2 hours ago, kjames2001 said:

i get this error in syslog: root: error: /Settings/AB.Main: missing csrf_token

https://forums.unraid.net/topic/46802-faq-for-unraid-v6/page/2/?tab=comments#comment-545988

On 7/24/2023 at 6:35 AM, sonic6 said:

Docker stop timeout is at 300seconds... but i think there must come addition delay from the plugin itself to work correctly

No. The attached log mentions errors coming out of docker daemon. One friend had similar issues and they were gone after switching fo ipvlan. No idea if that matters.

If you setup your Unraid notification correctly, you should have received an email for every error line.

 

20 hours ago, Slippin' Jimmy said:

Can you please add a backup type of backing up without stopping the docker containers?

This is already possible. See per container settings.

 

17 hours ago, OcifferAction said:

I did have one backup complete successfully and I could view the log then, but when it fails I can't see the log.

Any reboot since then?

 

16 hours ago, LoyalScotsman said:

please can i get some assistance with this

Sure. This is related to the common Plex backup issue: Some files inside Library/ApplicationSupport could not be backed up. It seems we have to collect common excludes for plex related apps.

 

Please try to exclude "Library/Application Support" - if that doesnt matter for disaster recovery.

8 hours ago, kjames2001 said:

i get this error in syslog: root: error: /Settings/AB.Main: missing csrf_token

Interesting, since I got another user (or was it you?) with the same issue recently. CSRF Token are handled by Unraid as far as I know. I have to check this later.

16 minutes ago, KluthR said:

were gone after switching fo ipvlan

i am on ipvlan.

i will try "stop all containers, backup, start all", but this isn't a fix... just a "workaround".. sadly to say it again... with the old appdatabackup it worked fine.

 

16 minutes ago, KluthR said:

If you setup your Unraid notification correctly, you should have received an email for every error line.

i think it is, because i got error notification via email/telegram/gui for other errors.

sometimes the logs from appdata backup completely missing.
 

 

8 minutes ago, sonic6 said:

i will try "stop all containers, backup, start all", but this isn't a fix

The stop mechanism is exactly the same for all options.

 

8 minutes ago, sonic6 said:

i think it is, because i got error notification via email/telegram/gui for other errors

Have to test this later then. Currently Iam not at home.

3 hours ago, KluthR said:

Please try to exclude "Library/Application Support" - if that doesnt matter for disaster recovery.

 

just seen this so i have excluded this from the next backup what contents is this excluding ?

i suspect this will exclude users watched data ? which if i am correct the metadata

only reason i question it is cause the old appdata backup ran fine and i don't want to completely remove the old one until i know this is working correctly

Edited July 25, 20232 yr by LoyalScotsman

2 hours ago, LoyalScotsman said:

only reason i question it is cause the old appdata backup ran fine and i don't want to completely remove the old one until i know this is working correctly

Which old one? From squid or my fork of the previous appdata?

5 hours ago, KluthR said:

Interesting, since I got another user (or was it you?) with the same issue recently. CSRF Token are handled by Unraid as far as I know. I have to check this later.

cool, thanks. look forward to your update.

 

NB: No, its not me.

Edited July 25, 20232 yr by kjames2001

Something strange I noticed about compression - when using multicore tar.zst the resulting backup for me was 11gb but when using normal tar.gz the backup was compressed much more resulting in only a 9GB backup.

 

Is this to be expected? My impression is that zstd would normally result in better compression, not worse. However with this plugin it seems that using the multicore compression "newer" option has made the backup take more space than just the older style compression.

 

Even stranger is before when using the old backup (I migrated today to this new plugin) the backups were smaller still, being only around 6GB each. Unclear what options have changed surrounding compression to have made such dramatic increases in size. Perhaps exposing the compression level to the user would be beneficial so we can set a max level 9 if desired?

 

@PeteAskingCant tell. The compression is at default settings. Only the T setting is set to 0 to use all cores.

Setting the compression level is something for the wishlist

On 7/23/2023 at 6:35 PM, KluthR said:

You excluded /mnt for Krusader resulting in failing verification. Please remove the exclusion or skip the whole container.

On 7/18/2023 at 1:42 PM, LoyalScotsman said:

Probably a silly question, but would removal of the exclusion cause backing up /mnt too?

