Hello

I have set up an base station with a sector connected to a XR5 mini pci card on a RB433AH board.
Clients connected to this base use RB411A and R52/R52H.

All use ROS 3.24

Problem is that clients disconnect all the time, one and one, not all at the same time.
Also with different inter walls, 3-30min.

All have good signal and LOS.

In log "disconnected, extensive data loss"

Any tips, please.

Interference or lack of signal. Post a screen shot of your associations.

I replaced RB532 with SR5, now I have 4 RB433AH each with 2 XR5 to 90º pannels (yes, they are redundant), more than 10 clients per radio, CM9 most of them.

First RB433AH didn't work right, I had to power it off and forget about it.
On the other 3 RBs, I've been having irregularities. Sometimes client disconnect, or throughput falls to less than 100Kbps (and they all should traffic more than 1 Mbps)
If I change frequency, and then back to original frequency, they start working well again.

Radios are RSTP bridged to ethers, use WDS, RouterOS 3.24
One is NStreme polling wds dynamic mesh
Other are wds dynamic, no NStreme. These behave worst.

We think we had received a bad lot of XR5... we're about to replace them.

Interference or lack of signal. Post a screen shot of your associations.

I think this is much more likely than a bad lot of XR5 cards - they all go thru QC.

Chofex, did you check for interference sources, especially other wireless equipment like cordless phones?

Sorry, I don't have the equipment to do that... all I can do is to show you Freq Usage...
It's under 1 per all frequencies, except:
5560 54,7
5640 10,4
5660 2,8
5680 3,3
5700 8,5

And those are all mine (all XR5, 90º pannels, and yes, there's one extra pannel pointing to a populated area)

Hello.

We too have the exact same setups and the exact same problem !! Thank god someone else is seeing this issue.

Our client are on 411's using R52's running 3.27

The AP is a 433AH with 2 x XR5's connected to Horiz and Vert.

The AP enclosure is a dual polarised 16dBi 90 Deg Panel like this one http://www.itelite.net/products-desc.php?id=389
I am seeing the the P Throughput on the CPE links going from 28Mb-s to 160kb and fluctuating wildly. The links are clear line of sight and less than 4km. Nearly all CPE TX/RX signals are in the -70dbm range. I only have Mikrotik connected the the AP's and all CPE are running ROS 3.27 only. See CPE Registration Details and P Throughput in photo below.
I have also seen very high ACK's on the AP's Registration Table +400us. See photo below.
I have tried turning off the Horz sector but the problem still exists on the vertical. I have also tried changing the channel but this does not fix it either.

This is driving me crazy and I really need to get this fixed. Please contact me directly if you want to team up on solving this issue and comparing notes.. I am ready to throw away the XR5's and go back to R52's but there must be a fix for this.

Bump.. Anyone out there ?

Lock your speeds down to 24 megs.

What is your noise floor? While you may have good SNR, you may still have a high noise floor with two high power radios in the same enclosure with no space diversity between antenna elements. We don't use the dual polarity antennas because there is no space between the elements and you do get cross channel interference. Yes you sometimes can make it work, but why go with sometimes...you want all the time.

The one image shows a good SNR but a bad CCQ, which is indicative of channel congestion. My guess is that you have a wireless frame timing issue where the two radios are sending at exactly the same time, or close enough that they are killing each other on the antenna element (with no space diversity). It appears as lost frame, not as noise to the receiving end. So you have a collision between frames on different frequencies.

A guess, but probably close to what is going on.

You could also have the PA shutting down momentarily due to overload on receive of RF from the radio in the same enclosure. We have seen PA blow due to this type of condition in dual radio systems.

Generally we try to keep everything as far apart as possible in a dual radio system.

Good luck.

