[stgt] [PATCH] check for valid parameters when passing tgtadm commands

[FUJITA Tomonori]

On Wed, 08 Oct 2008 10:04:40 +0200
Doron Shoham <dorons at Voltaire.COM> wrote:

> > This checks only 'update' option. With your change, we can't catch a
> > mistake like:
> > 
> > tgtadm --op delete --mode target --name hoge --value xyz
> > 
> > 
> 
> this specific mistake will be caught because --tid isn't provided.

That's not my point.


> following your method we'll need to check many more combinations.

That's not issue.


> For example, why not checking:
> tgtadm --mode target --op show --tid=1 --initiator-address=xyz

We should do.


> In my opinion, as long as the user has provided all of the valid options
> the command is a valid command.
> so it won't be a mistake to write 
> tgtadm --op delete --mode target --tid=1 --name hoge --value xyz
> it will just delete tid 1.

IMH, we should give an error or a warning in such case.


> > The null backing store code doesn't need the path option because it
> > doesn't perform any I/O. It's just for performance measurements.
> > 
> > Check out usr/bs_null.c
> 
> As far as I understand you have to use -b (path) option
> in order to create a null backing store.

Can you read the previous mail again?

Now we always require the path option but the null backing store
doesn't need it. So we should let users create a new logical unit
without the path option.
--
To unsubscribe from this list: send the line "unsubscribe stgt" in
the body of a message to majordomo at vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html