I'm trying to backup whole my Unraid conf on an external HDD attached to USB-C port.

I mount it using Unassigned plugin, and made two folder for backup.

 

 

 

 

When it started manually backup, it get an error for privilegies od something.

 

 

I tried everything i know about changing write/read privilegies, changing owner on the folder, recreate them few times, but nothing solves it until now.

 

Tnx for help.

i got a couple errors on my most recent backup:

 

Description: Please check the backup log!

Copying flash backup to destination failed!

 

Description: Please check the backup log!

Error While backing up VM XMLs. Please see the debug log.

 

I couldn't find the log from either of these locations.

 

It's backed them up without any problems for months. it also did send another email saying the backup completed.

 

 

I just got into the world of Unraid and started to create some containers and was wondering if this plugin backs up also the database such as for vaultwarden

Afternoon, I'm getting errors throughout the backup (mainly with arrs) telling me that the file is not found in the archive.


[25.03.2025 04:13:25][❌][ClamAV] tar verification failed! Tar said: tar: /mnt/user/appdata/clamav: Not found in archive; tar: Exiting with failure status due to previous errors
[25.03.2025 04:14:30][❌][bazarr] tar verification failed! Tar said: tar: /mnt/user/appdata/bazarr: Not found in archive; tar: Exiting with failure status due to previous errors
[25.03.2025 04:15:55][❌][sonarr] tar verification failed! Tar said: tar: /mnt/user/appdata/sonar Not found in archive; tar: Exiting with failure status due to previous errors

My debug ID is: 75af9cd9-60e5-4bc0-8e40-8710bfb50b49

I've got some of my important containers not stopping during backup (i.e. emby, home assistant, unraid etc. as this causes other issues. Any idea what i might have misconfigured to be causing these errors please?

 

When creating a backup for qBittorrent, I keep getting this error message:

tar creation failed! Tar said: tar: /mnt/user/appdata/qbittorrent/BT_backup: file changed as we read it

 

My backup folder also gets a disconcerting suffix "failed":

ab_20250325_222009-failed

 

How can I prevent this error?  What is causing this error?

 

Backup Type = Stop, backup, start for each container.

Skip stopping of containers? = No.

 

Thank you.

I'm struggling with a frigate container that gets reset on backup. The config file gets deleted and when the container reboots it defaults to standard out of the box settings. 

I cant find anything in the backup log. Mapped the log catalog in the frigate container in hopes of catching the bug, but that only shows a boot of a fresh install of frigate. 

The same thing happend a while back with a redis and postgresql16 (trimmed for immich) container. The solution in that case was to replace both containers with the same containers form forks. 

Ill post some logs if this persist. 

 

Just wanted to check if anyone else have had similar problems. 

 

I am having a recent issue where the plug in no longer skips the dockers I specify. I used the debug form and it indicates that I have selected the dockers to be skipped. Has anyone had any recent issues with this or solutions? Example debug text below.

 

            [adminer] => Array
                (
                    [skip] => yes
                    [group] => 
                    [backupExtVolumes] => no
                    [updateContainer] => 
                    [exclude] => 
                    [skipBackup] => no
                    [verifyBackup] => 
                    [ignoreBackupErrors] => 
                    [dontStop] => 

 

How many containers? If you have tons see earlier posts about the settings getting truncated.

2 hours ago, Kilrah said:

How many containers? If you have tons see earlier posts about the settings getting truncated.

I definitely have tons of containers after installing the nextcloud-aio. I think I found the correct work around you are referring to. My understanding is I need to run this command only if I reboot the server?

 

 

curl -s https://raw.githubusercontent.com/Commifreak/unraid-appdata.backup/refs/heads/master/scripts/hotfix_php_max_input_vars.sh | bash

 

Edited March 29, 20251 yr by benmchale

Yes, only a reboot would reset that fix.

On 3/27/2025 at 1:26 AM, Jaybau said:

How can I prevent this error?  What is causing this error?

Seems like something else is accessing files of this container. Is some other container using the same data?

On 3/22/2025 at 9:45 PM, steff_92 said:

I'm trying to backup whole my Unraid conf on an external HDD attached to USB-C port.

I mount it using Unassigned plugin, and made two folder for backup.

 

 

 

 

When it started manually backup, it get an error for privilegies od something.

 

 

I tried everything i know about changing write/read privilegies, changing owner on the folder, recreate them few times, but nothing solves it until now.

 

Tnx for help.

I found the solution.

I have previous setup destination folder and save the configuration.

When i try now to backup, a change the destination folder, but never click SAVE, instead i try direct manual backup. But the system reads from the saved configuration which is no longer available. I have noticed after few day strugling.

 

So the answer is, after you change destination driver or some other setting, you must SSAVE IT, and then DO the manual backup.

I am trying to use a custom cron schedule but it doesn't seem to be (fully) working.  The cron entry is 

 

0 3 1-7,15-21 * 4

 

which should run at 0300 on a Thursday in days 1-7 and days 15-21 (essentially twice a month). However, this month the backup ran on 6 March but not on 20 March.

