Replies: 3 comments 3 replies
-
I believe that yes, it would be better.
|
Beta Was this translation helpful? Give feedback.
-
Depends on what actual devices do. I thought they all set default mtu to 1500 as part of the nos itself. |
Beta Was this translation helpful? Give feedback.
-
The lad doth protest too much, methinks ;) Let's look at the OSPF test results: https://tests.netlab.tools/_html/coverage.ospf.ospfv2
It looks like most devices use the 1500 MTU as the default, so we have yet again a problem with a quirk of one of the platforms and you're trying to make it a "netlab should set a provider default" problem. There is a way to set the default MTU per device per provider (I happen to know it works because we're doing it for the cumulus device); please feel free to do that for Cumulus 5.x and document what you did in the caveats. |
Beta Was this translation helpful? Give feedback.
-
OSPF validation in integration/lag/01-l3-lag.yml fails for Cumulus_NVUE on Libvirt, because no default MTU is defined for that case on both FRR and Cumulus side. It works on clab
Wouldn't it be better if Netlab would define the default MTU for all providers, instead of only clab?
Beta Was this translation helpful? Give feedback.
All reactions