Controllerless Networks

I am new to Aruba so some of this maybe easy. I have a small wifi network configuration setup, but it is giving me big headaches. There is a 205H and a 205 on the system. The 205 has been acting erratically. I noticed looking at the system through inssider that there were three MAC addresses being broadcast. Two for the 205H, one for each spectrum, but only one for the 205 which was handling both spectrum's is this normal? To me it looked as though it was continously switching back and forth. The second issue is that when I have the two AP's running all of the client's attach themselves to the 205H, even though the 205 is much closer. The final issue I am having is that both AP's drop off the network at different times. Is there an issue with having these two types of AP's on the same system?

#AP205

How many SSIDs are you broadcasting?

Are these IAPs part of a Cluster ?

What's the current Min/Max power(EIRP) under the RF settings set to ?

I am only broadcasting one SSID.

 

They are the only two IAP's on the system.

 

Min is set to 18

 

Max is set to Max

Do you have both wired and wireless clients on the same VLAN?

 

Yes they are on the same VLAN

Okay:

 

Edit your WLAN and under Advanced, Turn Broadcast filtering to ARP.  That will drop wasteful and performance impacting broadcasts that wired clients emanate as well as do proxy ARP for your wireless clients:

 

Next, Change your ARM minimum transmit to 12 and Max Transmit to 18 to prevent Sticky clients from holding on to access points when they are out of range.  This will improve performance by putting a client nearest to the AP it should be on.

 

If you have 4 APs or more, you should make sure that 80mhz support is set to disabled.

 

ARP is already set to on

 

I changed the Min and Max to 12 and 18

 

80mhz was already disabled

 

When I arrived onsite today both AP's were down

the 205H had red lights on the power and connection LED

the 205 was green on power, Amber on connection and nothing on broadcasting

Let's take a step back:

 

Did this ever work?  When did it stop working and what happened?

I can not say it has ever worked consistently. I have been able to get the network working appropriatly but then the customer would call a few hours later and say his signal strength was down to two bars, that could have been from his computer latching on to the 205H which is across the house and not the closer 205. The bigger issue is why the AP's drop off completely.

Try to SSH into both access points and get the tech support.  SSH in and type "show tech-support".  That is very wide-ranging and would give you and idea about what is happening with the AP and is a good place to start if you don't have a clue what is going on.

I'll try that now. I would like to know why the 205H is broadcasting two mac addresses one for the 2.4mhz and one for the 5mhz. The 205 is broadcasting one mac address for both. Is this normal?

Run the command, "show ap bss-table" on both.  That will show what is being broadcast.  It is possible that the AP is broadcasting on a DFS channel and your device cannot see DFS channels.  The show ap bss-table command will show exactly what it is broadcasting.

I see both channels being broadcast they arejust under the same mac address which on inssider software makes it look like it is constantly shifting between 2.4 and 5mhz

What version of instant code is this?
Can we see the output of show bss-table and an insider screenshot?

Kitchen# show ap bss-table

Aruba AP BSS Table
------------------
bss ess port ip phy type ch/EIRP/max-EIRP cur-cl ap name in-t(s) tot-t
--- --- ---- -- --- ---- ---------------- ------ ------- ------- -----
40:e3:d6:00:ac:50 TP-LINK_2.4GHz_967EFD ?/? 192.168.1.241 a-VHT ap 36+/18/21 0 Kitchen 0 34m:25s
40:e3:d6:00:ac:40 TP-LINK_2.4GHz_967EFD ?/? 192.168.1.241 g-HT ap 1/18/22 0 Kitchen 0 34m:24s

Channel followed by "*" indicates channel selected due to unsupported configured channel.
"Spectrum" followed by "^" indicates Local Spectrum Override in effect.

 

Equipment Closet# show ap bss-table

Aruba AP BSS Table
------------------
bss ess port ip phy type ch/EIRP/max-EIRP cur-cl ap name in-t(s) tot-t
--- --- ---- -- --- ---- ---------------- ------ ------- ------- -----
84:d4:7e:84:37:f0 TP-LINK_2.4GHz_967EFD ?/? 192.168.1.240 a-VHT ap 52+/18/24 0 Equipment Closet 0 43m:31s
84:d4:7e:84:37:e0 TP-LINK_2.4GHz_967EFD ?/? 192.168.1.240 g-HT ap 6/18/22 4 Equipment Closet 0 43m:31s

Channel followed by "*" indicates channel selected due to unsupported configured channel.
"Spectrum" followed by "^" indicates Local Spectrum Override in effect.