 

I have checked /etc/cron.d/root and the entry seems to match what I am trying to achieve:
 

# Appdata.Backup cron settings
0 3 1-7,15-21 * 4 php /usr/local/emhttp/plugins/appdata.backup/scripts/backup.php > /dev/null 2>&1

 

Any ideas - does UNRAID not interpret the 1-7,5-21 correctly?

Edited March 30, 20251 yr by Ludditus
misspelling

37 minutes ago, Ludditus said:

0 3 1-7,15-21 * 4

crontab.guru (not necessarily definitive) interprets that as

Quote

At 03:00 on every day-of-month from 1 through 7 and every day-of-month from 15 through 21 and on Thursday.

Not sure how the particular linux Unraid is using might interpret it.

 

Suggest you try something simpler.

On 3/28/2025 at 11:53 PM, KluthR said:

Seems like something else is accessing files of this container. Is some other container using the same data?

That's probably the problem.  qBittorrent might be accessed by the other containers qbt_manage or cross-seed.  I also have appdata configured to stop a container, do the backup, start the container; which leaves qbt_maanage and cross-seed potentially accessing the same folders, though shouldn't be writing to the folders.

 

If this is the root cause, it would be better if I can stop and group multiple containers for backup, and start the containers in the same original order, and with the same "wait".  Stop/start all at the same time is not ideal because I have too many containers, and I am backing up 50 GB, all which takes a lot of downtime.

 

SOLUTION:

Clicking on app to get the menu for group and excluding folders is awesome!  Thank you.

 

Edited March 31, 20251 yr by Jaybau

4 hours ago, Jaybau said:

If this is the root cause, it would be better if I can stop and group multiple containers for backup, and start the containers in the same original order, and with the same "wait".

That's why the plugin has this exact functionality  

Edited March 30, 20251 yr by Kilrah

5 hours ago, Ludditus said:

I am trying to use a custom cron schedule but it doesn't seem to be (fully) working.  The cron entry is 

 

0 3 1-7,15-21 * 4

 

which should run at 0300 on a Thursday in days 1-7 and days 15-21 (essentially twice a month). However, this month the backup ran on 6 March but not on 20 March.

 

I have checked /etc/cron.d/root and the entry seems to match what I am trying to achieve:
 

# Appdata.Backup cron settings
0 3 1-7,15-21 * 4 php /usr/local/emhttp/plugins/appdata.backup/scripts/backup.php > /dev/null 2>&1

 

Any ideas - does UNRAID not interpret the 1-7,5-21 correctly?

duplicate the script. and set 1 cron type unraids didn't process the separate for the and condition...

as the cron expression will never execute due to the comma... as it seems to read and stop at the comma...
https://www.crondrive.com/test-cron-expression?utf8=✓&expression=0+3+1-7%2C15-21+*+4&time_zone=UTC&commit=Test

 

 

3 hours ago, Jaybau said:

That's probably the problem.  qBittorrent might be accessed by the other containers qbt_manage or cross-seed.  I also have appdata configured to stop a container, do the backup, start the container; which leaves qbt_maanage and cross-seed potentially accessing the same folders, though shouldn't be writing to the folders.

 

If this is the root cause, it would be better if I can stop and group multiple containers for backup, and start the containers in the same original order, and with the same "wait".  Stop/start all at the same time is not ideal because I have too many containers, and I am backing up 50 GB, all which takes a lot of downtime.

you will need to make a docker stop script to start before the backup starts and then make a docker start script to run after the backup stuff Finnish to start them in a order....

also you may need to adjust the docker order in the plugin.

1 hour ago, bmartino1 said:

you will need to make a docker stop script to start before the backup starts and then make a docker start script to run after the backup stuff Finnish to start them in a order....

Not at all, as said just above the plugin supports grouping.

11 minutes ago, Kilrah said:

Not at all, as said just above the plugin supports grouping.

then waht and were are you defining grouping... As I don't see that option and have to use script to stop the docker beforehand to not cause database corruptions...

you referring to the start order?



Immich can't be on if prostgrees and redis are not on...

7 minutes ago, bmartino1 said:

then waht and were are you defining grouping..

Click on container, set "Member of group", then you'll have them in a separate group you can reorder.

 

1 minute ago, Kilrah said:

Click on container, set "Member of group", then you'll have them in a separate group you can reorder.

 

Thank you. hidden menu under the docker name...

 

new testing to do. Thank you.

sad panda. I double click and try to open the group menu. tried other browers chrome, firefox, edge... incognito modes etc. I can't seem to make a group in the plugin...

 

Type a group name, save settings, then it'll be available to choose for others.

5 hours ago, bmartino1 said:

duplicate the script. and set 1 cron type unraids didn't process the separate for the and condition...

as the cron expression will never execute due to the comma... as it seems to read and stop at the comma...
https://www.crondrive.com/test-cron-expression?utf8=✓&expression=0+3+1-7%2C15-21+*+4&time_zone=UTC&commit=Test

 

 

Thanks. That seems like a very straightforward solution - I will try it out.