OSPF error log on MPLS interface

[rys]

                   Dear Friends,

I have noticed a log message of  "%OSPF-5-ADJCHG:............ Too many retransmissions "on of our core router connected another core router, which are served as the backbone between the datacentres.

These routers are running MPLS in these interfaces, and has an MTU size of 4470 on both the interfaces.

But I am not able to  ping to the other end IP with this size, the Max size I can ping is 1590.

So is this MTU causing the OSPF logs.....

Kindly share your suggestions and thoughts  and what all things i need to verify.

Thanks in advance.

Riyas

[medhat elsaeed]

Could you please tell us the type of interface you are using

[rys]

Its a Gigabit Ethernet, using media type SX.

[Vivek Ruhil]

Hi

Are these routers connected back to back or is there a Telco link wiht their EoSDH network.

If it's telco, please check with them for the supported MTU value and kindly use that MTU value on your Cisco devices.

The retransmissions message and your ping indicates that you have a higher MTU value configured than the one supported by your Telco.

If asking Telco is taking time, you can try changing the MTU value to 1550 (will cover all possible MTU values for MPLS enabled network) and see that these messages should disappear.

Just a thought, do you have CRC errors incrementing on your Cisco devices.

[rys]

HI Vivek,

Its connected using a fibre link provided by the local provider,

Regarding the retransmissions what I see most is mtu issue.

but eventually this set up is being running like this for a long time.

I noticed a few input errors in the one end of this link interface.

a day before the same interface went to err-disables state and cleared after we reset the interface.

we have back up link connected between these 2 sites via another router, all these routers are forming the

mpls cloud which serves the network.

How long it takes for the redundacy to kick in an mpls network??

[Vivek Ruhil]

Hi

The provider most likely has some kind of a switch or a SDH device inbetween that has a maximum MTU limit. Please check with them for the maximum MTU on that link.

As you said that this interface went in err-disable mode and after interface clearing the link came up, most probably during the same time the link provider must have done some changes on his end maybe a switch change or port change due to which now you get lower MTU. The only way out is to check with the provider and ask them to change the MTU to 4470 that you need.

Redundancy in an MPLS network entirely depends on your IGP, if you are using IGP fast hellos or BFD it is typically sub-second convergence these days but if you are dependent(very unlikely) on the IGP standard hellos that it should take somewhere around 30 secs to converge.

[Muhammad Thanveer]

Dear Vivek and Riyas,

I hope you are doing good. Vivek has given an very intelligent answer, please check it and when the matter comes to your question

How long it takes for the redundacy to kick in an mpls network??

I can understand why did you ask this question, Friend, MPLS makes your packet to reach faster by ignoring the checking process of source, destination and all those stuff, when the matter comes to your redundancy it is matter of time which is played by the timers.

If you are using BFD then it is a different matter.

BFD is a detection protocol designed to provide fast forwarding path failure detection times for all media types, encapsulations, topologies, and "routing protocols"

Just for fun--Here in our place we where taught to knock the door three times, if no responce comes we decide that there is no one to answer.

Dear Vivek,

It was an intelligent answer you gave you deserve fie stars..*****... It comes with the experience in fact.

Regards
Thanveer
"Everybody is genius. But if you judge a fish by its ability to climb a tree, it will live its whole life believing that it is a stupid."

[Vivek Ruhil]

Hi Mohammed

Thanks a lot for your words. Much appreciated.


Sent from Cisco Technical Support Android App

[rys]

Dear All,

Thanks for all your valuable comments, I sheduled a testing with my local provider since the input errors are keep

seeing in the interface. It not seems like an MTU issue, the interface is able to suuport jumbo frames and if mtu issue should be intermiitent.

Let see whats the out come after testing.

Thanks again for your all support.

Riyas

[Vivek Ruhil]

I hope by interface you meant that the Telco provider can support Jumbo MTU. It is good to have the same MTU value end to end so that you don't end up in unncessary fragmentation along the path.

[rys]

Dear All,

We ask the provider to have test on this link and they confirmed ts error free and MTU is within the limit, 1450.

After 1 week, we still got these errors again in the otehr end, too many retransmissions and ospf is not seeing stable.

Could it be because of the fibre patching or a Port issue ???

Thanks

Riyas

[Vivek Ruhil]

Hi

could you provide the below items:

1. Ping the other end with df-bit set and packet size of 1500 repeat for 100 packets

2. OSPF configuration on both ends and also the interface configuration.

Please remove all passwords when pasting your configurations online.

[rys]

HI ivek, Attached is the config for ur info

[Vivek Ruhil]

And the other end config plus ping

[rys]

HI Vivek,

The other config is exactly the same except the IP address. See the Ping below.

The