[Plugin] Appdata.Backup

[Kilrah]

2 hours ago, ssmith369 said:

Maybe Syncthing can syn to One Drive? Worth looking into.

No, but rclone is the common solution for that.

[schlappy]

I'm trying to restore a corrupted Plex DB.

 

I started with deleting the current version, then using CA Cleanup Appdata to clear out the plex directory.

 

I then downloaded the last stable version of the linuxserver plex app. 

 

In the restore plugin:

• I click on restore, select my USB.
• Select a backup from about a week ago; the problems started about 3 days ago
• The template is correct, so I just select the container for plex and restore it.

The app opens but then starts to configure plex as if it were a brand new install.  I was hoping to not have to do something like that, i would want my install to be the same.

 

Any information on why the appdata backup/restore app isn't actually backing up my exact server or why this isn't working?

[ssmith369]

22 hours ago, Kilrah said:

No, but rclone is the common solution for that.

Thank you. I just took a look and there are several rclone variants. I tend to use binhex dockers where possible. Do you just use the command line rclone, or do you use the rclone with GUI?

 

 

cheers

 

SS

Edited July 6 by ssmith369

[Kilrah]

I don't use it myself.

 

[Glassed Silver]

@KluthR

The needed change for the debugging command you want us to insert has to go into

/usr/local/emhttp/plugins/appdata.backup/pages/content/settings.php

 

Probably a difference between the published version everyone else is using and you using a dev environment?

 

Either way, I redacted the resulting file to include just one container that doesn't belong to any group as a baseline and then all containers of my MariaDB group.

 

I noticed that the sort order isn't stored in ANY item in the unredacted file either, I wager that the order saving method must drop the actual writing, but I don't know.

 

Anyhow, here goes nothing:

 

 

abdebugPOST_redacted.txt

Edited July 6 by Glassed Silver
add ping

[oliver]

hi im trying to make a manual backup with default settings and get an error the destination is not writeable.  I am trying to write to a folder inside of appdata:

 

debug id is 9860c8a1-b3cd-4349-b3a1-0178ca1ef98b

 

EDIT: fixed by manually creating the folder as described here:

there's something wrong with the scripts ability to create the folder properly itself.

Edited July 7 by oliver

[itimpi]

3 hours ago, oliver said:

I am trying to write to a folder inside of appdata:

This seems to defeat the normal use of this plugin to make a backup in case the drive holding appdata fails.   Normally you would backing up to the array or to a UD managed drive.

[oliver]

5 hours ago, itimpi said:

This seems to defeat the normal use of this plugin to make a backup in case the drive holding appdata fails.   Normally you would backing up to the array or to a UD managed drive.

 

For now im manually moving it off the array, I just wanted to get an initial backup created manually.  

[blitzio]

My backup started failing this week, and it seems to be due to Plex. 

 

It was working just fine before and I didn't change anything in my exclusions which are set up like this:

 

 

Does anyone have any suggestions on how to troubleshoot?

 

Thanks!

[blitzio]

Potential fix: I saw another member changed /user/ to the server name, which worked.

 

So I tried, and this time, the backup did not have any errors:

[KluthR]

13 hours ago, blitzio said:

Potential fix: I saw another member changed /user/ to the server name, which worked

That makes absolutely no sense. Changing those paths are exclduing the exclusions.

 

Your issue is the missing "appdata/plex" within the paths. Your exclusions should start with "/mnt/user/appdata/plex/"

[blitzio]

10 hours ago, KluthR said:

That makes absolutely no sense. Changing those paths are exclduing the exclusions.

 

Your issue is the missing "appdata/plex" within the paths. Your exclusions should start with "/mnt/user/appdata/plex/"

 

Understand, thanks for clarifying. So can I check that instead of the initial recommended exclusion list:

 

/mnt/user/Library/Application Support/Plex Media Server/Cache
/mnt/user/Library/Application Support/Plex Media Server/Media
/mnt/user/Library/Application Support/Plex Media Server/Metadata

 

My setup is different and I should set it instead to the following:

 

/mnt/user/appdata/plex/Cache/
/mnt/user/appdata/plex/Media/
/mnt/user/appdata/plex/Metadata/

 

Because when I check my folders, I only seem to have the Cache one. The other 2:

 

/mnt/user/appdata/plex/Library/Application Support/Plex Media Server/Media

/mnt/user/appdata/plex/Library/Application Support/Plex Media Server/Metadata

 

Don't exist and when I run backup, that's where I get the failed error.

Edited July 9 by blitzio

[KluthR]

Regarding contaioner order: I created myself ~160 dummy containers. The AppdataBackup page settings is now huge but ordering works as expected.

 

@affected users, could you tell me which browsers you use? Due to a debug logging issue (https://forums.unraid.net/bug-reports/stable-releases/61210-javascript-consoledebug-and-consolelog-disabled-due-to-vue-i18n-r3084/), you have to make another change to display me some information.

