Hello. NV2 have still random problems with locked modulation on 6,5Mbit. This is P2P link. On both sides is ROS 5.25
AP side show 6,5Mbit TX / 130 Mbit RX. In the same time client side of the link show 130/130Mbit.
Sometimes it happens after 30 days, sometimes after 80 days. It’s completely random. Throughput is very poor because of the locked modulation
See Ticket # 2012122166000702
I think so this is problem on AP side. Just turn off/on wireless and everything is ok - 130/130Mbit
Not sure this will help but I have set all my AP’s from default frequecy scan to +/- 20 to 40Mhz above and below the AP frequency, for example AP=5220, frequency scan 5180-5260, the same at the other end of PTP
See BW test over wlan interface. Maximum troughput in tx is about 5Mbit. It is simple TX is locked to 6,5Mbit, but why? You don´t have sometimes this problem?
Disable/enable wireless, then is rate 130/130 a troughput in tx is over 60Mbit. This problem is only with NV2 protocol
We are observing this problem too (for long time - maybe 1year+). Reported to support several times.
Typical signs of the problems is that latency on the link raises (but it depends on traffic on the radios) and the throghput lowers and when you try to find out what happened you see that AP is showing some low modulation in TX direction to a station. And the modulation doesn’t change during time. Usually you will see a low modulation like 6.5mbps and even there is a good quality the modulation is still the same. Client side in the same time will show RX modulation which is different (higher). If you check times of RX modulation last measurement showed by client (winbox’s Signal Strengths or print stats in registration-table) the there is only one or 2 modulations with lower times (measured recently). If these modulations are low ones you have a problem.
It is possible to write a script which will check all the client stations and inspect the modulation table’s times (btw: thank you MT for the stupid time representation used in CLI)
And it looks like any modulation can be the locked one (we have seen already locks on 6,9,12,18,36). The lower ones have higher impact on the network. Btw MT shows another modulation name in ‘Signal strengths’ and in the registration table (6 versus 6.5 etc).
I understand the MT’s support has not enough information to kill the problem (it occurs rather randomly) but from another point of view I thing they should take the problem much more seriously. What is Mikrotik without functional wireless layer? Nothing. If the problem had enough priority it could be solved before months.
the only way to fix this low modulation lockup is by uploading special packages and monitor that link till it locks up and then we could try to apply a special package for that and monitor again. And continue to do so till we will find the problem and solution. But to fasten all that we would need a link where it could happen more often - not once in a month.
soon we will have ready a mass monitoring script which will tell us whether there is a location in our network where the lockups occurs frequently. Probably it will not be on a PtP line but on an AP resp CPE routers at customers. The problems on PtP links are relatively rare (although not so rare we can ignore them) . I will contact you.
I’ve seen the situation that AP and CPE has different readings for the modulation in their registration table. AP was 5.25 and CPE 5.6 (rb133c). AP shows 6MBit while CPE shows 18MBit for the same connection. Protocol is NV2. Reboot of the cpe solved this.
Other CPEs on the same AP did not show this behavior.
Wireless layer - funkcionality is very very important. Now we use NV2 everywhere. But this know issue is very annoying.
I understand that support wants access to the AP where it happens often. But how to do that if it is random problem? The problem occurs maybe once in 30 days. Then clients call and complain about the speed.
I give remote access to “Locked AP” for support@mikrotik.com, but nothing changed.
They need to problem occurred frequently. But how to do that?
Ive seen this a lot on nstreme data rate will lock on 6Mbit after 60-90 sec link will disconnect and reconnect
On nv2 Ive seen it on 3 or 4 times - wrote to support but no feedback
I have some stations which lock near every day. But most of them are RB112/133.
Each time I run the script I find several locked CPEs. If all is OK I will contact Mikrotik support tomorrow to arrange the remote access to the stations/APs
the rates on CPE (shown in Registration Tab) are higher than the corresponding ones on AP (usually there is maximum rate displayed). Typically you can find that a ‘locked’ CPE shows 100% quality in RX direction (probably because it gets all the packets on very low modulation) and when you have “Last Measured” of modulations ordered by their time you see only one modulation (typically basic modulation i.e 6mbps) or two modulations which were measured before very short time but and all other modulation were measured deep in the past (I use 5 minutes as a threshold in the script).
On the AP you can identify ‘locked’ station when you check all the stations with low TX modulation - run (MAC) ping to them with delay 100ms and wait a minute if the TX modulation goes higher. If it remains the same (and the TX quality is not poor) the station has probably locked modulation. Then kick it off - seelct it and press the “-” (minus) button near to the top left corner of the Registration Tab window). After the station reconnects check if the modulation goes higher.
I have sent the passwords to support before one hour. Unfortunatelly, from offered list of clients (most of them slow boards), they selected the one which is RB433 and locks not so often. And I am not sure they will want to solve the problem in 5.X since I had to upgrade the client and AP to 6.5 version…
