John Baublitz

This seems reasonable to me! Do you think this would be a good issue for @bmr-cymru?

Currently, I think we're tending towards extending the static header with a timestamp of "last updated".

@bleve Does disabling SELinux or updating the policy solve the issue of the block device being inaccessible at boot? Have you tried it?

Blocked by #3035

Hi @rvql! This may actually be a bug. `Rtgenmsg` was added by an external contributor and it looks like attributes are not allowed to be set for that type yet....

@rvql I took a look at `addattr32` and, as I expected, it's following the typical rtnetlink format. Netlink in general has the following format: ``` |-----------------|----------------------|-----------------------| nlmsghdr subheader attributes (genlmsghdr/rtgenmsg)...

This is now resolved by f05308cd0cab47bd89c40df744a9a1bd661ce473 in #182

I'm beginning work on this issue in the `v0.5.0-dev` branch.

Work started in branch `v0.5.0-dev`.

@isomer You've previously provided some incredibly helpful feedback around how you're using neli in the context of both multicast groups and request-reponse workflows. This is exactly what I want to...