[solved] Dashboard broken after CORE to Scale migration

[reppard]

I recently migrated from TrueNAS CORE to SCALE. Pools and everything came over as expected and pretty much everything seems to work except the main dashboard. The widgets never populate. If I navigate to reporting I can view metrics for everything. The main dashboard is throwing this javascript error in the console:

Code:

error-handler.service.ts:35 TypeError: Cannot read properties of undefined (reading 'state')
    at resources-usage-store.service.ts:106:68
    at Array.forEach (<anonymous>)
    at resources-usage-store.service.ts:100:40
    at Array.forEach (<anonymous>)
    at Object.next (resources-usage-store.service.ts:79:20)
    at A.subscribe.P (tap.js:17:81)
    at d._next (OperatorSubscriber.js:13:21)
    at d.next (Subscriber.js:31:18)
    at take.js:12:32
    at d._next (OperatorSubscriber.js:13:21)
handleError @ error-handler.service.ts:35
next @ core.mjs:27015
next @ Subscriber.js:91
_next @ Subscriber.js:60
next @ Subscriber.js:31
(anonymous) @ Subject.js:34
d @ errorContext.js:19
next @ Subject.js:27
emit @ core.mjs:22733
(anonymous) @ core.mjs:26275
invoke @ zone.js:368
run @ zone.js:129
runOutsideAngular @ core.mjs:26148
onHandleError @ core.mjs:26275
handleError @ zone.js:372
runTask @ zone.js:176
invokeTask @ zone.js:483
M.invoke @ zone.js:472
C.M.args.<computed> @ zone.js:2347
Show 18 more frames
Show less

 

[sretalla]

Have you forced a refresh of your browser cache for that page?

 

[reppard]

Have you forced a refresh of your browser cache for that page?

Indeed. I've also tried different browsers.

 

[sretalla]

I can confirm that on Cobia (TrueNAS-SCALE-23.10.1) I see items populated in my dashboard and also see data in the reporting page.

It must be something with your system, so either re-install or look further into logs.

 

[reppard]

I can confirm that on Cobia (TrueNAS-SCALE-23.10.1) I see items populated in my dashboard and also see data in the reporting page.

It must be something with your system, so either re-install or look further into logs.

I installed SCALE via the CORE update process. Are you suggesting a fresh install of SCALE? Can I do with while preserving the apps I've already installed and configured?

 

[sretalla]

Take a config backup, install fresh, then restore the config.

As long as you leave the pool disks where they are, it should all come back as before (but with a clean OS copy).

Take care of SSH keys and encryption keys/ passphrases if you have those.

 

[reppard]

Take a config backup, install fresh, then restore the config.

As long as you leave the pool disks where they are, it should all come back as before (but with a clean OS copy).

Take care of SSH keys and encryption keys/ passphrases if you have those.

I gave that a shot. Install and config went fine. Dashboard is still borked. Seems like something in the config carried over from CORE to SCALE making it not happy

 

[sretalla]

Seems like the hot patch released may help you:

TrueNAS SCALE 23.10.1.1 has been released!

We are pleased to release TrueNAS SCALE 23.10.1.1! This is a small hotpatch with fixes for web interface issues discovered after the 23.10.1 release. The full 23.10.2 maintenance release is anticipated later in February. Changes: Fix non-physical interface (Link Aggregation, VLAN, Bridge)...

www.truenas.com

 

[reppard]

Seems like the hot patch released may help you:

TrueNAS SCALE 23.10.1.1 has been released!

We are pleased to release TrueNAS SCALE 23.10.1.1! This is a small hotpatch with fixes for web interface issues discovered after the 23.10.1 release. The full 23.10.2 maintenance release is anticipated later in February. Changes: Fix non-physical interface (Link Aggregation, VLAN, Bridge)...

www.truenas.com

I think I may have found out what was causing the issue. One of the nics in my bond interface wasn't attached and another non-existing interface was in its place. I fixed that and now the dashboard seems to be working. Definitely seems like a bug in the dashboard. The dashboard would actually be a great place to alert on this type of config error rather than just not working.