Wireless Access

Anyone had experience with dense environments causing packet loss? The installation has been over specified and I am experiencing packet loss on ping from the station to an internal network resource. My workstation is connected to the same switch as the AP and experiences no packet loss. The route the AP takes to the controller to tunnel is actually shorter, but on the same path add my workstation. This less me to believe it is unlikely to be the supporting wired instructure. I plan on taking advantage of some of the automatic ARM signal strength adjustments, which do not appear to be in use at present. Anyone aware of recommendations or experience?

This will certainly happen in a wireless environment if there is too much contention.  Ethernet does not have hub-like contention like wireless does.  The best thing that you can do is drop broadcasts at the virtual AP level.  You can also try to turn on ARM mode-aware if you feel you have too many access points in the area to decrease interference.  Lastly, you could have external interference that could be causing the same issue.  It alll depends on the percentage of dropped packets.  If 1, 2 or 3 out of 100 packets is being dropped, that is not bad for wireless.

Hi cj,

 

Apologies if I am being slow here but if i drop broadcasts at the virtual AP level wont that have a connectivtiy effect on the wireless stations?  Or is this common practive with aruba?  Sorry just never come across this before

 

thanks

only if you have applications that depend on broadcastSTo function

I guess dhcp would still be permitted right? Or is the direction of the broadcast blocking from the wired network to the wireless?

Hi everyone, I have the same problem, this issue comes after i upgrade to 6.1.3.0 version, I have the next config, but i still have the same problem 

VAP---Drop Broadcast and Multicast

VAP---Convert Broadcast ARP requests to unicast

VAP---Band Steering -----prefer-5GHz

Mode Aware Arm

QoS ---- Traffic Managment profile----Mode Station Shaping Policy

 

Any other idea??

Dhcp would be permitted... yes.

You need to drop broadcasts in all virtual aps for maximum effect. You do not need mode aware arm

I only noticed this after we had consultants come I and installs the system and since I have started configuring dot 1x for internal access and noticed it was a bit Kathy, then I noticed the loss. Can I ask what was your previous known good working version? I am perfectly prepared downgrade the code to get a result.also I have a couple of 650a I can test this on

soapdish. my previos version was 6.1.0.0 and it works well, but I need to upgraded cause i have an "Web authentication disable" problem in that version.

@ness were you using the native captive portal or amigo pod?

 @tennisball...

I am using the native captive portal

Sorry my predictive text is a bit wrong. I will certainly act on info from cj but can I ask what software you were using before your upgrade?

Soapdish I did not upgrade. That is just a general recommendation.

Sorry cj I came across very badly. Ness said he had these issues only after he updated the software. I was interested in the version he was using prior to the upgrade so I could replicate this in a lab. So this question was directed at ness, and I mentioned I would definitely be following your advise regarding broadcast dropping. I am very sorry for the confusion.

@ness what kind of station was having a problem was it Mac or windows or doesn't matter? I am experiencing much worse loss on apple... Not sure why yet

---

@soapdish wrote:
@Ness what kind of station was having a problem was it Mac or windows or doesn't matter? I am experiencing much worse loss on apple... Not sure why yet

---

Soapdish,

 

This seems to be a bad problem.  I want to advise that you open a TAC case in parallel, so that they can work on your individual issue at the same time.  It could take awhile to find out what is going on here in the forum.