I have recently deployed a MT hotspot running over quite a large area. It is several apartment blocks in a complex where there is lots of concrete, glass and steel, which has necessitated use of 10 repeater stations running WDS. The Hotspot itself works perfectly - no problems there (running a P3-733 with MT Level 6 license, Hotspot and User Manager). The problems are with the WDS system. Running dynamic WDS just did not work. Using the RSTP Bridge Test package I was lucky to get 128k across the network. I then switched to Static WDS which works a little better - I now average 2 - 3 mbps across the network, but it still is not suitable for the 4mbps of available bandwidth I have (soon to be increased to 8mbps). The system is very unstable though - adding another repeater in the wrong place can easily kill the network - and we often need to add mobile WDS repeaters for really difficult to reach apartments where users have laptops with poor wireless cards. Every time I add anything I have to re-calibrate the whole system.

If I take down say half the network (any half) then I can easily get 5-6mbps across the network. It does not really matter which units - as long as there are less than about 5 running the network is blazing fast. I suspect it is the bridging system re-broadcasting the signal to all units that is causing the problem - but I dont know an effective workaround.

That is without trying to mesh the system - as soon as I add in a redundant WDS link the network slows down hideously. Currently I am sticking to 1 link per repeater only.

Currently all systems are running 2.9.32 and it is reasonably stable (if slow) provided I dont change anything. Hardware is a mix of RB532 and WRAP 2C on the highsites connected to small 10.5dB Sector antennas, and RB112 for the mobile repeaters. Eventually this system will be supporting up to 200 clients simultaneously and running in excess of 25mbps of bandwidth (although no 1 unit will have to handle that amount - it will be spread over several units)

Has anyone else had experience with bridged WDS networks of this size and have you had similar problems? Should I maybe look at more powerfull NLOS antenna's and cut down on the number of highsites? Any comments are appreciated.

Hi David,

If you remember the last course you presented in PTA (still waiting for the certificate btw) I approached you with this very same problem.

I refer you to this topic:
http://forum.mikrotik.com//viewtopic.php?t=10197
Where I discussed the exact same teething problems as you are experiencing now. 5 routers were fine, but any more and it all fell apart.

I have a 10-router mesh network running very well, with between 2 and 5 WDS connections per node. I get 2 Mbps to the furthest node, and about 5-7 Mbps to the node closest to the gateway, running 802.11b/g. The mesh spans an area of about 500m x 300m, using CM9s and single 8dBi omni on each node. With lots of concrete, trees and on a slope; I have laptop coverage over 90% of the area.

The answer is to use connect-list, with either static or dynamic WDS and no earlier than 2.9.30.
2.9.30 connect-list was fixed when I presented my problem to specify a min-signal of -120 on the last no-connect rule. See the Mesh Wiki for the configuration example. This is the key to stable connections, and aim for signals at least in the -70's.
Either STP or RSTP work fine, but obviously RSTP reacts quicker to changes in the network.

Be careful of interference, which will break the mesh easily too. Use snooper to determine if there is any significant traffic on your chosen channels.

I also found that using WPA did not work very well - if a conection was lost, it would sometimes not re-authenticate for some time.

My final configuration consists of bridged WDS interfaces using WEP 128 as a 'backbone' and Virtual AP's running PPPoE via RADIUS.
Hotspot did not seem to be reliable if run off a central interface, but you could run hotspot on every node with RADIUS authentication.

Hey GWISA - yes I remember the PTA course well. However if I remember your problem seemed to be reaching the hotspot across several WDS interfaces - If it transited a WDS node then the hotspot was not reached.

I am not experiencing that particular issue - the hotspot always works. My problem is getting enough bandwidth through the WDS bridged network.

Having re-looked and re-thought the layout I am probably going to go with 3 seperate hotspot nodes running channels 1,6,11 and link them together on a 5.8 backbone. I am reasonably confident this will sort out my problem.

Hey GWISA - yes I remember the PTA course well. However if I remember your problem seemed to be reaching the hotspot across several WDS interfaces - If it transited a WDS node then the hotspot was not reached.

I am not experiencing that particular issue - the hotspot always works. My problem is getting enough bandwidth through the WDS bridged network.

Having re-looked and re-thought the layout I am probably going to go with 3 seperate hotspot nodes running channels 1,6,11 and link them together on a 5.8 backbone. I am reasonably confident this will sort out my problem.

It's not necessary to split the backbone portion.

If you have read the topic I referred you to, you'll see it works 100% with one interface on each node.

Your stability issues are definitely due to the dynamic WDS and marginal signal strengths.

Your solution might increase your throughput, but I suspect you're going to face exactly the same issue on 5GHz - do you need to get 4Meg to the furthest node? I would only split the backbone & AP's if the amount of concurrent clients connecting warrented it (overloading AP's).

Remember the range will be less on 5GHz, so you're going to have to boost your output to get the same throughput you should be getting on 2.4... Of course if 2.4 noise is killing your mesh, then it's another story...

*---How can I contact you? I've misplaced your email... I'd like to find out what happened to the course attendance certificates?---*

Currently I am running static WDS only with a single link per AP - no redundant links. Signal strengths between units are -60 to -70 at worst. I still get issues once too many units (more than about 10) are online. Bandwidth starts dropping to below 1.5mbps

Hi David,

Can you give me an answer on this please?

*---How can I contact you? I've misplaced your email... I'd like to find out what happened to the course attendance certificates?---*

thanks

Sorry - I thought my details were easy to find - it's on the Mikrotik site under trainers for South Africa. My email is d [at] etuan co za

I have mailed that address twice already and got no response... I will try again... thanks

You could be generating too much noise if you have too many APs close together - this will kill bandwidth and still reflect decent signal. Also, keep in mind that every hop you add will reduce your bandwidth by almost half. If you are using b/g mode, try swithing to b-only - we have seen over and over how differently networks operate (outdoors) when in b/g mode - b-only mode always works better (if you can deal with the bandwidth limitations). Adding a different frequency for your backbone will solve a few problems and speed things up alot.

I am aware of the noise problems associated with too many AP's close to each other. This could well be a problem, but there is no other choice as the apartment layout is extremely complex and difficult to project signal everywhere - in fact we will probably need another 2-4 highsites to completely cover everything. Not to mention several large bodies of water separating apartment blocks! That is why I am considering migrating to a backbone type layout with multiple WDS nodes that could run on different frequencies.

I have experimented with b only, b/g and g only. Currently b only offers the best performance.

Please consider that I would have tried nearly every possible combination of channel / mode / frequency before even trying to post.

I am not looking for a "magic bullet" solution here. I am mainly trying to get feedback on what other users have tried and what the results were.