Cedric-
On Sep 11, 2008, at Sep 11, 2008, 5:02 AM, Cedric Le Goater wrote:
The conservative play is to always force use of the initial namespace
and to deny creation of mounts that would use different
namespaces. In part
because the initial version of the namespace always exists. Which
means
as relates to Cedrics initial patch we would still need to know which
mounts should cause us to use a different uts namespace so we can
deny
them.
I will send the initial patch which forces use of the initial
namespace because
it does fix a real oops.
Today's patch looks reasonable to me. I would add a FIXME comment in
net/sunrpc/clnt.c near the site of the fix.
Then, I should be able to find some time to work on improving the
uts namespace
checks when NFS mounts are done.
As a final note, I don't think this is an issue only for NFS mounts.
The problem is with how AUTH_SYS RPC credentials are formed.
cl_nodename is a copy of utsname kept in the rpc_clnt structure for
efficiency. I think the auth_unix code has to be more sensitive to
which utsname is "in effect" for each RPC request. We probably can't
use the same utsname for all RPC requests for the life of an RPC
client. And since RPC credentials are cached, we should be more
careful about which cached credential is selected.
--
Chuck Lever
chuck[dot]lever[at]oracle[dot]com
--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
