Hey All,

Some background on the issue: For the past 3 months we have been plagued with an un-solvable issue at 3 of our sites. High ping times regardless of whom we allow to connect to the ap. We have ensured that this is not an interference issue, from outside or from co-location This is happening on ap's with xr2's. One site is a horz omni, the other two sites are vert 120 sectors with another ap colocated on same tower but with approx 20' horizontal seperation. At the sectored sites the issue only occurs on ONE of the sectors. We have an eclectic grouping of cpe's. Primarily Demarc Tech, a few ns2's and now RB411s. We have spent ALOT of time and lost customers during this time. We are a very experienced wisp with 7 years under our belt and 13 years under mine (I only stated this in an effort to give some background). We have a total of 27 ap sites, this is happening at only these 3 for the past few months. Ping times are in the min of 500ms and up to 2kms.

Last night I recalled an issue (that we still support) that happened a few years back. Our first cpe's we installed in 2003 were Senao Bridges in Rootennas (please dont give me any flack), which btw are still operating strong! Anyhow, When we upgraded the AP from a Ruby Chipset to an Atheros (at the time was on StarOS), our network came to a halt on that AP. Took about 3 weeks to solve, the issue was simply due to the chipset being incompatible with the senos AND the demarc cpe's on the same AP. We moved ALL of the senaos onto their own Atheros interface and have never had a problem since.

Having remember that scenerio I downgraded the problematic AP's from 4.14 to 3.3. Problem solved so far, need more time to verify but looks promising. There wasnt ONE setting changed by me, just a downgrade. We use very little configurations in our APs, especially on the wireless side. My thought was that the grouping of CPE's on those paticular APs were conflicting in some way that the 4.x OS either was causing or could not accomodate.

Could this be???

NVM, problem is still there! I've learned to be patient knowing the solution will eventually come, but I am at a loss.

Dallas or any other, smarter than I person be willing to consult? If so please PM me asap.

HAVE U IMPLEMENTED RTS/CTS AND SEEN THE RESULT

No I havent with the mikrotiks, do you have suggestions on best practices?

I have no idea if this will help or not since we are running all StarOs on our AP's. I have experienced the same issues as you have described, and determined that it was one customer that was causing the issue. He has a -79 signal, and only happened when he was uploading to many different hosts. I knock him off and the AP returns back to normal again. Took me quite a while to figure this one out.

I do have a question for you though given your StarOs background. Are you still running any StarOs, or have you given it all up, and how does a MikroTik AP compare to a StarOs AP? Would be easy to change out for us since we are running all XR2's at the moment.

the wiki has some guidance on using rts/cts and it has helped me when i was facing similar issues. it has toned up the traffic for each client but u need to configure it for all clients and also each for ur AP radio. u also need to keep the upload and download of clients in mind.
the actual effect is seen in the pthrougput and after configuring each client on an AP. also check ur network to see if any client has weak signal. remove them. ccq should be 90% + for clients with good signal.