Re: NFS/credentials leak in 2.6.29-rc1

["J. Bruce Fields" <bfields@fieldses.org>]

On Tue, Jan 27, 2009 at 12:26:59PM +0300, Evgeniy Polyakov wrote:
> On Mon, Jan 26, 2009 at 07:49:55PM -0500, J. Bruce Fields (bfields@fieldses.org) wrote:
> > > With additional put_cred, i.e.:
> > > 
> > >         put_cred(override_creds(new));
> > >         put_cred(new);
> > >         return 0;
> > > 
> > > I got following fun tcpdump and failed mount (it stuck, but can be interrupted):
> > 
> > I'm not able to reproduce this.
> > 
> > Have you been able to reproduce this consistently?  Any further results?
> 
> Yes, I can easily reproduce it with the tree I work with.
> But I will update to the latest git soon (likely today) and rerun the
> tests, so it may happen that rc1-rc2 frame fixed the issue at least
> partially. I can also turn on some debugging options if needed.

I took a closer look at the tcpdump output.  Those numbers aren't port
numbers, they're rpc xid's: see "NFS Requests and Replies" in the
"OUTPUT FORMAT" section of the tcpdump man page.  Or compare the output
of tcpdump with wireshark's output.

That still doesn't explain why you were seeing a mount hang--but some
network or other configuration problem may be the most likely
explanation.

--b.