Once I did the debug, the red client turned yellow and it shows it started but I am still stuck with the following:

 

Think I got it figured out.   Once it got the error (above) I clicked on a new menu option on the client (Continue) and it's backing up the rest now.   Not sure why it matters but the PC has 2 drives and the machine boot is actually from the 2nd drive and I think it's getting confused.  But the big mystery is why changing log from warn to debug made the client wake up (it was red initially and now it's yellow)  I restarted the urbackup 50 times in troubleshooting lol.

Is there a client docker to go with the server? That way we can use this to backup some unraid files?

Is there a client docker to go with the server? That way we can use this to backup some unraid files?

Not required, this has already been discussed in this thread, you simply download the static binary client and run it on unRAID.

Sent from my 22021211RG using Tapatalk

is anyway we can set permissions for folders of backups clients ?

 

i found i have all clents dir/subdirs have drwxrwx--- but the old one have 

drwxrwxrwx ... and container UMASK 000 ...

On 5/13/2022 at 12:18 AM, turbojazz said:

I had to run the container as privleged for it to be able to complete the btrfs tests correctly and use it.  Check the container logs on startup.

According to the "running" section of the urbackup docker README, btrfs support requires adding the `--cap-add SYS_ADMIN` parameter to the docker run command (it can be added as an `EXTRA_PARAMETER` in the unraid template).

Quote

For BTRFS-Support add --cap-add SYS_ADMIN to the command above

After adding this parameter, I no longer need to run the container as privileged to use btrfs.

 

@binhex Thanks for all your work creating and maintaining this container!

 

I've just filed an issue on github to request that the file /etc/urbackup/backupfolder/ be automatically populated with the backup folder in order to improve support for btrfs snapshots.

 

I turned btrfs snapshots on by creating the file within the container last year without thinking and just discovered that it was (of course) blown away when the docker image changed during the update from urbackup 2.4 to urbackup 2.5. Completely my fault for not adding it in a truly persistent way. Urbackup is very resilient so it continued to make backups without btrfs. However, the fallout over ~6 months was hundreds of GBs of error messages in my urbackup log file along with backups that totaled 1-2TBs bigger than they should be.

 

In case anyone else hits this, my workaround for the time being is to add the following command which creates the file as a post-arg (note the leading semicolon):

; docker exec <urbackup-container> bash -c "mkdir -p /etc/urbackup; echo '/media' > /etc/urbackup/backupfolder"

 

EDIT: And it's already been fixed. Thanks again for the amazing support binhex!

Edited May 2, 20233 yr by pants

34 minutes ago, Masterwishx said:

 

Should we build from source the client ?

nope, no need, simply follow the install instructions for heading 'Binary Linux client (command line only; with auto-update; x86/AMD64/ARMv6+/ARM64) from the following url:- https://www.urbackup.org/download.html it worked for me (tested inside a docker container for safety).

35 minutes ago, binhex said:

nope, no need, simply follow the install instructions for heading 'Binary Linux client (command line only; with auto-update; x86/AMD64/ARMv6+/ARM64) from the following url:- https://www.urbackup.org/download.html it worked for me (tested inside a docker container for safety).

 

Thanks, do you mean https://hndl.urbackup.org/Client/2.5.23/UrBackup Client Linux 2.5.23.sh file ?

1 hour ago, Masterwishx said:

 

Thanks, do you mean https://hndl.urbackup.org/Client/2.5.23/UrBackup Client Linux 2.5.23.sh file ?

yep, so use the command shown on that page to download and execute that script:-

TF=$(mktemp) && wget "https://hndl.urbackup.org/Client/2.5.23/UrBackup%20Client%20Linux%202.5.23.sh" -O $TF && sudo sh $TF; rm -f $TF

once you have gone through the prompts you will end up with the urbackup client being on the path, if you want it to persist reboots then copy the binary file created by the script to persistent storage and then add a line to your 'go' file to copy it from persistent storage to somewhere on the path, e.g. /usr/local/bin/

@binhex can you say if was something changes in permissions of container in past ?

Because i have old client PC with 777 permissions for folder :

 

 

but all newer clients have 770 :

 

 

or maybe i missing something ?

