[BBLISA] Does read only really mean it?

[Dewey Sasser]

On 12/04/2013 10:37 PM, Alex Aminoff wrote:
> Nevertheless, I tested it and unless I messed up my test, an NFS mount
> with -o ro, you read a file on the mounted FS, and the access time is
> updated.
>
> For the test the server was a NetApp, the client was Linux.

> There is a mount flag -o noatime that does what I want. But I would
> argue that this is not right. The simplest behavior - nothing is ever
> written period - should be what you get by default, and then there could
> be a flag that enables exceptional behavior, that is updating the access
> time.

I would expect "read only" to mean "no writes of any kind". However, I 
could see it being valid if an NFS client might mount read only and not 
send any writes (nor any metadata updates) yet the server updated atime 
itself due to a client reading.

For example, SAMBA is a user mode process and a share exported "ro" is 
still reading files on the local file system, so the kernel would have 
no way of distinguishing (unless SAMBA opened all files with "O_NOATIME" 
for a read-only share).

NFS on Linux is typically a kernel level process, so the rules *could* 
be different.  I have no idea how NetApp implements NFS.

If a local file system is mounted "ro" I would expect it to mean "no 
writes go to this file system".

--
Dewey