Re: [RFC][PATCH 0/9] Network receive deadlock prevention for NBD
From: Evgeniy Polyakov
Date: Mon Aug 14 2006 - 02:52:56 EST
On Mon, Aug 14, 2006 at 08:45:43AM +0200, Peter Zijlstra (a.p.zijlstra@xxxxxxxxx) wrote:
> > Just for clarification - it will be completely impossible to login using
> > openssh or some other priveledge separation protocol to the machine due
> > to the nature of unix sockets. So you will be unable to manage your
> > storage system just because it is in OOM - it is not what is expected
> > from reliable system.
> > > But really, if you expect to run reliable block IO to Zanzibar over an ssh
> > > tunnel through a firewall, then you might also consider taking up bungie
> > > jumping with the cord tied to your neck.
> > Just pure openssh for control connection (admin should be able to
> > login).
> These periods of degenerated functionality should be short and
> infrequent albeit critical for machine recovery. Would you rather have a
> slower ssh login (the machine will recover) or drive/fly to Zanzibar to
> physically reboot the machine?
It will not work, since you can not mark openssh sockets as those which
are able to get memory from reserved pool. So admin unable to check the
system status and make anything to turn system's life on.
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/