SMB2 and allocation size

[Stefan Metzmacher]

Hi Andreas,

I'd actually like to have the server fixed.
We need to provide the behavior of a Windows Server

Should we only look at the block number if size != 0?

Otherwise we would have to apply the change to a lot of torture tests!

BTW: did you noticed that the comment and the define are not using the
same value?

> my gitlab runner for MIT KRB5 failed when running the samba3.smb2.session test 
> with the following error:
> 
> Exception: Exception: ../../source4/torture/smb2/session.c:85: (&io1)-
>> out.alloc_size was 1048576 (0x100000), expected 0 (0x0): out.alloc_size 
> incorrect
> 
> I've tracked it down to a difference in vfswrap_get_alloc_size().
> 
> On the docker image an empty file gets 8 blocks:
> 
> $ touch foo && stat foo
>   File: foo
>   Size: 0         	Blocks: 8          IO Block: 4096   regular empty file
> Device: 809h/2057d	Inode: 1593246     Links: 1
> Access: (0644/-rw-r--r--)  Uid: (    0/    root)   Gid: (    0/    root)
> Access: 2018-11-30 11:12:16.000000000 +0000
> Modify: 2018-11-30 11:12:16.000000000 +0000
> Change: 2018-11-30 11:12:16.000000000 +0000
>  Birth: -

What's the filesystem there?

We have some private runners where we have an ext4 filesystem...

metze

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: OpenPGP digital signature
URL: <http://lists.samba.org/pipermail/samba-technical/attachments/20181130/17214218/signature.sig>