|
I have seen access points malfunction in this manner for three reasons:
- Some APs have an option to automatically disable themselves when they sense they have lost wired network connectivity. This feature is intended to stop clients from connecting to an AP with a strong wireless signal but no upstream connectivity in multi-AP WLANs. I suggest plugging the AP, its normal Ethernet uplink, and a LAN analyzer into a hub. Use the analyzer to monitor packets to see what happens just prior to the AP losing signal. If you don't see anything, try running a continuous ping (ping –t) against the AP until signal is lost. (I'm assuming your AP does not have a serial port through which you could simply check its log to see what's happening.)
- Another less-likely possibility is that your AP is experiencing a periodic denial of service (DoS) attack that causes it to freeze. The DoS attack could be intentional -- for example, an attacker sending random connection requests to cause the AP's association table to overflow. Or it could be accidental, like a nearby source of RF interference that overwhelms your AP's own signal. 802.11 DoS attacks can be seen by monitoring the air with a WLAN analyzer like Wireshark. Non-802.11 interference can be diagnosed using a wireless spectrum analyzer like Wi-Spy.
- Alternatively, many APs exhibit this kind of behavior when they experience a firmware bug or hardware failure like overheating. A quick fix for corrupted firmware is often to reinstall the latest and greatest firmware update. Unfortunately, hardware failures usually require AP replacement -- I have seen more than a few APs turned into doorstops for this reason.
 To learn more wireless troubleshooting tips, view SearchNetworking.com's wireless troubleshooting learning guide.
|