Num APs:2
Num Associations:4
Equipment Closet#

 

 

Instant Version:6.4.3.0-4.2.0.0_49453

That looks fine. Channel 52 is typically a dfs channel. What regulatory domain is configured on your IAPs?

Where would I find that info?

Country Code US

Is there a way to see a log of why the 205H went down in SSH?

Is there an issue with having a 205H and and a 205 acting as a the same network? I wouldn't think so, but I'm looking for anything that can tell me what's going on?

---

jbaker@thesoundroom.com wrote:

Is there an issue with having a 205H and and a 205 acting as a the same network? I wouldn't think so, but I'm looking for anything that can tell me what's going on?

---

There should be no issue, as long as they are on the same version of code.

2 things.

 

1.  You can use "show log system" on both APs to see possibly why it went down.

2.  DFS channels are being allowed/used, which means that the AP would switch channels if it "sees" radar on that channel.  I would uncheck all DFS channels so that you can avoid that.  Many clients also still do not scan for DFS channels, so if you have just two APs, it would make that radio unavailable on any client that cannot see DFS.  Do this to eliminate DFS channels:

 

Go to RF>Show Advanced> Checkbox in "Customize Valid Channels".  Move channels 52+, 60+,100+,108+,132+ and 140+ to the left and click OK:

 

 

See if that fixes your issue.  You could be having radar events.  Type "show ap arm history" to see if that is the case.

 

Equipment Closet# show ap arm history

Interface :wifi0
ARM History
-----------
Time of Change Old Channel New Channel Old Power New Power Reason
-------------- ----------- ----------- --------- --------- ------
2015-11-04 09:17:54 52+ 52+ 24 18 P-
2015-11-04 09:14:10 108+ 108+ Max 24 P-
Interface :wifi1
ARM History
-----------
Time of Change Old Channel New Channel Old Power New Power Reason
-------------- ----------- ----------- --------- --------- ------
2015-11-04 09:17:56 6 6 24 18 P-
2015-11-04 09:14:11 1 1 Max 24 P-
I: Interference, R: Radar detection, N: Noise exceeded, Q: Bad Channel Quality E : Error threshold exceeded, INV: Invalid Channel, G: Rogue AP Containment, M: Em pty Channel, P+: Increase Power, P-: Decrease Power, 40INT: 40MHZ intol detected on 2.4G, NO40INT: 40MHz intol cleared on 2.4G, OFF: Turn off Radio, ON: Turn on Radio

 

 

Kitchen# show ap arm history

Interface :wifi0
ARM History
-----------
Time of Change Old Channel New Channel Old Power New Power Reason
-------------- ----------- ----------- --------- --------- ------
1970-01-01 02:01:09 36+ 36+ Max 18 P-
Interface :wifi1
ARM History
-----------
Time of Change Old Channel New Channel Old Power New Power Reason
-------------- ----------- ----------- --------- --------- ------
1970-01-01 02:35:59 1 11 18 18 M
1970-01-01 02:01:09 1 1 Max 18 P-
I: Interference, R: Radar detection, N: Noise exceeded, Q: Bad Channel Quality E : Error threshold exceeded, INV: Invalid Channel, G: Rogue AP Containment, M: Em pty Channel, P+: Increase Power, P-: Decrease Power, 40INT: 40MHZ intol detected on 2.4G, NO40INT: 40MHz intol cleared on 2.4G, OFF: Turn off Radio, ON: Turn on Radio
Kitchen#

 

 

Look like a power issue??

Interface :wifi0
ARM History
-----------
Time of Change Old Channel New Channel Old Power New Power Reason
-------------- ----------- ----------- --------- --------- ------
1970-01-01 02:01:09 36+ 36+ Max 18 P-
Interface :wifi1
ARM History
-----------
Time of Change Old Channel New Channel Old Power New Power Reason
-------------- ----------- ----------- --------- --------- ------
1970-01-01 02:35:59 1 11 18 18 M
1970-01-01 02:01:09 1 1 Max 18 P-
I: Interference, R: Radar detection, N: Noise exceeded, Q: Bad Channel Quality E : Error threshold exceeded, INV: Invalid Channel, G: Rogue AP Containment, M: Em pty Channel, P+: Increase Power, P-: Decrease Power, 40INT: 40MHZ intol detected on 2.4G, NO40INT: 40MHz intol cleared on 2.4G, OFF: Turn off Radio, ON: Turn on Radio
Kitchen# show log system

