RouterOS equiv. of Cisco's "ip ospf mtu-ignore"?

Is there a RouterOS equivalent to Cisco’s “ip ospf mtu-ignore” to tell a MikroTik to ignore MTU mismatches? (4.x or 3.x…)

Thanks!

Aaron out.

As far as I know, there is no such feature in MikroTik RouterOS OSPF.
MTU is only adjusted for PtP links.

Any progress on implementing an OSPF MTU-ignore feature for RouterOS’s ospf and ospf-v3?

Currently I cannot get a Cisco and a RouterOS box to talk OSPF to each other over IPv6 due to MTU mismatches. But likewise due to other (currently working) devices on the segments in question, I cannot reduce the Cisco’s MTU to match the MikroTik, and the 'Tik’s MTU cannot get any larger.

Here’s what I see in the logs:

22:00:46 route,ospf,info Discarding Database Description packet: too large MTU 
22:00:46 route,ospf,info     mine=1524 
22:00:46 route,ospf,info     received=1546

Cisco routers can ignore MTU mismatches quite easily and establish working OSPF sessions with other devices, just not RouterOS boxes… yet…

Here’s hoping the Mikrotik people add an option like “/routing ospf-v3 interface set 0 mtu-ignore=yes” in the near future…

Aaron out.

I know it is an old post that I am replying to but it is one without a resolution…

FYI I got around this problem by setting the MTU on the ethernet interface that hosts the OSPF instance equal to the “received=” value shown in the log.

You shouldn’t have mismatching MTU’s on links anyways, so you’d best just fix the MTU and then as an added bonus, you’ll avoid all the ugly issues that arrise when you have an MTU mismatch.

Hi, 11 years after…is there any news regarding this?

On 11 years still you have not fixed the MTU error you have on your network?

For those who understand why they need this feature, would be nice to have it.

Its 2022 and Mikrotik still has no fix for this. It pretty simple and they need to understand that in a multi vendor environment across multiple hops this is a REAL issue. Fix your stuff as we don’t always have control over the other vendor devices.

I believe it was already discussed in the forum. Adding hack to make OSPF work over broken network will not fix the problems for actual routed traffic over this broken path.

Yes that correct this is a broken network for sure that I took over. MTU should never be touched unless it is for a VPN tunnel and only proven to be needed. With that said what I found is if I create a second OSPF instance area 0 and add the peering interface to the downstream Cisco I can get the OSPF to form.

However, I cannot seem to be able to add the networks to both instances and get them to advertise out to both peers. I am pretty sure that configuring the knob to advertise statics on both instances will work.

Does anyone know how I can get the configured subnets under the network tab to advertise to 2 ospf instances?

Any suggestions is greatly appreciated.