Federico Paolinelli
Federico Paolinelli
Currently frr-k8s is meant to run per-node, as it affects the configuration of the node network. If we move the paradigm, we should allow to have a configuration per pod...
Yes we are trying to get back on track and support a newer frr version. It just will take some time as we need to understadn / narrow down if...
Is this about Openshift? If so, I'd suggest reaching the customer support through the official channels. I am still unsure about the workflow, Openshift uses both ovn kubernetes and metallb...
@gsk1985 mind expanding on what's broken in your scenario?
I am going to close this as obsolete. It's not clear what the use case is. Feel free to reopen if you provide more details. My understanding is, we do...
@gregsulek wouldn't the same result be possible by using the local pref we already have?
Fair enough, you are right. Makes sense to reopen it, frr supports prepending so shouldn't be super hard to implement.
CLosing this as no feedback was provided
@karampok assigning this to you
I don't think it makes sense for L2, and if redundancy is needed probably it's worth moving to BGP. In any case, if L2 is required, part of the request...