Jan 1 02:00:34 syslog: <341005> <ERRS> |AP Kitchen: | open core directory failed
Jan 1 02:00:34 syslog: <341005> <ERRS> |AP Kitchen: | terminal access is enabled but sshd is n ot running, starting
Jan 1 02:00:38 nanny[1712]: <303022> <WARN> |AP Kitchen: | Reboot Reason: AP rebooted Thu Jan 1 00:00:09 PST 1970; Restore AP ENV.
Jan 1 02:00:48 cli[1760]: <341317> <WARN> |AP Kitchen@192.168.1.241 cli| Regulatory table file init at version 1.0_49142, build Mar 19,2015.
Jan 1 02:00:49 cli[1760]: <341004> <WARN> |AP Kitchen@192.168.1.241 cli| Extended ssid enabled
Jan 1 02:00:49 cli[1760]: <341207> <WARN> |AP Kitchen@192.168.1.241 cli| AP support up to 16 S SID.
Jan 1 02:00:49 cli[1760]: <341172> <WARN> |AP Kitchen@192.168.1.241 cli| Find enet0 name eth0.
Jan 1 02:00:49 cli[1760]: <341004> <WARN> |AP Kitchen@192.168.1.241 cli| cli starting, clearin g any dpimgr
Jan 1 02:00:50 cli[1760]: <341004> <WARN> |AP Kitchen@192.168.1.241 cli| cli_choose_eth_uplink :eth current uplink set to DHCP
Jan 1 02:00:50 cli[1760]: <341169> <WARN> |AP Kitchen@192.168.1.241 cli| Add uplink Ethernet 0 .
Jan 1 02:00:50 cli[1760]: <341167> <WARN> |AP Kitchen@192.168.1.241 cli| Uplink eth0 type Ethe rnet, state INIT->LOAD.
Jan 1 02:00:50 cli[1760]: <341169> <WARN> |AP Kitchen@192.168.1.241 cli| Add uplink Wifi-sta 6 .
Jan 1 02:00:50 cli[1760]: <341169> <WARN> |AP Kitchen@192.168.1.241 cli| Add uplink 3G/4G 7.
Jan 1 02:00:50 cli[1760]: <341167> <WARN> |AP Kitchen@192.168.1.241 cli| Uplink type Dummy, s tate INIT->LOAD.
Jan 1 02:00:50 cli[1760]: <341004> <WARN> |AP Kitchen@192.168.1.241 cli| receive sapd hello fo r process recovery
Jan 1 02:00:51 cli[1760]: <341174> <WARN> |AP Kitchen@192.168.1.241 cli| No current uplink, pi ck the highest one - Ethernet eth0.
Jan 1 02:00:51 cli[1760]: <341263> <WARN> |AP Kitchen@192.168.1.241 cli| enable uplink eth0.
Jan 1 02:00:51 cli[1760]: <341167> <WARN> |AP Kitchen@192.168.1.241 cli| Uplink eth0 type Ethe rnet, state LOAD->PROBE.
Jan 1 02:00:51 cli[1760]: <341265> <WARN> |AP Kitchen@192.168.1.241 cli| enable ethernet uplin k eth0.
Jan 1 02:00:51 cli[1760]: <341004> <WARN> |AP Kitchen@192.168.1.241 cli| the ethernet uplink i nuse in first time, get ip address.
Jan 1 02:00:51 cli[1760]: <341167> <WARN> |AP Kitchen@192.168.1.241 cli| Uplink eth0 type Ethe rnet, state PROBE->UP.
Jan 1 02:00:51 cli[1760]: <341185> <WARN> |AP Kitchen@192.168.1.241 cli| Retrieving ip address from br0, ip 192.168.1.241, mask 255.255.255.0.
Jan 1 02:00:51 cli[1760]: <341274> <WARN> |AP Kitchen@192.168.1.241 cli| Update election ip fr om br0, election ip 192.168.1.241/255.255.255.0.
Jan 1 02:00:51 cli[1760]: <341004> <WARN> |AP Kitchen@192.168.1.241 cli| build_my_ip_address: old ip and new ip same; skipping
Jan 1 02:00:51 cli[1760]: <341006> <CRIT> |AP Kitchen@192.168.1.241 cli| Sending uplink_up msg to sapd
Jan 1 02:00:51 cli[1760]: <341006> <CRIT> |AP Kitchen@192.168.1.241 cli| trap uplink info.
Jan 1 02:00:51 cli[1760]: <341004> <WARN> |AP Kitchen@192.168.1.241 cli| last uplink is null, it's the first uplink, ignore this trap.
Jan 1 02:00:51 cli[1760]: <341173> <WARN> |AP Kitchen@192.168.1.241 cli| Current uplink set to Ethernet, state UP.
Jan 1 02:00:51 cli[1760]: <341135> <WARN> |AP Kitchen@192.168.1.241 cli| Master Changed - new 192.168.1.240 old 0.0.0.0 current swarm state 0.
Jan 1 02:00:52 cli[1760]: <341135> <WARN> |AP Kitchen@192.168.1.241 cli| Master Changed - new 192.168.1.240 old 0.0.0.0 current swarm state 1.
Jan 1 02:00:55 cli[1760]: <341004> <WARN> |AP Kitchen@192.168.1.241 cli| Swarm State Change fr om Found to Image_ok
Jan 1 02:01:00 nanny[1712]: <303073> <ERRS> |AP Kitchen@192.168.1.241 nanny| Process /aruba/bi n/radiusd-term [pid 1880] died: got signal SIGTERM
Jan 1 02:01:00 nanny[1712]: <303079> <ERRS> |AP Kitchen@192.168.1.241 nanny| Restarted process /aruba/bin/radiusd-term, new pid 2061
Jan 1 02:01:00 cli[1760]: <341004> <WARN> |AP Kitchen@192.168.1.241 cli| restart dhcp helper a nd dnsmasq!!!

