chente

just have to make sure that the traffic from those different domains reaches the NPM IP.

Quote from chente

NPM will be in charge of redirecting traffic to each server or service (different IPs) depending on whether the origin is one domain or another and vice versa.

I understand how it can/should work.

What I never thought was that the Proxy Server (be it NPM, SWAG, whatever) could translate 2x (or more) names to IPs.

Maybe I need a better understanding on REGEX or whatever its called.

Since I never needed it, never wondered about it, but I can only consider that SWAG will also be able to do this.
[EDIT]

It seems, it is:

SWAG Reverse Proxy, handling multiple domains across multiple servers | WalnutHomelab
[/EDIT]

And more:
swag reverse proxy 2 different domains at DuckDuckGo