MikroTik

I'm getting "OspfNeighbor { router-id: X.X.X.X state: Full } received wrong LS Ack" in my v7.1.1 routers from my v6.48.6 routers.
It seems like restarting the OSPF instance on the v6 routers fixes it for a while.

Any idea what is causing this?

We even got this between a 7.1.1 and 7.1 today. Only visible in the 7.1 router.

Yeah, same issue here ... we have a mikrotik only OSPF with about 40 switches and the logs get spammed a lot. All running 7.1.1

There is 7.1.2 now, but there is no mention of solving this issue ....

Hello there

I have set up a GNS3 lab with only 7.1.2 cloud hosted routers and I am facing the same issue

Did you manage to resolve this?

Hello,
any news regarding this problem?
I've tried 7.1.3 and it's still present. Is somehow related to BFD?

thank you

Hello In my case in the VM lab I had to upgrade and restart the VM host.

After that I have not experienced the same

We have same problem. Very simple OSPF setup, 3x v6 routers, 1x v7 (now 7.3beta37)

On the v7 route OSPF fails with the same error "ignoring LS Ack - wrong neighbor state" , sometimes after few seconds, sometimes 5-10 minutes..

BFD on v6 is enabled.

Same problem here...

Hello,

We had the same problem where V7 routers were getting hammered by LS ACK messages.

What fixed it for us was to upgrade adjacent V6 routers to V7. We have 40 routers in our network and have now upgraded half of them running V7.2.3. So far everything is working well.

We also turn off BFD on V6 routers OSPF interface that connect to V7 routers. Be aware that if you are using OSPF and NAT on the same V7 router you need to disable connection tracking for OSPF.

Not suggesting that you do any of that, but that is what works for us.

Regards,

This still appears to be an issue in routeros 7.3.1. I'm trying to form an OSPF adjacency from a CCR2116-12G-4S+ running 7.3.1 to a CCR1036-8G-2S+ running 6.47.10 or 6.45.9. The adjacency gets stuck in the 'Exchange' state and the ROS7 box constantly reports "received wrong LS Ack for..." Any news on when this will be resolved?

I'm seeing the same error every time a client disconnects from the PPPOE server on the router. Only happens on RouterOS7. Currently using 7.2.3. Most of my OSPF network is still running ROS6, only seeing the error on the router with ROS7

I will try updating to 7.4 later tonight and see if that resolves the issue or not.

I am running v7.4.1 and I am seeing this on a link to a router running v6.47.9. I added a second link between the routers and I am getting the same error on the secondary link.

Hey Mikrotik could we please get an answer on this?

on my case moving to v7.4.1 fixed the issue

Im preparing to upgrade my companies network from 6.49.6 to 7.5 but we cannot do this without showing the transition will be smooth.
In our test lab Im upgrading a single router to 7.5 and connecting over a GRE tunnel. When I do this our OSPF doesnt work like it does from 6.x to 6.x.
We initially get the routes: state Change to Full
We see all routes for a split second and then we then receive these errors on the 7.5 router: received wrong LS Ack for inter-area-prefix 10.10.0.0 10.10.9.14 0x80000006 expected 0x80000007.
This then brings DOWN the gre tunnel entirely, which is very interesting as without OSPF the tunnel is stable.
After this the gre tunnel comes back online and we start the loop again.

We see all routes for a split second and then we then receive these errors on the 7.5 router: received wrong LS Ack for inter-area-prefix 10.10.0.0 10.10.9.14 0x80000006 expected 0x80000007.
This then brings DOWN the gre tunnel entirely, which is very interesting as without OSPF the tunnel is stable.
After this the gre tunnel comes back online and we start the loop again.

You probably already got this figured out, but it sounds like OSPF is installing a route for your external IPs to go across the tunnel. The moment the routes get installed, your routers start sending your tunnel management traffic across the tunnel. You need to filter out the external IPs of the tunnel from OSPF using a Route Filters. You can see my related post about this over here: viewtopic.php?p=961905

did anyone figure out the proper fix for this ? i see it on CCR2216 running 7.7 with 3 CCR's in a triangle.

The OSPF setup is straightforward, consisting of three v6 routers and one v7 (currently 7.3beta37). However, the v7 router experiences an OSPF failure with the error message "ignoring LS Ack - wrong neighbor state," occurring intermittently, ranging from a few seconds to 5-10 minutes. BFD is enabled on the v6 routers.