Jan 1 02:01:01 cli[1760]: <341307> <WARN> |AP Kitchen@192.168.1.241 cli| recv_config_init: rec eive config from master 192.168.1.240, using url 0, auto save disable 0.
Jan 1 02:01:02 cli[1760]: <341004> <WARN> |AP Kitchen@192.168.1.241 cli| Extended ssid enabled
Jan 1 02:01:02 cli[1760]: <341207> <WARN> |AP Kitchen@192.168.1.241 cli| AP support up to 16 S SID.
Jan 1 02:01:02 cli[1760]: <341172> <WARN> |AP Kitchen@192.168.1.241 cli| Find enet0 name eth0.
Jan 1 02:01:02 cli[1760]: <341004> <WARN> |AP Kitchen@192.168.1.241 cli| cli starting, clearin g any dpimgr
Jan 1 02:01:03 cli[1760]: <341205> <WARN> |AP Kitchen@192.168.1.241 cli| Applying enet config : port eth1.
Jan 1 02:01:04 cli[1760]: <341205> <WARN> |AP Kitchen@192.168.1.241 cli| Applying enet config : port eth2.
Jan 1 02:01:05 cli[1760]: <341205> <WARN> |AP Kitchen@192.168.1.241 cli| Applying enet config : port eth3.
Jan 1 02:44:32 cli[1760]: <341271> <ERRS> |AP Kitchen@192.168.1.241 cli| ethernet uplink eth0 link down.
Jan 1 02:44:32 cli[1760]: <341004> <WARN> |AP Kitchen@192.168.1.241 cli| stopping DHCP on bond 0
Jan 1 02:44:32 cli[1760]: <341185> <WARN> |AP Kitchen@192.168.1.241 cli| Retrieving ip address from uplink, ip 192.168.1.241, mask 255.255.255.0.
Jan 1 02:44:32 cli[1760]: <341274> <WARN> |AP Kitchen@192.168.1.241 cli| Update election ip fr om br0, election ip 192.168.1.241/255.255.255.0.
Jan 1 02:44:32 cli[1760]: <341004> <WARN> |AP Kitchen@192.168.1.241 cli| build_my_ip_address: old ip and new ip same; skipping
Jan 1 02:44:32 cli[1760]: <341167> <WARN> |AP Kitchen@192.168.1.241 cli| Uplink eth0 type Ethe rnet, state UP->DOWN.
Jan 1 02:44:32 cli[1760]: <341004> <WARN> |AP Kitchen@192.168.1.241 cli| uplink failure on sla ve
Jan 1 02:44:32 cli[1760]: <341263> <WARN> |AP Kitchen@192.168.1.241 cli| enable uplink .
Jan 1 02:44:32 cli[1760]: <341167> <WARN> |AP Kitchen@192.168.1.241 cli| Uplink type Dummy, s tate LOAD->PROBE.
Jan 1 02:44:32 cli[1760]: <341004> <WARN> |AP Kitchen@192.168.1.241 cli| enable dummy uplink
Jan 1 02:44:32 cli[1760]: <341004> <WARN> |AP Kitchen@192.168.1.241 cli| disable SSID[TP-LINK_ 2.4GHz_967EFD
Jan 1 02:44:33 cli[1760]: <341004> <WARN> |AP Kitchen@192.168.1.241 cli| uplink failure on sla ve
Jan 1 02:44:35 cli[1760]: <341167> <WARN> |AP Kitchen@192.168.1.241 cli| Uplink type Dummy, s tate PROBE->UP.
Jan 1 02:44:35 cli[1760]: <341004> <WARN> |AP Kitchen@192.168.1.241 cli| uplink failure on sla ve
Jan 1 02:44:45 cli[1760]: <341272> <ERRS> |AP Kitchen@192.168.1.241 cli| ethernet uplink eth0 link up.
Jan 1 02:44:45 cli[1760]: <341167> <WARN> |AP Kitchen@192.168.1.241 cli| Uplink eth0 type Ethe rnet, state DOWN->UP.
Jan 1 02:44:45 cli[1760]: <341166> <WARN> |AP Kitchen@192.168.1.241 cli| Get interface br0 ip: 0.0.0.0/0.0.0.0.
Jan 1 02:44:45 cli[1760]: <341005> <ERRS> |AP Kitchen@192.168.1.241 cli| ethernet up without i p address, setup ip now
Jan 1 02:44:45 cli[1760]: <341181> <WARN> |AP Kitchen@192.168.1.241 cli| Uplink eth0, setup ip for uplink - Ethernet.
Jan 1 02:44:45 cli[1760]: <341004> <WARN> |AP Kitchen@192.168.1.241 cli| bring up dhcp client for ethernet uplink
Jan 1 02:44:45 cli[1760]: <341004> <WARN> |AP Kitchen@192.168.1.241 cli| /etc/setup_ip_swarm E thernet 1
Jan 1 02:44:45 cli[1760]: <341166> <WARN> |AP Kitchen@192.168.1.241 cli| Get interface br0 ip: 192.168.1.241/255.255.255.0.
Jan 1 02:44:45 cli[1760]: <341185> <WARN> |AP Kitchen@192.168.1.241 cli| Retrieving ip address from br0, ip 192.168.1.241, mask 255.255.255.0.
Jan 1 02:44:45 cli[1760]: <341274> <WARN> |AP Kitchen@192.168.1.241 cli| Update election ip fr om br0, election ip 192.168.1.241/255.255.255.0.
Jan 1 02:44:45 cli[1760]: <341004> <WARN> |AP Kitchen@192.168.1.241 cli| build_my_ip_address: old ip and new ip same; skipping
Jan 1 02:44:45 cli[1760]: <341006> <CRIT> |AP Kitchen@192.168.1.241 cli| Sending uplink_up msg to sapd
Jan 1 02:44:45 cli[1760]: <341006> <CRIT> |AP Kitchen@192.168.1.241 cli| trap uplink info.