Hey everyone,

I figured I would post in this thread being its close to my setup. I have several RB433AH + XR5 AP with Ubiquiti NS5 for clients. My problem is with one RB433AH and 2 XR5, one radio has 15 clients, the other 14 clients. I've noticed lately that when we hit about 7-8Mbps of traffic (According to TheDude) from this RB433AH it seems to bottleneck. I am a newbie to all this and its a pretty down and dirty simple configuration. I do have my data rates locked to a max of 24Mbps for A/G. We sell packages of 1.5mbps all the way up to 6mbps, most customers are at 1.5mbps, but as we all know most business just use it for email and website traffic so they don't really peg that out. But like I said I've noticed when The Dude says we hit 7-8mbps out of this board I notice things slow down for the customers who aren't hogging the bandwidth. Can anyone think of a reason why this would be happening or any suggestions to solve this issue??

p.s. I have a DS3 backhaul which this board is tapped into via ethernet and switches of course.

figured I would post in this thread being its close to my setup. I have several RB433AH + XR5 AP with Ubiquiti NS5 for clients. My problem is with one RB433AH and 2 XR5, one radio has 15 clients, the other 14 clients. I've noticed lately that when we hit about 7-8Mbps of traffic (According to TheDude) from this RB433AH it seems to bottleneck. I am a newbie to all this and its a pretty down and dirty simple configuration. I do have my data rates locked to a max of 24Mbps for A/G. We sell packages of 1.5mbps all the way up to 6mbps, most customers are at 1.5mbps, but as we all know most business just use it for email and website traffic so they don't really peg that out. But like I said I've noticed when The Dude says we hit 7-8mbps out of this board I notice things slow down for the customers who aren't hogging the bandwidth. Can anyone think of a reason why this would be happening or any suggestions to solve this issue??

Actually, 7-8Mbit is a bottleneck for me too.. I think it's pretty normal...

Hello

I have set up an base station with a sector connected to a XR5 mini pci card on a RB433AH board.
Clients connected to this base use RB411A and R52/R52H.

All use ROS 3.24

Problem is that clients disconnect all the time, one and one, not all at the same time.
Also with different inter walls, 3-30min.

All have good signal and LOS.

In log "disconnected, extensive data loss"

Any tips, please.

I encountered the same issue with XR5. I had a batch of RB433AH + 2x XR5.. When both radios were up, clients were randomly disconnection for "extensive data loss" . When i turn off one of the two XR5 those issues are fixed. I'm not a physician, but it's pretty obvious that XR5s were making some interferences between them. It wasn't a problem of power supply, since they are supplied by 5x batteries 100A and the cable is only 4M long.. I tried to lower the tx-power but wasn't helpfull.
I replaced them with SR5 and fixed my issues. And i'm keeping those XR5 for Backhaul links..

figured I would post in this thread being its close to my setup. I have several RB433AH + XR5 AP with Ubiquiti NS5 for clients. My problem is with one RB433AH and 2 XR5, one radio has 15 clients, the other 14 clients. I've noticed lately that when we hit about 7-8Mbps of traffic (According to TheDude) from this RB433AH it seems to bottleneck. I am a newbie to all this and its a pretty down and dirty simple configuration. I do have my data rates locked to a max of 24Mbps for A/G. We sell packages of 1.5mbps all the way up to 6mbps, most customers are at 1.5mbps, but as we all know most business just use it for email and website traffic so they don't really peg that out. But like I said I've noticed when The Dude says we hit 7-8mbps out of this board I notice things slow down for the customers who aren't hogging the bandwidth. Can anyone think of a reason why this would be happening or any suggestions to solve this issue??

Actually, 7-8Mbit is a bottleneck for me too.. I think it's pretty normal...

Hello

I have set up an base station with a sector connected to a XR5 mini pci card on a RB433AH board.
Clients connected to this base use RB411A and R52/R52H.

All use ROS 3.24

Problem is that clients disconnect all the time, one and one, not all at the same time.
Also with different inter walls, 3-30min.

All have good signal and LOS.

In log "disconnected, extensive data loss"

Any tips, please.

