We're having a rather strange issue, and i'm hoping someone might have some insight.
We have 2 sites, each with it's own 4404 WLC. Each site is supposed to act as a backup to the other. We would like to use HREAP, so in the event that a site's controller goes down, the AP's can failover to the other but still provide local switching.
Whenever we enable HREAP on the AP's, clients are no longer able to access resources. Traffic simply does not work. For instance, accessing google, cnn, yahoo, etc. doesn't work. Pinging local equipement does work, however. All WLAN's are HREAP local switching enabled. All AP's are still associated to the local controller.
The WLC and the AP's are on the same subnet. They are connected via industrial ro-line switches. What's odd is that if we replace the ro-line switch with a Cisco/linksys switch, everything works fine. If the AP's are in Local mode, everything works fine. Clients wired to the ro-line switches work fine.
Just to test, we've tried a spare HP procurve and a spare D-Link switch, and these have the same problem as well. Only the cisco/linksys switches seem to work. However, due to cost, we can't go around and replace all of the LAN switches.
We're using HREAP at our other sites, which all have cisco switching, and everything is working fine there.