Wireless Access

Hi Everyone,

I setup an Aruba controller (ArubaOS (MODEL: Aruba3400-US), Version 6.4.2.0) with AP's 135.  My internet connection is a cable modem with a local provider.  After it was configured the Internet and wireless was running good for a day.  Now the Internet drops 5-10 times a day.  I cannot ping out to Internet during this time from the devices or the controller.  The only way to resolve this is to shutdown and no shutdown the int gi1/1 port which is the port connected to the cable modem.  Any thoughts I would appreciate it.  Thank you

Below is what I configured.  I did notice that I configured #ip default-gateway import dhcp but I do not see this in the running config.

interface gigabitethernet 1/0
description "GE1/0"
trusted
trusted vlan 1-4094
switchport mode trunk
switchport access vlan 7
!

interface gigabitethernet 1/1
description "GE1/1"
trusted
trusted vlan 1-4094
switchport access vlan 77

interface vlan 7
ip address 10.100.200.1 255.255.255.0
ip nat inside
!

interface vlan 77
ip address dhcp-client

ip dhcp excluded-address 10.100.200.1 10.100.200.10
ip dhcp pool local
default-router 10.100.200.1
dns-server import
domain-name test.com
lease 1 0 0 0
network 10.100.200.0 255.255.255.0
authoritative

(CKCTR01) # show ip route

Codes: C - connected, O - OSPF, R - RIP, S - static
M - mgmt, U - route usable, * - candidate default, V - RAPNG VPN

Gateway of last resort is Imported from DHCP to network 0.0.0.0 at cost 10
Gateway of last resort is Imported from CELL to network 0.0.0.0 at cost 10
Gateway of last resort is Imported from PPPOE to network 0.0.0.0 at cost 10
S* 0.0.0.0/0 [10/0] via 208.104.96.1*
C 10.100.200.0/24 is directly connected, VLAN7
C 208.104.96.0/20 is directly connected, VLAN77

Everything at and beind the 3400 loses internet? Can you still ping the upstream gateway (should be the next hop in the show ip route info)? It sounds like maybe your cable modem is forcing a DHCP refresh more often than your lease time, hard to tell without better logs to capture one of these events. Downloading the logs during the outage and opening a TAC case to have them look at the logs would be the best first step. But making sure from the controller you cannot ping the upstream gateway of your ISP during these outages would help celar up if it's something on the WAN interface or something further upstream. If you can ping your ISP's gateway just upstream from your cable modembut not the internet, it's something else not likely related to your LAN.

The 3400 and everything behind it loses the Internet.  I'll try pinging the ISP gateway when it loses its connection again.  Do you think (lease 1 0 0 0) needs to be adjusted under the dhcp scope?  Thanks

Your lease config is for the internal DHCP server. Your ISP's cable modem controls the lease time for devices pulling an IP address from the cable modem. No it wouldn't help. But downing the interface and then pulling it back up would force a DHCP renew to be issued from the controller, which is why I would lean towards a DHCP issue between the controller and the ISP/cable modem. It may be something else, so finding out if you can ping your cable modem when you are 'down' is a good indicator of whether it's DHCP or not.

I just had the outage occur.  I tried pinging the ISP default gateway and I could not get to it.

So pull the logs (Maintenance > Copy Logs > Download Logs and Include Tech Support) and send that to TAC. If you were trying to ping the WAN gateway (208.104.96.1 which I would assume is your WAN's gateway IP), and you were trying from the controller CLI or the Ping window in the GUI, then either your cable modem is dropping L2 for some reason (maybe they have a goofy L2 DHCP discovery thing, who knows), or the controller is seeing something on the WAN's VLAN that is causing issues. 

There is no other connection on VLAN 77 correct? It's just Gig1/1 connected directly into the cable modem, no switch no anything? Duplicate IPs can kind of do that behavior, though it usually happens much more frequently to lose L2.