I encountered the same issue with XR5. I had a batch of RB433AH + 2x XR5.. When both radios were up, clients were randomly disconnection for "extensive data loss" . When i turn off one of the two XR5 those issues are fixed. I'm not a physician, but it's pretty obvious that XR5s were making some interferences between them. It wasn't a problem of power supply, since they are supplied by 5x batteries 100A and the cable is only 4M long.. I tried to lower the tx-power but wasn't helpfull.
I replaced them with SR5 and fixed my issues. And i'm keeping those XR5 for Backhaul links..

The XR5's have been working very well for me I don't get any disconnects unless I reboot a unit or the customer reboots. I have my XR5's about 20 to 40 mhz apart when it comes to frequency. I just can't for the life of me figure out why when we hit 7mbps it slows down.

What is your noise floor? While you may have good SNR, you may still have a high noise floor with two high power radios in the same enclosure with no space diversity between antenna elements. We don't use the dual polarity antennas because there is no space between the elements and you do get cross channel interference. Yes you sometimes can make it work, but why go with sometimes...you want all the time.

Thanks for all the help guys. I have figured out what the problem was. NOTHING to do with radio cards interfering in the same enclosure. I have not change anything with the Data rate either (all set to default). I installed a second sector on the mast and migrated the clients one at a time until the issue was discovered.

It seems to have been a hidden node issue. I upgraded the ROS wireless package to the wireless-test package and turned on the HW-Protection RTS/CTS on all CPE and on the AP. The problem is gone and all is working really well. All customers are now connected and stable.

I thought RTS/CTS was standard in Mikrotik without having to use the wireless-test package.

This was done in 3.23: (I thought..)

*) wireless - fix for RTS/CTS when used together with dynamic ACK timeout;

I was using 3.25 and this problem existed. I am now using 3.28 on all CPE and AP for this entire site.

I am now running 2 x XR5's in the same enclosure (default power) and my CCQ's are 95%/90% on average.

I would really like to know more about this issue as it has caused me alot of problems and time to figure this out and solve it.

Mikrotik ?? what does the HW protection threshold work and do I need this on the AP as well ?The only thing is that I have noticed a small increase in jitter but the performance is far better and stable now. I can reduce the jitter by adjusting the HW protection threshold. I just don't understand it enough and have kept all CPE's the same while I do further tests.

bump...

Medianet mentioned that 7-8mbps is where things would bottleneck for him as well, has anyone else seen this and been able to resolve this as well. I would hate to have to put up more sectors because we can only use 8mbps of traffic per sector, anyone got suggestions?

ok so i uploaded the wireless test package and i think It seems to have improved, I've got download up to 14mbps overall and no one seems to lag. My issue lately though has been upload, i'll run a speedtest pull 6-8mbps and my upload will be 1-2mbps where it use to be more or less on par with the download. We also have an alvarion link at our office speedtest on that is 9283kbps down 8922kbps up, i've had several customers also complain that the upload is extremely slow even attaching a file to an email seems to take long. I had one customer call this morning, she's set for 2mbps upload, only able to pull 1100kbps, doesn't seem much but when I installed it we got 2048kbps solid each test. We only have 10 customers on that sector, overall Tx CCQ on that sector 94-100% the customer in questions ccq 98/100. Any help would be greatly appreciated, if you need more info let me know, I have people bugging me here at the office.

It sounds like most of the issues are hidden node related. 802.11abg are contention base protocols and as such will not degrade gracefully, that is why you hit your 7mb or so and things grind to a halt. Typically you will hit that wall between 10-20 users (traffic load dependent) There are 2 ways around this, one is to use RTS CTS settings as noted, the other way I would suggest trying is using Nsteme as it is a polling protocol and does not have the hidden node issues. But of course you need an all MT set to use Nstreme.

It would be real interested in comparing RTS/CTS to nstreme in terms of through put and latency.

Bump - me too!!!!

