v7.13rc [testing] is released!

A downgrade to 7.13beta3 on the cap ax “fixed” it for now.
All devices in this ssid are now in the correct vlan again.

Cross posting from SUP-134528:

Should the OSPF logging issues be fixed now, or still a work in progress?

Loaded 7.13rc2 and I am still seeing the following:
received wrong LS Ack for opaque-area
received wrong LS Ack for router

yeah OSFP routing stopped working correctly with 7.13rc2 :confused:

The routing table is populated as usual, but packets are not routed (tested on ax3)

no issues with ospf in 7.13rc2. Have it running on 2 separate installations, rather small topology, but routing and route exchange works.

Thanks for adding :deserialize. This is really useful for interacting with REST-Interfaces. I was already able to implement a scheduled auto-login for an annoying AP with a captive portal.

Also can’t wait for migrating my cAP acs to wave2 once 7.13 is out, it’s looking promising on my testing ac3.

RouterOS crashes when I am importing a WireGuard config file.

Can you share the config file (without any keys)?

Here you go:

# 12: 12 > wgclient_12.conf
[Interface]
Address = 10.255.2.12/24
DNS = 10.255.2.1
PrivateKey = 0C7w3meByHk1VQ2mu9QnbiUtcWdH2joB/aJFateQ6UE=
MTU = 1280

[Peer]
PublicKey = lVaGrM7TX/H1yPG5ZuwJvclWtaE+a1XwWYeS5to/CwU=
PresharedKey = 6AZ6Mvl+z1L2eKgifQrsi/tWPSN+mcsO+QjSmwTSJUs=
AllowedIPs = 10.255.2.0/24, 192.168.0.0/16
Endpoint = my-endpoint:51820
PersistentKeepalive = 25

I’ve generated the keys for this example.

Possibly an address clash? You habe 10.255.2.0/24 for interface and allowed-ips inside the tunnel.

Pardon the “me too” post, but I observed an OSPF regression between 7.13Beta 2 and 7.13RC2. The beta ran fine on my home/office CCR2116 for a number of days (a week or two I believe). Loaded the RC for a few minutes (it peers via OSPF and BGP to the core of my ISP network) and adjacencies kept bouncing. Unfortunately it wasn’t at a time where I could grab a supout to submit; I quickly had to switch partitions back to 7.12 to get things up and running.

Yes, this is malware. I came across it something like ten years ago on a few customer PCs when I was running our WISP. Common AV packages including Malwarebytes could not detect it, and since it was on the customers’ PCs, I was not able to take it into the shop for a deep dive.

Question about the wifiwave2 repartitioning:

Is any of this going to fix the issue where on an Audience, you can get WPA3 or you can get meshing, but you can’t get both?

If not, is it because the task is impossible due to hardware/protocol limitations, or possible but just not coded for?

I don’t think wifi driver repartitioning will help with meshing+WPA3 in any way.

I still have memory leaks even on 7.13RC2(over 500mb a day)
SUP-136530

It’s something that i am also interested in. Support for WDS meshing should be on MikroTik’s priority list for the new Wifi package, especially for the Audience use-case. In my opinion it could be possible, since in 7.12 there is already support for station-bridge mode, so the “road” is pretty much laid out in front. That being said, you could do a workaround for the mesh setup, using a PtMP topology where you select a well placed (centrally placed) device in AP mode and the other satellites in station-bridge mode.
Let’s hope that MikroTik is working on a mesh topology solution for Wifi…

All the best!

That should be no problem. The config is auto-generated by a script, so it should be fine.

And even if it was invalid, the router crashing is certainly not the right way to tell me that.

You should file a support ticket with a supout.rif from router, that should have details on the crash for Mikrotik. It could be the line endings, or the “#” comment…dunno, but agree it shouldn’t crash

FYI
I did this upgrade to my rb5009 which is capsman to some wifiwave2 devices as well. I had recently added Adguard Container and when I did the upgrade the container didnt work anymore despite still running, couldnt ping it. So I had a DNS issue that took me a while to fix (forgot about the Nat settings!!) before eventually going to my backup DNS which I run on my Nas or even cloudflare. On top of that though the router itself was really really slow, even typing in a terminal took ages.
Resources on the device appeared normal so not sure what it was.
It was enough for me to roll back to 7.12 stable I won’t be going near it until it comes out of test but even then I will need to be sure before going ahead as it didnt seem to recognise my Cap devices as they werent upgraded and as my client pc was running via one of them this created an initial issue. I will need to devise a good procedure for the next upgrade as I need the caps (Hap-ax2s and a Audience) to be able to work immediately.

I upgraded my RB5009 last friday, also acting as capsman, incl AX2 and AX3.
On that RB5009 are also some containers running (amongst them PiHole).
I had zero problems with that upgrade.

One exception: upgrade via capsman of the cap devices was not possible until I specified an upgrade path and the necessary files on RB5009.
This is a consequence of the split of wireless drivers from the base package so on itself logical.

Seems to be missing EAP Method Passthrough option.