Applications Access is Broken

šŸ›Žļø Get Help
1

This pertains to a C8 Hub running platform 2.4.3.123

When I try to go to the app page when connected directly to the Hubitat device (via the web interface) the app page hangs. Also I cannot reconnect the Amazon Alexa integration because authentication fails.

Currently existing integrations (i.e HomeKit) still work.

A soft reset and restore didnā€™t resolve the problem. A firmware update didnā€™t resolve the problem.

The logs reveal the following errors:

app:35 2025-09-24 03:10:26.883 PM error java.lang.NullPointerException: null (run)

app:170 2025-09-24 03:10:13.039 PM error java.lang.NullPointerException: null (refreshBridge)

Can anyone suggest a resolution short of a full reset of the device?

2

Try
http://:8080/installedapp/status/35
If that pages loads, click disable and disable http://:8080/installedapp/status/170
too..

If that works does the app page now load?

1 Like
3

Maybe try a rebuild database on reboot? I'm not sure if that will help, but it may be something to try.

4

Thanks! It didnā€™t quite work, but returned an interesting result. What I get is a blank page but no ā€˜hangā€™. So unfortunately I canā€™t disable anything. Still, itā€™s a different result than I had been able to achieve. Does this suggest any further action on my part to you?

5

Did you replace the "35" with the actual DNI of you Amazon Echo App? It should bring up the info page for the app, which is the same as pressing the circle "i" in the app list page last column.

6

I just now tried that. No luck. On the bright side I didnā€™t need to perform the soft reset and restoreā€¦

7

How would I determine the DNI for Amazon Echo app? Iā€™m up to try anything at this point.

8

if you cant disable the problematic apps, i think a soft reset and restore is your best bet. a shutdown and unplug for 5-15min might helpā€¦ but my money is on the soft reset.

9

Check the "Show App ID" from the setting gear on the Apps list page. It becomes the first column.

10

the app id is also in the posted logs, that why i recommend the urls that i did

11

Iā€™ve tried the soft reset. With the exception of powering it off for an extended period of time. Iā€™ll try your suggestion incorporating the power interruption step shortly.

I may be shooting in the dark here, as I donā€™t know enough about the OS. But to me it appears that perhaps I have a corrupted kernel?

12

The apps list page is what wonā€™t load. I just get the endless circling icon. Just for grins I left it on that page overnight to see if it would finally load. And of course it did notā€¦

13

What I will try right now is performing the soft reset and then power it off for 10 minutes or so before performing the restore.

14

i think i missed where you posted that you had already soft reset and restored, you may need to restore from an older backup.

Also on the diag page you can try to boot into safe mode. Iā€™ve never done it, but i assume you can try to disable the 2 apps in safe mode. if there is more going on the the logs than those 2 apps, sharing more of the logs would help.

If youre willing to wait till the morning @bobbyD should be on shift. Heā€™ll need your hub id to look into the issue further.

1 Like
15

Have you tried loading the page with the browser's developer console open to see if any specific errors are reported?

1 Like
16

Have you tried starting in Safe Mode from Diagnostic Tool, also, send a message to @support_team along with your hub id.

17

I appreciate your help and suggestions!

I have made some further observations:

  1. Itā€™s not a kernel issue. If I go to the apps page after the soft reset, but before performing the restore, the apps page shows up just fine.
  2. Using the safe mode after the restore doesnā€™t help. Same hang on the apps page.

So itā€™s a database corruption issue. Iā€™ll try and locate a backup made before the corruption issue occurred. Failing that Iā€™ll install the LZF library for Python and take a look-see at the database backup(s). In fact Iā€™ll do that no matter what just out of curiosity.

Iā€™ll report back on my result(s) in any caseā€¦

18

I managed to find an older backup of the database, restored it, and things worked. Well, to a degree. I played hell with restoring some of the integrations that had been implemented since that backup. But now it's all good, and I am running just fine. And you can believe it that I took a backup right now when everything is working!

I have not taken the chance yet to look at the backup(s) with Python, but I will do that when I have a chance.

Thanks again for yours and everyone else's help!

19

Iā€™m no Linux wiz but the hub platform runs in a Java virtual machine so Iā€™m not sure thatā€™s possible.

I believe the backups are encrypted.

1 Like
20

To help with future troubleshooting in general, if you are accessing a web server on any computing device (see application layer) it will be highly unlikely to be a kernel issue as you are already much higher in the compute stack. Itā€™s pretty safe to assume that if you are getting an IP address the kernel is probably fine regardless of compute platform.

Glad to hear a previous backup got you going, it never hurts to get an extra backup especially when youā€™re making major changes. I do 2 a day, 1 local (DLā€™ed to a mac and archived via timemachine) and 1 cloud and itā€™s only over kill until it saves hrs of work and creativityā€¦. once :stuck_out_tongue_winking_eye:

IMG_1140
IMG_11401652Ɨ1376 161 KB