|
||||||||
We had a problem with 1.2b10 with captive portal with no SSL. After approx 4 days of operation it stopped letting people through. No obvious errors, but no captive portal page and all traffic was blocked (apart from MAC and IPs specifically allowed). Turning the captive portal off got traffic flowing again without a reboot. The only log entries around that period are (it failed sometime the evning of:Oct 4): Oct 7 01:36:02 last message repeated 12 times Oct 7 01:35:29 /kernel: arplookup 0.0.0.0 <http://0.0.0.0> failed: host is not on local network Oct 7 00:54:07 last message repeated 21 times Oct 7 00:51:42 last message repeated 57 times Oct 7 00:41:38 /kernel: arplookup 0.0.0.0 <http://0.0.0.0> failed: host is not on local network Oct 5 09:48:03 last message repeated 5 times Oct 5 09:47:55 /kernel: arplookup 0.0.0.0 <http://0.0.0.0> failed: host is not on local network Oct 4 22:04:28 msntp: resetting on error 0.233 > 0.205 Oct 4 22:04:28 msntp: resetting on error 0.233 > 0.205 Oct 3 20:58:24 /kernel: arplookup 0.0.0.0 <http://0.0.0.0> failed: host is not on local network Oct 3 20:19:26 /kernel: arplookup 0.0.0.0 <http://0.0.0.0> failed: host is not on local network We don't have a convenient way of monitoring captive portal failure so we've left it off, but just in case someone else is having the same problem... The portal was created from scratch with 1.2b10 and uses the built-in login database with one user, guest, built into the signin page (that way we get a log entry). This was a Soekris 4801 (128MB RAM). I'll probably get brave, and try 1.2 and see what happens... cheers, michael |