Anyone using wpa_supplicant w/ 2.5?
-
Any of you using wpa_supplicant w/ 2.5 & att for bypass?
I use to have my script run early shell and it would auth just before WAN started and all is good.
On 2.5 this still works except got an odd thing. The WPA service pegs to 100% CPU forever.If I change script to just run at the end of everything this doesn't occur but this greatly delays boot-up because it has to wait for WAN to time out and then DNS to time out (since not auth) before boot continues.
At the moment I just run it twice with the second run killing the first one but this isn't ideal.
I'm just wondering if anyone else has played with yet?
To be fair bypass will be going away with AT&T soon anyway with the upgrades so it may not matter soon.
pre 2.5 everything works fine so guessing wpa is doing something too soon on 2.5 and freaking out (despite it does auth/work).
-
@cybercare I just moved to the 2.5 snapshots to try out wg, and to make sure the pfatt bypass still works.
I found the same thing you did: the bypass works, but wpa_supplicant pegs a CPU core at 100%. Have to kill the process to stop it. Not sure what's changed, but I guess the bandaid is to edit the script to automatically kill the process after authentication completes.
-
Not sure if this is related, but replying here again in case it is - internet dies about once a day (but pfSense still reachable on LAN). This is always in the middle of the night, so I just reboot the router due to lack of time, and that fixes it.
I'm not sure if the authentication is expiring/dropping, and since I killed wpa_supplicant, it is unable to reauth.
If I can't kill wpa_supplicant, but wpa_supplicant is using 100% CPU if I don't - what am I supposed to do?
-
Seeing the same issue with %100 cpu usage on wpa_supplicant on pfsense 2.5 RC
One current workaround I read was to kill wpa_supplicant after boot and rerun it right away. Then the abnormal cpu usage goes away but you still have proper authentication.
I wonder if someone is looking into this.
-
@netnerdy Looks like this isn't completely true. Authentication seems to go away if I kill and restart wpa_supplicant. Instead I tried reruning the whole script (pfatt.sh) after the boot sequence. When I do that, I get a hard crash and the firewall reboots.