Apps Broken after Upgrading to TrueNAS-SCALE-23.10.1

[Ralphshep]

Hello,

I upgraded to TrueNAS-SCALE-23.10.1 yesterday and a couple of apps did not start back up again. Some, that don't hold data and are simple web servers were simple to get back online. However, there are a few that require a bit more configuration that I can't get working.

jasongdove/ersatztv:latest-vaapi
This is an app I installed from scratch. However, the following error appears in the container log:

2024-01-02 14:10:41.166247-05:00---> System.IO.IOException: The configured user limit (128) on the number of inotify instances has been reached, or the per-process limit on the number of open file descriptors has been reached.

Photoprism (from Truecharts)
My initial Photoprism instance won't start because it can't find some of the container images so I rebuilt photoprism and migrated the PVC drives according to this guide. However, something seems to have gone wrong with the MariaDB migration.

2024-01-02 14:33:05 (combined from similar events): Unable to attach or mount volumes: unmounted volumes=[list-0 list-1], unattached volumes=[photoprismtemp kube-api-access-j7kt5 list-0 list-1 shm temp shared storage varlogs]: timed out waiting for the condition

NextCloud (From Truecharts)
With NextCloud it seems that the container images have disappeared.

Failed to pull image "tccr.io/truecharts/postgresql:14.6.0@sha256:101b616f2cd0d4d3c1963ecdffd5ffc5704f202967288e718f7aa2ae526600bc": rpc error: code = NotFound desc = failed to pull and unpack image "tccr.io/truecharts/postgresql@sha256:101b616f2cd0d4d3c1963ecdffd5ffc5704f202967288e718f7aa2ae526600bc": failed to resolve reference "tccr.io/truecharts/postgresql@sha256:101b616f2cd0d4d3c1963ecdffd5ffc5704f202967288e718f7aa2ae526600bc": tccr.io/truecharts/postgresql@sha256:101b616f2cd0d4d3c1963ecdffd5ffc5704f202967288e718f7aa2ae526600bc: not found

 

[LarsR]

For truecharts there was an upstream github bug which was resolved with the latest update from a couple of hours ago. Try a fresh catalogue refresh and you should get an update that should fix it, at least it did for me.

 

[Bear Tan]

Thanks. Refresh the Catalogs do help.

 

[Ralphshep]

@LarsR and @Bear Tan thanks for the recommendations. However, it seems my issues may have been much more than just needing to refresh the catalogs. Instead, I have found it simpler to just reinstall most of these apps. I don't have much data saved in the apps. Most of my important data is connected via host paths.

 

[destate]

My apps also won't start (or enumerate) after upgrading from 23.10.0.1 to 23.10.1.1, but unlike @Ralphshep, I really don't want to reinstall my apps from scratch. There's a big red "!" where the running apps should be, it says "Applications are not running" and it has a button to open settings (which takes me to Kubernetes advanced settings) which are all set to their default settings that match my previous setup. I tried updating the catalogs, but that didn't change anything. Any other ideas?

 

[aekt]

truechart has some major compatibility issues, and i'm not sure why they decided this is a good idea when there's a giant breakchange from stable -> enterprise branch that require data migration.

they have the tendency of only keeping recent releases (i believe approx 12 months?)
they have also removed older image versions that are dependencies of older application images.
you will have to:

1. go to https://github.com/truecharts/containers
2. find the image with exact same hash code, in your case it is sha256:101b616f2cd0d4d3c1963ecdffd5ffc5704f202967288e718f7aa2ae526600bc
3. go to System > Shell
4. elevate to root via sudo su
5. perform docker images pull (for bluefin) ctr images pull (for cobia)
6. tag the pulled image with the same name it is being used, in your case it is tccr.io/truecharts/postgresql:14.6.0@sha256:101b616f2cd0d4d3c1963ecdffd5ffc5704f202967288e718f7aa2ae526600bc
7. then start your app

otherwise once your local image cache is gone (after system update), your app won't be able to start