Re: [PATCH] [RFC] netfilter: nf_conntrack: don't relase a conntrackwith non-zero refcnt

[Florian Westphal]

Andrew Vagin <avagin@xxxxxxxxxxxxx> wrote:
> > I think it would be nice if we could keep it that way.
> > If everything fails we could proably intoduce a 'larval' dummy list
> > similar to the one used by template conntracks?
> 
> I'm not sure, that this is required. Could you elaborate when this can
> be useful?

You can dump the lists via ctnetlink.  Its meant as a debugging aid in
case one suspects refcnt leaks.

Granted, in this situation there should be no leak since we put the newly
allocated entry in the error case.

> Now I see only overhead, because we need to take the nf_conntrack_lock
> lock to add conntrack in a list.

True. I don't have any preference, I guess I'd just do the insertion into the
unconfirmed list when we know we cannot track to keep the "unhashed"
bug trap in the destroy function.

Pablo, any preference?
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/