Hostapd suddenly stopped working. Do you have any ideas how to fix it?

Dear IPFire community,

I have installed IPFire about a year ago. Interface blue was configured to serve as wireless access point - i.e. it provided the WIFI network. Interface red is connected to the internet and interface green was unused. Everything was working well until yesterday evening.

Today in the morning, all devices lost their network connection. Please note that no configuration changes were made and that no updates were installed. (Well, at least not that I know of.) Taking a look at IPFire revealed the following during boot

Starting ACPI daemon… [ OK ]
Could not find interface with address blue0 for wireless access point [ FAIL ]
FAILURE:
You should not be reading this error message.
It means that an unforseen error took place in /etc/rc.d/rc3.d/S18hostapd, which exited with a return value of 1
If you re able to track this error down a bug in one of the files provided by IPFire, please be so kind to inform us at…
Press Enter to continue or wait a minute…
Adding IPv4 address 10.10.0.1 to the interface…
Command line is not complete. Try option “help” [ Fail ]
Bringing up the blue0 interface…
Adding IPv4 address 10.10.0.1 to the blue0 interface [ OK ]
Bringing up the red0 interface…
Starting dhcpcd on the red0 interface… [ OK ]
DCHP assigned Settings for red0:
[ip address, hostname mask gateway dns are shown here]
RTNETLINK answers: No such file or directory
Adding static routes… [ OK ]

It seems like, all out of a sudden, hostapd fails to initialize. I assume it could be related to hostapd being started before interface blue0 is available. Could that be the case? How would you debug and fix this issue?

Right now the main concern is to get IPFire up and running again. However, there is another detail Id like to discuss. Its related to updating IPFire

After installing IPFire about a year ago, all updates were installed through the web interface. The system got rebootet after each update and I’ve verified that the updated version was displayed at the web interface. This led me to the assumption that version 155 was running yesterday evening. However, after running “pakfire upgrade” today, it did not just upgrade from 155 to 156, it upgraded from 146 to 156, making it look like none of the updates done through the web interface got installed. This appears to be either a bug at the web interface update, or the system somehow managed to revert itself back to version 145 (which I assume was the version installed about a year ago). This is a rather strange behavior. Has anybody experienced something similar? Or do you have any ideas what could have caused it?

Your help is very much appreciated,
Thank you.

1 Like

Same problem :frowning:

You are not alone. It seems hostapd is broken since 156 update.

Hi all,
this problem exists since Core 154.
Follow this link and perform the steps suggested by Wolf B in comment 7.
Before doing this you must delete hostapd from your instance. After performing these steps you have to complete your new hostapd configuration according comment 12. When all these done, restart your firewall. When checking the services you may have to start your new hostapd manually again.
As long as it does not work properly perform above steps in every new Core upgrade. :smile:

Cheers, Pierre

1 Like

Hi all,

it seems that hostapd do not allow 00 with the new versions and the double entries has been fixed with Core 156 → 12585 – Update with core 154 stops hostapd on some hardware. .

Best,

Erik

1 Like