[Solved] Lost Docker my*templates

[spidobot]

Hi folks,

 

I tested and ran unraid for some time and configured everything for my use cases perfectly. only thing missing was a flash backup routine.

fun thing is, that my flash failed during backup. physical machine wouldn't even boot up again, so i had to setup a new flash and configure everything again, no problem.

downside is, that i lost all docker (my)templates during this.

 

my issue is now, that all dockers still work, but updating and config-editing isn't possible anymore, since all templates are gone.

for example:

 

my question for you is, if there is a possibility to only get the "standard" config XMLs? i could take everything else from there through editing them in the my*templates folder on the flash.

 

 

thanks in advance

 

 

p.s. if there is already a thread with this question, i apologize. i googled and searched the forums, but couldn't find anything suitable. in that case i simply had tomatoes on my eyes. also feel free to move this thread, if it is wrong in this place.

Edited June 15, 2022 by spidobot
solved

[Squid]

You can uninstall each container one at a time and then reinstall via Apps and set them back up again.

 

Taking a note of what the paths / port mappings are on the docker tab (advanced view) will make the process easier.

[spidobot]

ok, i'll install a second instance of each docker and take that config for the old one. (want to be shure to not loose data from a database within a docker through deleting that docker and reinstalling it)

 

thanks, i'll try this weekend and report, if it worked, or if i encounter any issues. just in case, anyone has the same problem at some time

 

thanks to you for your quick reply

[trurl]

6 minutes ago, spidobot said:

install a second instance of each docker

You don't want a second instance, you want a new instance after deleting the old. As long as the new instance is setup the same, its appdata should work just as before.

[meat]

I had the same problem after upgrading to 6.10.2    Some of the dockers wouldn't start, some would start but I couldn't update or edit, and if I tried to uninstall / reinstall it would fail.  I downgraded back to 6.9.2 and so far everything seems to be working fine.  

[spidobot]

Update:

 

I tried to deleting an old docker (gameserver, which isn't used, so if there goes anything wrong, it wouldn't be bad)

and setting up a new instance of it with the same configuration. (as suggested by squid and trurl)

-> all previous data (inside of /mnt/user/appdata/satisfactory/) on the array from the old instance got deleted. (just some test-gamesaves, so nothing bad)

 

so what did work for me?

0. stop "main"/"first" container

1. setting up a secondary instance with exact same configuration

1.a but with a twist: for a link to the filesystem, i just added a "2" (just so the name is different for the directory)

2. install

3. let container run for a bit to set itself up (optional, just wanted to be sure the container can install itself once)

4. stop secondary container

5. edit it's config to the "first" container (see 1.a: now all paths should be the same)

6. start up secondary container

7. check data/webui/gameserver/whateverYourContainerIs and delete "first" container

8. delete the directory from 1.a

 

this list could be shorter i know, but i was willing to do the extra steps, just to be rrreeeaaallllyyyy sure, since the initial test with the gameserver showed, that some docker installations overwrite the directory, if there already exists one.

 

same counts for dockers, which need a db-docker. on their first installation, they "migrate" (but simply overwrite and clear) an existing db.

happened with "wekan" for me. so above steps should also work for database connections, but i didn't test it.

 

maybe it helps somebody else.

 

thanks all for your time and inputs. they helped a lot!

Edited June 14, 2022 by spidobot