7.9.1 gave me a lot of issues with this lsack issue. I've seen 7.9.2 released and the problem seems to have gone away. Was only affecting 1 device a RB951G-2HnD but made it unusable. Release notes don't mention anything related so maybe this was just luck

Hello,

I am upgraded to firmware version 7.10.2 and am receiving same warning. Is there any insight as to what might be causing this?


default-v2 { version: 2 router-id: 192.168.0.2 } ospf-area-2.0.2.2 { 2.0.2.2 } interface { p2p x.x.x.x%vlanxxx1 } neighbor { router-id: x.x.x.x state: Full } received wrong LS Ack for router 192.168.0.2 192.168.0.2 0x80003214 expected 0x80003215

I have multiple OSPF routes but this is the only one giving me this warning state

The same here ,I'm faceing the same issue on CCR 2116 With V 7.11 is connected to other routers (CCR2116 V 7.10.2 and CCR 1009 V 6.49.3 etc ...)
the log is : route ospf warning ospf-instance-1 { version: 2 router-id: x.x.x.x } ospf-area-1 { 0.0.0.0 } interface { broadcast x.x.x.x%vlan.21 } neighbor { router-id: x.x.x.x state: Full } received wrong LS Ack for network xx.x.x x.x.x.x 0x80000001 expected 0x80000004

no diffrence when we using Network type:ptp or Broadcast
can anyone give a hand or consider this issue an bug in V7

Hi all,

I'm also getting these errors between 7.10.2 and 6.48.6 routers. Is there any further information regarding these OSPF messages please?

"default { version: 2 router-id: 10.1.100.150 } 10 { 0.0.0.10 } interface { p2p 10.1.1.173%ether8 } neighbor { router-id: 10.1.100.151 state: Full } received wrong LS Ack for inter-area-prefix 100.67.12.0 10.1.100.150 0x80000014 expected 0x80000015"

Many Thanks

Bumpbump, on between any 7.11.2.
(the so called +1 , 2 -> 3)

Bumpbump, on between any 7.11.2.

Code: Select all

default-v2 { version: 2 router-id: x.x.x.13 } backbone { 0.0.0.0 } interface { p2p x.x.x.13%l2tp-x} neighbor { router-id: x.x.x.254 state: Full } received wrong LS Ack for router x.x.x.13 x.x.x.13 0x80000382 expected 0x80000383

(the so called +1 , 2 -> 3)

Responding to myself... ymmv, masqurade / nat seems to cause this (internal interfaces). Probably shouldn't, quite certainly a default type of logic bug.

Referenced here (without the +1 "received wrong LS Ack for router" keywords, why it can't be found, crossposting):
NAT killing OSPF

Bumpbump, on between any 7.11.2.

Code: Select all

default-v2 { version: 2 router-id: x.x.x.13 } backbone { 0.0.0.0 } interface { p2p x.x.x.13%l2tp-x} neighbor { router-id: x.x.x.254 state: Full } received wrong LS Ack for router x.x.x.13 x.x.x.13 0x80000382 expected 0x80000383

(the so called +1 , 2 -> 3)

Responding to myself... ymmv, masqurade / nat seems to cause this (internal interfaces). Probably shouldn't, quite certainly a default type of logic bug.

Referenced here (without the +1 "received wrong LS Ack for router" keywords, why it can't be found, crossposting):
NAT killing OSPF

Not for me. I'm using MT as a VPN concentrator and I don't have NAT configured and still receiving "wrong LS Ack" messages

So did anybody actually log a support case with MikroTik?

https://mikrotik.com/support

You funny. If they don't understand to enrich insight from their own user-base forum and want to hide the revelation process in private tickets then they may suffer their own ignorance. It's not like they would need to worry about competitive advances leaking on a proprietary platform. It's not like they're actually listening, so why bother. Just like with Microsoft, the answers never arrive from the manufacturers in their ivory towers.

@Joni, its almost as if the MikroTik community forums aren't an official tech support method hey?

MikroTik have stated several times that they don't monitor this forum closely as its a user forum for users to discuss amongst each other. Maybe you have noticed that when MikroTik staff do participate in the forums, they will encourage people having bugs or other misbehaviours to log a support case?

Even if they did see this thread I am curious how you think they are supposed to solve it when nobody has shared any particularly detailed information about their setup or what they are seeing, let alone a supout file (one of the first steps of logging a support case).

