Web interface unreachable after update

When updating from 2.3.1.140 -> 2.3.3.133 on a c5 I found I was unable to access the web interface at all, even while plugged directly into the same switch. I used the IP:8081 trick to roll back 2 versions and now I'm able to access the admin interface again.

I see a lot of threads about blocking proxy access and needing to permit non-local access, but nothing that matches this "on the same LAN and still cannot get in" behavior. Any pointers?

Normally I'd suggest that somehow your IP address changed but the fact that you were still able to get to port 8081 eliminates that. And the fact that you were able to roll back and successfully get to the UI definitely points to the upgrade as a likely culprit though I haven't heard of that being an issue. What happens if you try the upgrade again, just to see if it works this time? Otherwise I think this might be one for @bobbyd.

1 Like

The latest release has a new database engine and in rare cases, the database fails to update properly. It sounds like your hub is one of those cases. please perform the following steps:

  1. Go to Diagnostic Tool
  2. If you don't have the latest Diagnostic Tool (version 1.0.97) then follow these instructions to download the latest version.
  3. Perform a Soft Reset by following instructions in this document: Soft Reset | Hubitat Documentation
  4. Once the hub is back online, restore from a backup.

NOTE if the hub was unable to install the 2.3.3 platform version, and your hub is still showing that is running an older version, then go to the Diagnostic Tool and select "Download latest version". Once the download completes, click "Restore previous version" and select the latest 2.3.3 version.

I have the mac pinned in the DHCP server and can verify IP address did not change.

I rolled forward and back twice to verify that it is 100% correlated with the update and 100% fixed by the rollback. I am pretty sure I have the version numbers correct, but the first instance of the problem could definitely have been before this one, because I was a few versions behind in updating.

Happy to do troubleshooting if that's useful.

oh, interesting. ok. i'll try this when i can tolerate the downtime/risk. thanks!

1 Like

Well apparently @bobbyd doesn't actually sleep... so there ya go!

3 Likes

Ya know what they say, no rest for the wicked!

2 Likes

Who doesn't check their work emails / notifications first thing in the morning....? Or through the night... or notifications of any kind.... That's normal... isn't it.... :slight_smile:

2 Likes

How can you avoid them when the lights turn red if someone mentions your name?

2 Likes

I'd be careful how many people you make aware of that.... @bobbyD....

@bobbyD ... @bobbyD .... :rofl: :rofl:

2 Likes

@bobbyd

2 Likes

Perhaps that should have been "who" you let know.... :wink:

Say hello to one-eyed Forky, the 1st grade project...

3 Likes

No.

1 Like

Good point, that comes across as me wanting to exclude others..... Have you considered setting up an automation to inform you of the emails and other notifications you missed while asleep or divorced from your mobile device....? :wink:

2 Likes

It does feel like there should be a forky "catch-all" option in RM that turns on a light in BobbyD's place... Would that get traction..... It would certainly help with support of issues with rules..... :wink:

2 Likes

Just to close the loop here (and sorry for delay: was traveling and couldn't do this remotely): This worked, so something messed up the database and prevented 2.3.3.140 from working properly.

The backup+softreset (which then on reboot does an update to 2.3.3.140 anyway) and then restore worked great.

thanks!

3 Likes

This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.