 

Could you open the settings.php file once again, scroll to line ~907 (the line reading "$('#abSettingsForm').on('submit', function () {". A few lines later (~922) you should see two lines of "        });". The latter of those both matter for the next step:

 

Replace the WHOLE BLOCK with this:

 

    $('#abSettingsForm').on('submit', function () {
        console.error("SUBMIT!");
        let mainValue = $('#containerOrderSortable').sortable('serialize', {
            expression: /(.+?)_(.+)/
        });
        console.error('Main order value: ', mainValue);
        $('#containerOrder').val(mainValue);

        $("input[id^='containerGroupOrder']").each(function (index) {
            console.error("Processing groupOrder " + $(this).attr('id'));
            let groupValue = $('#' + $(this).attr('id') + '_Sortable').sortable('serialize', {
                expression: /(.+?)=(.+)/
            });
            console.error('Group order value', groupValue);
            $(this).val(groupValue);
        });
        console.error('Final form:', $(this).serialize());
        console.error("Submit halted!");
        return false;
    });

 

Then save, reload settings page and hit F12 (the developer console). Open the tab called "Console". Hit Submit on the settings page. Nothing will happen, which is fine. But the console will output some red lines. Maybe you can send me a PM with the result?

 

Should look like:

 

 

 

To re-enable submission either redo all changes ot just remove the "return false;" line at the last few lines.

[KluthR]

On 7/5/2024 at 11:28 PM, schlappy said:

I'm trying to restore a corrupted Plex DB

Need more info.

 

If the db path was included inside the backup, it should be restored correctly and the 1:1 sorce path. You could check the file contents by browsing the backup destination/run folder and open the archive which holds the database. the restore should also create a debug file which you could submit.

 

On 7/7/2024 at 10:54 AM, oliver said:

there's something wrong with the scripts ability to create the folder properly itself.

Just checked - the script does not CREATE the destination at all! One has to create the main destination itself.

 

 

[KluthR]

2 hours ago, blitzio said:

recommended exclusion list

The list has a info above: "Each following paths need to be prefixed with your correct appdata path!". And this prefix is (in your case) "/mnt/user/appdata".

 

2 hours ago, blitzio said:

My setup is different and I should set it instead to the following

Thats totally ok then.

[blitzio]

18 hours ago, KluthR said:

The list has a info above: "Each following paths need to be prefixed with your correct appdata path!". And this prefix is (in your case) "/mnt/user/appdata".

 

Thats totally ok then.

 

Ok thanks for the sanity check, I will give it try. Appreciate it

[fir3drag0n]

On 6/11/2024 at 7:45 AM, beasthouse said:

Hey legends, have a few errors popping up in my logs.

 

Not sure what this one means, let alone how to fix it:

 

[10.06.2024 02:28:03][❌][code-server] tar verification failed! Tar said: tar: Removing leading `/' from member names; mnt/user/appdata/homeassistant/home-assistant_v2.db: Mod time differs; mnt/user/appdata/homeassistant/home-assistant_v2.db: Size differs; mnt/user/appdata/homeassistant/home-assistant.log: Mod time differs; mnt/user/appdata/homeassistant/home-assistant.log: Size differs; mnt/user/appdata/homeassistant/home-assistant_v2.db-wal: Mod time differs; mnt/user/appdata/homeassistant/home-assistant_v2.db-wal: Contents differ; mnt/user/appdata/homeassistant/home-assistant_v2.db-shm: Contents differ

As above:

 

[10.06.2024 02:51:40][❌][imagemaid] tar verification failed! Tar said: tar: Removing leading `/' from member names; mnt/app_cache/appdata/plex/Logs/Plex Media Server.log: Mod time differs; mnt/app_cache/appdata/plex/Logs/Plex Media Server.log: Size differs; mnt/app_cache/appdata/plex/Plug-in Support/Databases/com.plexapp.plugins.library.db: Mod time differs; mnt/app_cache/appdata/plex/Plug-in Support/Databases/com.plexapp.plugins.library.db: Contents differ; mnt/app_cache/appdata/plex/Plug-in Support/Databases/com.plexapp.plugins.library.db-wal: Mod time differs; mnt/app_cache/appdata/plex/Plug-in Support/Databases/com.plexapp.plugins.library.db-wal: Contents differ; mnt/app_cache/appdata/plex/Plug-in Support/Databases/com.plexapp.plugins.library.db-shm: Mod time differs; mnt/app_cache/appdata/plex/Plug-in Support/Databases/com.plexapp.plugins.library.db-shm: Contents differ

 

I have double checked the mappings for this one and it keeps occurring:

 

[10.06.2024 02:26:23][❌][cross-seed-misc] '/mnt/user/data/torrents/misc/cross-seed' does NOT exist! Please check your mappings! Skipping it for now.

