OPNsense Interface No Longer Accessible After 26.1.8 Update

[nepopnuser]

Hello. After upgrading from 26.1.6_2 to 26.1.8 I can no longer access the web interface. Not sure if this is related to the other "access" issues posted.

The console works fine and when try to use "Assign interface" it says the following.

Code Select Expand

Fatal error: Uncaught Error: Call to undefined function confgig_read_array() in /usr/local/etc/inc/console.inc:150
Stack trace:
#0 /usr/local/etc/inc/console.inc(81): _set_networking_interfaces_ports(false)
#1 /usr/local/opnsense/scripts/shell/setports.php(37): set_networking_interfaces_ports()
#2 {main}
  thrown in /usr/local/etc/inc/console.inc on line 150

We are running OPNsense on an older Dell just as a VPN appliance. It was configured and working with one NIC and the firewall completely disabled. We are relying on our existing firewall to protect this connection.

Please let me know if you need any other details. Thanks.

[franco]

My mistake, see https://github.com/opnsense/core/commit/1aa4254289c

It will be hotfixed in a few minutes.


Cheers,
Franco

[nepopnuser]

Since the interface is inaccessible and this is a hotfix, how do you recommend that I fix this?

Should I reinstall and then reapply the config? Or can I somehow overwrite the system file that contains those lines?

[nepopnuser]

I updated the "/usr/local/etc/inc/console.inc" file and then restarted, but the appliance is still inaccessible.

It should be stated that this change fixed the error in the original post but the unit is still down.

I ran through assigning an interface and setting the IP all over again but the issue remains.

[franco]

I'm not sure what went wrong for you. You simply stumbled over the syntax error in recovery but it was not the cause of your problems.


Cheers,
Franco

[nepopnuser]

I've been messing around with it and had gotten back into the web interface by adding a default route. Thinking I should get back to a "good" state, I restored the prior config. After the restart I'm locked out of it again. Adding the default route again says it already exists.

Based on other recent posts, it seems like there might be underlying problems with NAT and/or NUT. Not sure how that applies to this situation. As you said, the typo was not the cause of my issue but the update itself is.

After more tinkering I was able to get back into the web interface. This was done by selecting DHCP for the LAN. In the web interface I noticed that "IPv4 gateway rules" says "Select a gateway from the list to reply the incoming packets to the proper next hop on their way back and apply source NAT when configured. This is typically disabled for LAN type interfaces." As mentioned before, the appliance is only set up with LAN. It does not have WAN. This seems to explain why access was broken, but it doesn't explain why the update broke it.

It had been working fine for the last week (brand new install) and it stopped with the update. How do we determine what the problem is?