On 03/29/2016 05:01 PM, Eric Blake wrote:
> The existing transmission phase protocol is difficult to sniff,
> because correct interpretation of the server stream requires
> context from the client stream (or risks false positives if
> data payloads happen to contain the protocol magic numbers).  It
> also prohibits the ability to do efficient sparse reads, or to
> return a short read where an error is reported without also
> sending length bytes of (bogus) data.
> 

> +* `NBD_CMD_READ`
> +
> +    If `NBD_OPT_STRUCTURED_READ` was not negotiated, then a read
> +    request MUST always be answered by a single non-structured
> +    response, as documented above (using magic 0x67446698
> +    `NBD_REPLY_MAGIC`, and containing length bytes of data according
> +    to the client's request, although those bytes MAY be invalid if an
> +    error is returned, and the connection MUST if an error occurs

MUST be closed

(I hate it when I send patches before 'git commit --amend' on the final
unsaved changes in my editor...)

> +    after a header claiming no error).
> +

-- 
Eric Blake   eblake redhat com    +1-919-301-3266
Libvirt virtualization library http://libvirt.org