On 26/08/2025 11:39, Daniel Kral wrote:
> As stated in the ha-manager #2 patch, I'm inclined to register this at
> /nodes/{node}/ha/... instead of /cluster/ha/nodes/{nodes}/..., because
+1
> then we wouldn't need to duplicate the nodelist in the HA endpoints in
> the latter, because it is already provided at GET /nodes. We could still
> provide information on GET /nodes/{node}/ha for each node, e.g. the LRM
> state, mode, etc.
Just to be sure: do you mean for "each"–as in separately for the respective
node–or for "all"–as in always for all nodes independent on which {node}?
The latter should rather be returned by /cluster/resources, if needed.
_______________________________________________
pve-devel mailing list
pve-devel@lists.proxmox.com
https://lists.proxmox.com/cgi-bin/mailman/listinfo/pve-devel
