Accepting ip6 addresses, deploying as ip4 fails.

[hcoin]

microceph cluster config set cluster_network /<prefix reports no error, treats them as ip4. As a result, nothing works. That command should also update ceph.conf

reef/stable

[UtkarshBhatthere]

Thanks for the issue @hcoin. I am afraid the description lacks actionable information. Can you please follow the ISSUE TEMPLATE and populate required information ?

The cluster config CLI does not need to update the ceph.conf file because it uses ceph config CLI to internally populate ceph-mon configs. You can check the populated ceph configs using $ ceph config dump global.

[hcoin]

Just to confirm, is ip6 supported? Because I experienced such as #microceph cluster config set cluster_network fc00:1000:0:b00::/64 (or maybe it was [fc00:1000:0:b00::]/64)

which was accepted, but it broke the cluster.

[UtkarshBhatthere]

Yes, there should be no problems configuring IPv6. Cluster Networks are pretty configurable, so you may reset # microceph cluster config reset cluster_network and things should go back to normal. In order to find out what went bad, we would need more details.