[libvirt PATCH 0/5] Formalize the deprecation of arguments in virsh

[Peter Krempa]

On Tue, Mar 23, 2021 at 14:36:19 +0100, Michal Privoznik wrote:
> On 3/22/21 5:09 PM, Tim Wiederhake wrote:
> > virsh has several arguments that are better not used. This series introduces
> > a formal way of marking them as deprecated.
> 
> Commit messages are rather sparse. What we currently have is hiding options
> we deem obsolete from users and replacing them with better ones (just :Ggrep
> VSH_OT_ALIAS). No error message, no warning. What makes these you picked
> special? I'm not against reporting that an option is obsolete, but I don't
> quite understand why we need a different way for obsoleting those three.

There's one exception seen in 5/5 and that is the 'base64' parameter of
'secret-set-value'. (It's not a boolean enabling base64 mode but rather
a base64 value of the secret passed on the commandline).

That option doesn't have an _ALIAS, but really shouldn't be used at all,
thus we do print a warning.

For this particular case we could consider hiding it, but it might be
problematic as it doesn't use VSH_OFLAG_REQ_OPT, so is applied without
the option name prefix [1] , which could be very misleading to users if it
were hidden from the help output.

Additionally for that particular case, printing a generic deprecation
warning wouln'd IMO be enough as it's really a security issue, not just
deprecation.

[1]:

$ virsh secret-set-value c021dc3e-8227-4bfa-8f1c-ebd0356fb872 YmxlCg==
error: Passing secret value as command-line argument is insecure!
Secret value set