Thank you,
Rod

well as of late the problem hasn't been the bottleneck, its been uploads are very slow. Customers are hitting their down speeds but I got an email from a customer today saying he was only getting 50k when he's paying for 1024k UP. I've messed with rts/cts turned them both ON, 433ah+xr5 and the NS5's and haven't seen a difference when they are enabled or disabled. Here's the config as it pertains to the bridge, wireless if anyone can help it would be greatly appreciated, again using RB433ah with xr5 with NS5 as client subscriber units, RB433ah is running v.3.28 with wireless test package..

/interface bridge
add admin-mac=00:0C:42:29:B9:78 ageing-time=5m arp=enabled auto-mac=no \
comment="" disabled=no forward-delay=15s l2mtu=1526 max-message-age=20s \
mtu=1500 name=bridge1 priority=0x8000 protocol-mode=none \
transmit-hold-count=6
/interface ethernet
set 0 arp=enabled auto-negotiation=yes comment="" disabled=no full-duplex=yes \
l2mtu=1526 mac-address=00:0C:42:29:B9:78 mtu=1500 name=ether1 speed=\
100Mbps
/interface wireless
set 0 ack-timeout=dynamic adaptive-noise-immunity=ap-and-client-mode \
allow-sharedkey=no antenna-gain=0 antenna-mode=ant-a area="" arp=enabled \
band=5ghz basic-rates-a/g=6Mbps,9Mbps,12Mbps,18Mbps burst-time=disabled \
comment="" compression=no country="united states" default-ap-tx-limit=0 \
default-authentication=yes default-client-tx-limit=0 default-forwarding=\
no dfs-mode=none disable-running-check=no disabled=no disconnect-timeout=\
3s frame-lifetime=0 frequency=xxx frequency-mode=manual-txpower \
hide-ssid=yes hw-fragmentation-threshold=disabled hw-protection-mode=\
rts-cts hw-protection-threshold=0 hw-retries=4 l2mtu=2290 mac-address=\
00:15:6D:64:A5:14 max-station-count=150 mode=ap-bridge mtu=1500 name=\
"wlan1 180 sector" noise-floor-threshold=115 on-fail-retry-time=100ms \
periodic-calibration=default periodic-calibration-interval=60 \
preamble-mode=both proprietary-extensions=post-2.9.25 radio-name=\
xxxxxxxxxxx rate-set=configured scan-list=default \
security-profile=default ssid=xxxxxxx station-bridge-clone-mac=\
00:00:00:00:00:00 supported-rates-a/g=6Mbps,9Mbps,12Mbps,18Mbps \
tx-power-mode=default update-stats-interval=disabled wds-cost-range=\
50-150 wds-default-bridge=none wds-default-cost=100 wds-ignore-ssid=no \
wds-mode=disabled wmm-support=enabled
set 1 ack-timeout=dynamic adaptive-noise-immunity=ap-and-client-mode \
allow-sharedkey=no antenna-gain=0 antenna-mode=ant-a area="" arp=enabled \
band=5ghz basic-rates-a/g=6Mbps,9Mbps,12Mbps,18Mbps burst-time=disabled \
comment="" compression=no country="united states" default-ap-tx-limit=0 \
default-authentication=yes default-client-tx-limit=0 default-forwarding=\
no dfs-mode=none disable-running-check=no disabled=no disconnect-timeout=\
3s frame-lifetime=0 frequency=xxxx frequency-mode=manual-txpower \
hide-ssid=yes hw-fragmentation-threshold=disabled hw-protection-mode=\
rts-cts hw-protection-threshold=0 hw-retries=4 l2mtu=2290 mac-address=\
00:15:6D:64:A4:D7 max-station-count=150 mode=ap-bridge mtu=1500 name=\
"wlan2 60 sector" noise-floor-threshold=127 on-fail-retry-time=100ms \
periodic-calibration=disabled periodic-calibration-interval=60 \
