With 7.9rc1 I suddenly see VRRP announcement traffic on interfaces that have a group-master (which is not themselves). This is causing rapid address flapping Logs 15:50:13 vrrp,info vrrp-security-dns now MASTER, master down timer 15:50:13 vrrp,info vrrp-security-dns now BACKUP (set by VRRP Group) .....

Support at MikroTik helped me along and now it works. Here my steps for posterity and in case someone else runs into the same issue. 1. Update (packages) of the router to 7.4. If you can't see them, choose "upgrade" channel, that should give you the option. Packages 4.x wont do 2. After t...

This happened to me too, but it started working suddenly after I disabled both the zerotier interface and the zerotier instance and then re-enabled both. Interesting. I tried that a couple times, doesn't work. Now, for the even weirder part: The only way I can get the ZT interface to come up is: - ...

I had the same issue. I have fixed it by setting "disable-running-check=yes" on my interface. This made the interface work. The parameter is, however, completely undocumented, so I don't know exactly what it does (beyond what it sounds like), or why it is required. I hope when the package ...

So you need the ZT client/app/package on each device as well as ZeroTier enabled on the MT Router? You do not. You can use the router to forward packets to the ZT interface like any other interface. You can even put the ZT interface on a bridge (don't know if ROS supports that, yet, but ZT does). B...

How does ZeroTier license works with Mikrotik? https://www.zerotier.com/pricing/ Does it count all user on the inside, or only see the Nat traffic? ZT only counts actual nodes connected to the network, as in, devices that run the ZT client/app/package. You can forward as many devices through as you...

I have played around a bit on my RB4011 with this. It is working beautifully now. However, I found I had to set disable-running-check=yes to make the interface come up. That parameter is not described anywhere, what does it do (I just found out it fixes my problems by experimenting around)? My confi...

I am using DHCP(v4) to obtain IP addresses and default gateways. In ROS 6 I could enable the "check-gateway" feature on DHCP provided routes using route filters with the "dynamic-in" chain. This chain does not seem to "trigger" onto DHCP obtained routes in ROS7's /routi...

Recently bought an RB4011. So far the missing IPv6 FastTrack doesn't drag me down (simply because it has so much CPU available to it). However, I'd still love to keep my CPU less utilized when I push gigabit IPv6 (jumps to 25% use with gigabit in one direction, so it'd probably be fine for bidirect...

Recently bought an RB4011. So far the missing IPv6 FastTrack doesn't drag me down (simply because it has so much CPU available to it). However, I'd still love to keep my CPU less utilized when I push gigabit IPv6 (jumps to 25% use with gigabit in one direction, so it'd probably be fine for bidirecti...

Are you using the MTik device as the router (it establishes the PPPoE session or it is behind a router but still does NATing) or as a switch (it bridges all ports / switches them)?

export + import via text rsc file. Even if you do not want to import all settings then you can paste them one by one into terminal window with Ctrl+C + Paste When I tried this, the import errored (for example, I have a bridge br-lan, instead of default bridge-local, so the import errored in that ph...

That is interesting. But is this just a coincidence (client process), or is controlled by the CAPsMAN? I think it does depend on the Wi-Fi client hardware you are using. However it does make sense only same-channel roaming works seamless: - To check for APs on different channels it would have to go...

strong crypto enabled. some tool for generating public openssh key? In puttygen you have a field for the public key ("to paste in authorized_keys"). You take the entire contents of that field, paste them in a file and then put that on the router and import it. See link: http://blog.muhamm...

In my own experience (got several RBs in my home) if you set them up with CAPsMANv2 (wireless-cm2) and set all of them to the same frequency/ssid/key (just the exact same provisioning profile), the devices freely roam on their own (my iPhone very quickly swaps access points when I move through my ho...

All scripts are in my GitHub repo at https://github.com/Doridian/FoxdenNetworkScripts There currently is: - certget (automatically get an SSL certificate and re-import it. I use with LetsEncrypt on my RasPi to have working LE certs on the routers fully automatically) - dnssync (syncs DNS static entr...

The CAPsMan configuration is very similar: https://blog.linitx.com/howto-improved-capsman-wireless-client-roaming/ Thanks for that, works perfectly. Only thing I wish RB could do was "zero-downtime" roaming, so that one access point can "seamlessly" hand off clients to another. ...

You'll want to set up both as AP-Bridge with the same SSID and same encryption, etc.

Same channel or different channel? I would assume different channels?

I did this at my home once, and was very unhappy with the instability of things - I had a 951 acting as the wds slave.... it was constantly losing connection to the master or to my devices, but when I returned it to being just a bridge client, it was stable again. Remember that when you use such a ...

Setup is as follows: Modem <-> RouterA (RB2011) <-> RouterB (hAP) RouterA and RouterB both have wired devices attached (computers, IPTV boxes, ...) Both of them also provide a wireless network. What I want now is both of these access points to provide "the same" wireless network. As in, wh...

Apple has released OSX 10.11.2 today which "fixes Wi-Fi stability". So far these disconnects have not repeated since I updated...

More findings: - MBP's BSSID / channel still correctly reflect the RouterBOARD as the AP - Disabling/Enabling TKIP/AESCCM/both does not fix nor break anything - Disconnecting from the Wi-Fi gets seen by the RB (disconnected: station leaving) - Wi-Fi debug logs show absolutely no useful information a...

Disable the TKIP encryption and leave only AES enabled.

This is what I used to have it at. I enabled TKIP to see if it would fix the issue.

The problem is simple: After waking the MacBook (Pro, mid-2015) from sleep, the Wi-Fi stays disconnected until RouterOS says "group key exchange timeout", at which point it will reassociate and everything is fine (both the MacBook and RouterOS still claim they are connected to each other)....

Somehow OpenVPN has become unstable with 6.9. It was super stable in 6.7 now I get disconnects every few hours...

If you really need those features at the moment. Run openwrt image in metarouter and run ovpn from there with full feature set.

This is not the solution. With metarouter my router's CPU gets horribly overloaded so much that my whole LAN starts lagging (RB2011).

I don't care too much about LZO. But I need UDP tunnels (TCP is unusable for many situations as a tunneling protocol as outlined by others previously).
Again, MetaROUTER is not the solution here, especially by causing a major CPU load just to run on my RB...