I updated a handful of nodes the other day. Upon a brief loss of WiFi this morning, it appears all of the nodes went into fallback mode. I can see their SSIDs, but when I try to connect, they are asking for a 4-8 digit PIN or a password.
I cannot connect to any of the nodes. My WiFi client attempts to connect, it appears to accept the "allstarlink" password, but then fails with "couldn't connect to network".
Anyone know what the PIN is, or how it is derived? Maybe that will work?
Trying to avoid having to go connect all of them to wired connections to recover.
I simply rebooted all of the nodes and they reconnected to their respective WiFi networks. But still curious as to why I could not connect to the fallback access point SSID.
The next step will be to go back into all of the nodes and disable the fallback service again.
ensure that you have administrative (not "limited") access
visit the Cockpit WiFi manager and update the list of WiFi networks. This might involve adding a new WiFi network, updating the information for an existing WiFi network, or re-ordering the list of networks already present (in theory, the server/node should attempt to join the WiFi networks in the displayed order)
The fallback-ap "should" allocate a 192.168.187.64/29 address. If you are not getting an IP in that block then there's something off with the configuration in that system ... and I will defer to @N8EI on any next steps.
If you can manage logging into your server/node then it would be good to know what specific versions of the asl3-appliance packages are installed. If possible, please capture the output of the sudo asl-show-version command.
As I stated in another thread, I had the problem where I was able to connect to the access point, but it didn't assign me an IP address. I guess there is more than one of us with that issue.
Instead, my phone used APIPA, 169.254.X.X, which is not useful. I didn't try manually assigning myself an address in the 192.187.64./29 range.
I'm not even sure how the systemd timer got turned on again. I disabled it a while back when this happened the first time.
The fallback-ap "should" allocate a 192.168.187.64/29 address. If you are not getting an IP in that block then there's something off with the configuration in that system ... and I will defer to @N8EI on any next steps.
I thought I got pushed to beta software again - The same problem is happening again - Node drops off the network to never look back or even consider retrying. Apparently giving its own hot spot of 192.168.187.65. During problem I would shrug and just reboot the node a handful of times during the last week or two - I thought today "I'll just run updates", It's been a while... Well, yes, 50 something updates. Here we go - it wiped out my custom MOTD (yes I have backup) But having to re implement... Then on the latest s/w, just today, it happened again. Node dropped off network and threw up its own hotspot.. why my node is losing its network connection again? I am sure it loses WiFI all the time, but it's been smart enough to connect back (the way it should be) Well, yes, seems it's the lovely fallback Wi-fi again. (Ignore the video running behind me of Office Space movie scene - baseball bats vs copy machine)
This is my solution - may it be vetted / confirmed by others if possible:
I've looked into this. The latest version of asl3-appliance-pi is, in fact, enabling the Fallback AP feature on new installs or upgrades of the package. This was not intentional but an overlooked facet of the package. At this point, the cat is out of the proverbial bag so I'm working on the documentation for it. The upcoming major UI update will make this much easier to manage.