Latest leases patchset - getting there !

[Jeremy Allison]

On Mon, Nov 17, 2014 at 08:02:38PM +0100, Stefan (metze) Metzmacher wrote:
> 
> It would also be good to verify this section with a test:
> 
>         if (lp_locking(fsp->conn->params) && file_has_brlocks(fsp)) {
>                 DEBUG(10,("grant_fsp_oplock_type: file %s has byte range
> locks\n",
>                         fsp_str_dbg(fsp)));
>                 granted &= ~SMB2_LEASE_READ;
>         }
> 
> We need to make sure we never grant "W" or WH" leases.

Yes, looking at this right now it would seem
that if we did:

open RWH LEASE1 ->
		<- reply RWH, H1
locking H1	->
		<- OK
open RWH LEASE1 ->

we would reply with (WH, H2) - which should
not be allowed. Windows clients don't do this,
which is why we wouldn't have caught this before.

I'll check with a test.

> It's also unclear to me why a client shouldn't be able to get a read lease,
> when it already has brlocks (and a possible write lease).
> 
> Do we have an oplock related test for this already?

Yeah, we do have some SMB2 oplock + byte range
lock tests, but I don't think it checks this
specific case.

Jeremy.