Re: [pve-devel] [PATCH pve-docs 5/6] pvecm: add note about corosync killnode error

[Fabian Ebner <f.ebner@proxmox.com>]

Am 15.09.21 um 15:36 schrieb Dylan Whyte:
> Mentions the misleading error message shown, when deleting a node,
> because of the failing command:
>      corosync-cfgtool -k x
> 
> Some forum users were confused by this, and believed that the removal of
> the node was unsuccessful.
> 
> Signed-off-by: Dylan Whyte <d.whyte@proxmox.com>
> ---
>   pvecm.adoc | 5 +++++
>   1 file changed, 5 insertions(+)
> 
> diff --git a/pvecm.adoc b/pvecm.adoc
> index af9cf33..e7140a6 100644
> --- a/pvecm.adoc
> +++ b/pvecm.adoc
> @@ -340,6 +340,11 @@ After powering off the node hp4, we can safely remove it from the cluster.
>    Killing node 4
>   ----
>   
> +NOTE: At this point, it is possible that you will receive an error message
> +stating `Could not kill node (error = CS_ERR_NOT_EXIST)`. This does not
> +signify an actual failure in the deletion of the node, but rather a failure in
> +corosync trying to kill an offline node. Thus, it can be safely ignored.
> +

Currently our API call/CLI command fails in this scenario. Can we 
reliably detect the situation and not run
     corosync-cfgtool -k x
then? Or always run it to be sure, but gracefully catch the specific 
error and print it as a warning (maybe adding something like "- node 
already offline?") instead of failing?

>   Use `pvecm nodes` or `pvecm status` to check the node list again. It should
>   look something like:
>   
>