preamble-mode=both proprietary-extensions=post-2.9.25 radio-name=\
xxxxxxxxxx rate-set=configured scan-list=default security-profile=\
default ssid=xxxxx station-bridge-clone-mac=00:00:00:00:00:00 \
supported-rates-a/g=6Mbps,9Mbps,12Mbps,18Mbps tx-power-mode=default \
update-stats-interval=disabled wds-cost-range=50-150 wds-default-bridge=\
none wds-default-cost=100 wds-ignore-ssid=no wds-mode=disabled \
wmm-support=enabled
/interface wireless manual-tx-power-table
set "wlan1 180 sector" comment="" manual-tx-powers="1Mbps:24,2Mbps:24,5.5Mbps:\
24,11Mbps:24,6Mbps:24,9Mbps:24,12Mbps:24,18Mbps:24,24Mbps:24,36Mbps:24,48M\
bps:24,54Mbps:24,HT20-0:0,HT20-1:0,HT20-2:0,HT20-3:0,HT20-4:0,HT20-5:0,HT2\
0-6:0,HT20-7:0,HT40-0:0,HT40-1:0,HT40-2:0,HT40-3:0,HT40-4:0,HT40-5:0,HT40-\
6:0,HT40-7:0"
set "wlan2 60 sector" comment="" manual-tx-powers="1Mbps:17,2Mbps:17,5.5Mbps:1\
7,11Mbps:17,6Mbps:17,9Mbps:17,12Mbps:17,18Mbps:17,24Mbps:17,36Mbps:17,48Mb\
ps:17,54Mbps:17,HT20-0:0,HT20-1:0,HT20-2:0,HT20-3:0,HT20-4:0,HT20-5:0,HT20\
-6:0,HT20-7:0,HT40-0:0,HT40-1:0,HT40-2:0,HT40-3:0,HT40-4:0,HT40-5:0,HT40-6\
:0,HT40-7:0"
/interface wireless nstreme
set "wlan1 180 sector" comment="" disable-csma=no enable-nstreme=no \
enable-polling=no framer-limit=4000 framer-policy=none
set "wlan2 60 sector" comment="" disable-csma=yes enable-nstreme=no \
enable-polling=no framer-limit=3200 framer-policy=dynamic-size
/interface bridge port
add bridge=bridge1 comment="" disabled=no edge=auto external-fdb=auto \
horizon=none interface=ether1 path-cost=10 point-to-point=auto priority=\
0x80
add bridge=bridge1 comment="" disabled=no edge=auto external-fdb=auto \
horizon=none interface="wlan1 180 sector" path-cost=10 point-to-point=\
auto priority=0x80
add bridge=bridge1 comment="" disabled=no edge=auto external-fdb=auto \
horizon=none interface="wlan2 60 sector" path-cost=10 point-to-point=auto \
priority=0x80
/interface bridge settings
set use-ip-firewall=yes use-ip-firewall-for-pppoe=no \
use-ip-firewall-for-vlan=no

If down loads are good and up loads suck it's interference either self induced or some other gear near by. Remember that non 802.11 bases signals are not picked up and seen as noise. You can sit right on a Canopy and not see a thing. So take any reading you get from the radio with a caution. A spectrum analyzer might be of some help if you can get your hands on one. An alternative the the little ones from ubiquti, one for 2.4 and one for 5ghz there about $100 or so and plug into your usb port. Have not tried them but looks like they should do the trick in most cases.

Hello mherbst01,

Did you test or have any news about this problems?
Anyone have new info about rts/cts to solve the "hidden nodes" problems?

Thanks,

I have the same setup , but I made a research of the signal interference per channel to pick the best channel because my own base stations was made interference even at 4 K.M as far. The pics are for one base statio of a cluster of 3 with 3 120° Pac Wireless Sectors RB433AH +Ubiquiti XR5 3 Radios one XR5 per RB Pac Wireless Poe. Ping for the Worse client -71 of signal
all clients using Nanostation5 3 K.M max distance from base station.