UMASK=000 in container .

On 1/10/2023 at 12:43 PM, Kilrah said:

BTW I just changed log config to info after noticing my urbackup.log was 1.2GB because of a slew of

 

Did you fixed this log problem ? (ERROR: Sending broadcast failed! (ipv6) )

I think should be fixed by set broadcast interface : 

#Interfaces from which to send broadcasts. (Default: all). #Comma separated -- e.g. "eth0,eth1" BROADCAST_INTERFACES="eth0"

https://forums.urbackup.org/t/error-sending-broadcast-failed-ipv6/8375/5

 

 

I still have them but will try that...

 

EDIT: eth0 gives no connectivity, needs br0 instead. Will see in a while if there are still log entries...

Edited May 13, 20233 yr by Kilrah

44 minutes ago, Kilrah said:

 

EDIT: eth0 gives no connectivity, needs br0 instead. Will see in a while if there are still log entries...

 

i still was no time to check it ,do you mean only br0 working ? its bridge no ? urbackup uses host .....

so bond0 that i have will also not wotk i think ...

Yeah container is on Host and looking at the interfaces in the container the one called br0 was the one with the server's ip so I tried that and it worked.

 

Looking at older log it seems I got the error 3 times every minute and haven't seen any yet since changing so I guess that works.

Edited May 13, 20233 yr by Kilrah

25 minutes ago, Kilrah said:

Looking at older log it seems I got the error 3 times every minute and haven't seen any yet since changing so I guess that works.

 

Yes its working , but also found some other errors

I don't have those, but my destination is an array-only share of xfs drives.

Edited May 13, 20233 yr by Kilrah

1 hour ago, Kilrah said:

I don't have those, but my destination is an array-only share of xfs drives.

 

Yes , im also have only xfs drives but only today founded this error , and also SQLite Error 14: 'unable to open database file' 

so i think its broken somehow tryed to fix it by older backups but still have it :(

 

On 1/9/2023 at 3:15 AM, JonathanM said:

Are any other people seeing this in their logs?

 

WARNING: SQLite: os_unix.c:36982: (40) openDirectory(/var/urbackup) -  errorcode: 14

 

The only google result that looks relevant references symlinks for var/urbackup no longer working and needing bind mounts, so I'm wondering if I messed something up in my config.

 

To be clear, I can't see any issues besides a WARNING in the log file, everything seems to be working fine.

 

@JonathanM , @cayman Found this error , tryed to fix by older CAbackups of urbackup app , but still ,did you fixed this ?

@binhex the container working with 99:100 by default ,but in log i see user set to : 0 after 99 ? maybe becose of it i have error 14 of db , however all working fine ?

 

 

list of /var/urbackup :

 

I have some as well but it doesn't seem to be a problem, this instance has been running fine since last summer...

 

 

It might be because /var/urbackup is a symlink to /config/urbackup instead of being the actual storage location uerbackup expects?

Edited May 14, 20233 yr by Kilrah

42 minutes ago, Kilrah said:

I have some as well but it doesn't seem to be a problem, this instance has been running fine since last summer...

 

Yes all working fine , also found that i had this warning from 2 month ago, but yesterday i was thinking that it new and my sqlite DB is broken , so 2 hours i tried to fix it with CAbackups ...

 

45 minutes ago, Kilrah said:

It might be because /var/urbackup is a symlink to /config/urbackup instead of being the actual storage location uerbackup expects?

 

Dont know really, maybe @binhex should knows better ?

i checked forum urbackup and this something with permissions .... someone get rid of this after clean install ...

12 hours ago, Masterwishx said:

 

@JonathanM , @cayman Found this error , tryed to fix by older CAbackups of urbackup app , but still ,did you fixed this ?

no - but everything seems to work properly

11 minutes ago, cayman said:

no - but everything seems to work properly

Same ..

On 9/12/2022 at 7:21 PM, KluthR said:

Basically you remove UMASK, PUID and PGID variables and change the the variable

 

if you still use official container did you added PUID=1000, PGID=100 ? founded it have newer  v2.5.31 ...