So if you just want to sit around and feel sorry for yourself that's fine and you have my sympathy. But its been nearly 2 years doing the same thing without any results, so maybe its time to try another way?

BTW to add something productive to the discussion: I am having the same issue with CCR1036 running 7.12 (also seen in 7.11 and 7.10) speaking OSPF with CCR1036 running 6.48.6

While the OSPF adj does form, I get the following log message: I have firewall filter rules but no NAT. Connection tracking was auto, however no change when I set to no.

I will advise the case ref when I log it.

SUP-134528 logged, comments from MikroTik as follows:

Regarding "received wrong LS Ack for external"

Thank you for the report. It is cosmetical bug and does not influence OSPF working. It is reported to our specialists but currently I can't provide ETA.

Regarding similar messages but for "received wrong LS Ack for router"

If there is ROS version on v6 on one router and v7 - on the second side and adjacency is formed - then yes, it is same issue.

You're welcome.

Great, thanks for the info

No word via the support ticket yet, but this looks interesting...

viewtopic.php?t=201465&start=300#p1038543

What's new in 7.13beta3 (2023-Nov-24 13:52):
...
*) ospf - fixed LSA Type3 advertisement for OSPFv2;

I loaded this onto a CCR in my test setup...

Before I was getting "wrong LS ack for router" messages. After upgrading I am still getting those and now also "wrong LS ack for opaque-area".

I am only getting "wrong LS Ack for external" in my live network, so can't confirm if those are fixed just yet.

still has the same issue...

received wrong LS Ack for router

7.13rc3

the problem hasn't gone away

7.13 - received wrong LS Ack


are you going to fix this?

This is the most important info: "It is cosmetical bug and does not influence OSPF working." Only if you like your log to be clear, than I understand you concern

When rc2 released I asked if this was now fixed and got this:

This is a minor problem and is not fixed yet. You will receive a notification when the problem is resolved.

Hopefully 7.14 can fix this???

...

Isn't it just great how you don't get a notification when your support requests are closed. Tried the newly released 7.13.4 but no change.

Since the problem was still happening on 7.14 I asked if there was a particular target date or release for fixing this, this was the response.

it is reported, but as it was mentioned it has very low priority, I cannot say when it will be fixed.

same problem v7.14 stable

I recently encountered this problem after a network overhaul. Turns out I had accidentally used the same router-id on two different routers. Might try checking to confirm that all your routers have a unique OSPF router-id.

i have the same issue and my action is i downgraded my core router to 7.4 from 7.11 and all my other router connected to my OSPF CORE router the issue never seen again in logs

Same here

7.14.2... and it was showing the same error when maintaining adjacency with 6.49.8. Investigated, and ineed the router-id was duplicated in the network. Modified it and it never gave error again

Same here

7.14.2... and it was showing the same error when maintaining adjacency with 6.49.8. Investigated, and ineed the router-id was duplicated in the network. Modified it and it never gave error again

This mistake will result in its own specific and unique error message, will it not? Seems like there are at least 5 different variants of this log message all being thrown into the one discussion here...

I just wanted to update in my cases from observations upgrading to 7.14.3:

• Regarding "received wrong LS Ack for router" this has indeed stopped once all neighbours got up to v7 per advice from MikroTik
• Regarding "received wrong LS Ack for external" I found that routers which had a out-filter on the OSPF instance were originating these messages. In my case out filter for OSPF is not needed so this went away once I disabled it

I recently encountered this problem after a network overhaul. Turns out I had accidentally used the same router-id on two different routers. Might try checking to confirm that all your routers have a unique OSPF router-id.

THIS was my solution. ROUTING > Router ID > Add >
Name: main bridge ip of your router
ID: same as name, main bridge ip of your router (ex 192.168.1.1)
Check only vrf (that works for me cuz dynamic ospf set that)
Select from VRF: main

after that:
Routing > OSPF > Instances > Select your instance and set RouterID to recent created
Disable and enable ospf instance and log is clean

In OSFP Interface template is important the order of networks, for example, first set VPN and then second the local network.
Or try the other way.
Make that in every ospf included router, verify everyrouter use different ip address.

I have 7 routers tunneling via openvpn and accessing everything from the central around my country.
And everything is OK after some hours.
Before this problem, i was losing ospf routing by some minutes

ping latency and router cpu has better performance right now.
Maybe helps anyone

ROS 7.13.3