32 minutes ago, Gico said:

Probably a silly question, but would removal of the exclusion cause backing up /mnt too?

no, because of this:

Edited July 26, 20232 yr by sonic6

On 6/24/2023 at 9:01 AM, jcofer555 said:

how do i backup the actual appdata like the old plugin? when i do a backup all i get is the .xml files
 

I I have the same issue... I a glad i saw it before i ru into an issue and then having no recent backup.. do you already know how to fix it?

1 hour ago, quiksilver2 said:

I I have the same issue... I a glad i saw it before i ru into an issue and then having no recent backup.. do you already know how to fix it?

@quiksilver2 and @jcofer555 Please share the debug log or the view for Internal volumes of the container sections. It feeld like all volume mappings are treated as external because incpmplete Appdata sources configuration. Note the help block for that field.

8 hours ago, KluthR said:

@quiksilver2 and @jcofer555 Please share the debug log or the view for Internal volumes of the container sections. It feeld like all volume mappings are treated as external because incpmplete Appdata sources configuration. Note the help block for that field.

my issue is resolved after figuring out which volumes i needed to select and which to deselect, a much more cumbersome process. though i guess being able to backup more things can be a good thing

Edited July 26, 20232 yr by jcofer555

I'm getting a invalid source error. I made a backup of my appdata and it appears to have been successful. I've attached the debug log from the backup. When I try to restore from that exact same backup I get the invalid source error. I am not trying to restore my appdata from the old plugin, you can see in the first image I am selecting the ab_ folder. 

 

ab.debug.log

You have to select the parent folder that holds the ab_ folders.

On 7/25/2023 at 12:37 PM, KluthR said:

Which old one? From squid or my fork of the previous appdata?

yes the oen from Squids i still have it installed but finally got around to updating to your one as per his notes when accessed his one 

I have an other q

On 7/25/2023 at 8:02 AM, KluthR said:

Interesting, since I got another user (or was it you?) with the same issue recently. CSRF Token are handled by Unraid as far as I know. I have to check this later.

Kindly let me know if i can config the plugin some how? like a config.yml file somewhere? and point me to a sample config plz, so that i can make use of it to back up my appdata. feeling very insecure right now.

On 7/25/2023 at 8:34 AM, KluthR said:

The stop mechanism is exactly the same for all options.

 

Have to test this later then. Currently Iam not at home.

the same againe:

 

same Container aren't stopped when the backup started, so they aren't backuped and also not started again...

and there is again no log which i can post... no normal log file, no debug log.

now i was using the "Stop all container,..." option. This works, because the timeout for those containers are high enough.

ab.debug (1).log

 

using 

docker container stop xteve

works fine, the container stops immediately

Edited July 31, 20232 yr by sonic6

On 7/25/2023 at 2:02 AM, KluthR said:

Any reboot since then?

I have rebooted a few times since the successful backups with logs. Now I've only been receiving failed backups with no logs.

 

I just did another fresh reinstall of the plugin and ran the backup manually. I'm seeing the log now.

 

Errors are due to having multiple game servers relying on the appdata folder steamcmd. I have the plugin set to stop the containers sequentially. While one of my game servers does a backup, another game server changes some contents in the steamcmd folder. I have the internal volume steamcmd skipped for now during the backup.

So when i first set up this it was failing so i now have a selection of failed backups that i cannot delete, i have tried to delete them via kurseder but it refuses to delete any advice on how to delete the failed one because taking up unnessaray space

 

On 7/30/2023 at 8:58 PM, kjames2001 said:

Kindly let me know if i can config the plugin some how? like a config.ym

Yes but manual configuration is not that easy. I got no feedback yet on the CSRF thing I asked internally. Did you tried another browser? Mobile phone?

 

On 7/31/2023 at 6:32 AM, sonic6 said:

and there is again no log which i can post... no normal log file, no debug log

Anything in the system log in that case?

On 7/31/2023 at 7:29 AM, sonic6 said:

now i was using the "Stop all container,..." option. This works, because the timeout for those containers are high enough.

The mechanism works the same, its just the order of execution that changes, so no difference here. I cant tell whats the issue here.

 

4 hours ago, LoyalScotsman said:

i now have a selection of failed backups that i cannot delete

File owner is root and I assume the rights are "rwxr-xr-x", so I add a permission fix in a later update.

12 hours ago, KluthR said:

Yes but manual configuration is not that easy. I got no feedback yet on the CSRF thing I asked internally. Did you tried another browser? Mobile phone?

Yes, still the same problem.