Re: [Nbd] [PATCH v5] doc: Add NBD_CMD_BLOCK_STATUS extension

[Alex Bligh <alex@...872...>]

> On 13 Dec 2016, at 16:06, Wouter Verhelst <w@...112...> wrote:
> 
>> 
>> "If a server supports the `base:allocation` metadata context, then writing
>> to an extent which has `NBD_STATE_HOLE` clear MUST NOT fail with ENOSPC
>> unless for reasons specified in the definition of another context."
> 
> That is essentially what I mean, yes. The point though, is that I also
> think a client should assume the worst.
> 
> (additionally, there may be reasons for ENOSPC that the server isn't
> aware of; so even if NBD_STATE_HOLE is clear, it should still not be an
> error for the server to return ENOSPC)

All of this suggests 'SHOULD NOT' would be more appropriate than
'MUST NOT'.

-- 
Alex Bligh