 

And last one, and why I am here 😉:
 

[10.06.2024 02:57:10][⚠️][Main] An error occurred during backup! RETENTION WILL NOT BE CHECKED! Please review the log. If you need further assistance, ask in the support forum.

 

 

Any help would be greatly appreciated.

 

Did you resolve it? I do have quite the same error, interestingly as well with home assistant container:

 

[08.07.2024 07:22:06][❌][hassConfigurator] tar verification failed! Tar said: tar: Removing leading `/' from member names; mnt/user/appdata/Home-Assistant-Container/.storage/scheduler.storage: Mod time differs; mnt/user/appdata/Home-Assistant-Container/homematicip_local/cache/RaspberryMatic_homematic_devices.json: Mod time differs; mnt/user/appdata/Home-Assistant-Container/homematicip_local/cache/RaspberryMatic_homematic_paramsets.json: Mod time differs; mnt/user/appdata/Home-Assistant-Container/home-assistant.log: Mod time differs; mnt/user/appdata/Home-Assistant-Container/home-assistant.log: Size differs

 

[bmartino1]

I would like some scripting help and a bit more understanding of the option in the app.

More info needed:

I'm trimming my docker applications but at once time, I was running 6 dockers to complete 3 server needs. and using this plugin to backup.

Old needs:
2x mysql via maraia DB
A database for nextcloud

A Database for Photoprism
1x MongoDB
A database for unifi

 

3x service docker applications:
nextcloud

photoprism

unifi-network-appliaction

Problem:
So when using the backup tool it will backup but due to the start order of shutdown and startup... i would like to stop all the containers and or stop the server docker first then the database to maintain data and not have errors inside the serve docker because the database is not there...

It would be ideal to have a option to not start dockers back up. and to have a custom cmand to start them when done. ATM I only have a database and photo prism...

I have 2 scripts a teardown and startup

teardown:

docker stop PhotoPrism
docker stop Mariadb-SQLPhotoPrism
#docker stop nextcloud
#docker stop Mariadb-SQLnextcloud
#cd /mnt/cache/appdata/netprobe_lite && docker compose down

startup:
 

#docker login  -u dockerusername --password apipasswordkey

docker start Mariadb-SQLPhotoPrism
#docker start Mariadb-SQLnextcloud
sleep 5
docker start PhotoPrism
#docker start nextcloud
#cd /mnt/cache/appdata/netprobe_lite && docker compose up -d
exit 0

Now to the script issues and hopefully a better understanding.
I had teardown sh script as a pre run
startup as a post run.

While it did stop the dockers, it appears that the backup starts the docker after the backup.

This wouldn't be a problem as I stopped the docker before the plugin begins its settings of backup and update if available... this appears to also start the docker after backup. Which would be fine if it followed a start order.

But the start order will not save nor follow that rule set. Each time I load the plug page via setting, the start order is different. and the logs when this does an auto backup show different dockers starting, stopping when backed up...

 

Ideally I would want it to save and run in this order:


any idea on how to fix this to not have the service docker photo prism running without mariadb?

Edited July 13 by bmartino1
spelling

[Kilrah]

Sounds like all you're asking for is handled by the grouping feature, no scripts needed.

 

Edited July 14 by Kilrah

[LadyHaley]

Hello, I am having issues getting the backups to actually run, they dont seem to be running with the schedule I have set, and manual backups are not working. My debug ID is 1687cfe1-8348-4322-8c64-a736e3eeee46
 

[Kilrah]

It says right there it can't write to the destination, so check that.

[bmartino1]

16 hours ago, Kilrah said:

Sounds like all you're asking for is handled by the grouping feature, no scripts needed.

 

It's not keeping the order even with a save  that my main issue. how hard would it be to make a separate save button for that feature?

[KluthR]

So it seems you are affected by this bug as well. I cant reproduce it yet. Please view a fee posts before yours, I wrote a few things about it. Maybe you could male the mentioned changed and report back?

[NeoJoris]

On 7/14/2024 at 11:14 AM, LadyHaley said:

Hello, I am having issues getting the backups to actually run, they dont seem to be running with the schedule I have set, and manual backups are not working. My debug ID is 1687cfe1-8348-4322-8c64-a736e3eeee46
 

I have the same issue. AB can't find the 'remotes' location (for me that is /mnt/remotes/SMB_NAS_Backup/Areopagus/)

@Kilrah via console the folder is available, i think this is another issue.

[LadyHaley]

2 hours ago, NeoJoris said:

I have the same issue. AB can't find the 'remotes' location (for me that is /mnt/remotes/SMB_NAS_Backup/Areopagus/)

@Kilrah via console the folder is available, i think this is another issue.

Yeah I have tried different file paths and creating them manually beforehand, I dont know what im doing wrong, none of my docker containers seem have trouble using the same directory