264> <WARN> |AP Kitchen@192.168.1.241 cli| disable uplink .
Jan 1 02:44:45 cli[1760]: <341167> <WARN> |AP Kitchen@192.168.1.241 cli| Uplink type Dummy, s tate UP->DOWN.
Jan 1 02:44:45 cli[1760]: <341004> <WARN> |AP Kitchen@192.168.1.241 cli| disable dummy uplink
Jan 1 02:44:45 cli[1760]: <341004> <WARN> |AP Kitchen@192.168.1.241 cli| enable SSID[TP-LINK_2 .4GHz_967EFD
Kitchen# Kitchen# show log system

The kitchen AP looks like it has no access to the internet, because it does not have the correct time.  Are they both in the same instant cluster?  Which AP is the VC?  Can you see both APs from a single AP in the GUI?

They are both in the same cluster. In the instant GUI I see both pucks now, however the kitchen had dropped off for to days.

 

The virtual controller has shifted but it should be on the 205 equipment room AP

How is the kitchen AP powered?

What kind of switch is it plugged into?

I see the link going up and down...

Who is disabling the uplink?

264> <WARN> |AP Kitchen@192.168.1.241 cli| disable uplink .

No one should be unless it disables when the VC is updating

They are both being powered by a seperate POE injector pd-9001gr/at/ac

How do I make sure the time and date is correct in both AP's

They need access to the internet, so they can hit an NTP server...

Both of AP's have the current time as of now. I will see if the settings we changed have an effect I appreciate the help. I do want to confirm that there is no issue with having a 205H and a 205 on the same cluster though.

There should be no issue.

 

Did you also make the regulatory domain channel changes?

The domain was already set to US. I did go in and remove the channels in RF to not broadcast in the DFS channels.

Yes. That is what I meant.

Can you now see all 4 Mac addresses in insider?

I always could. The issue was on the 205 I was seeing only on mac address which in inssider just kept flipping back and forth from 5mhz to 2.4mhz, which on my time graph looked like noise.