Re: [Devel] Re: swsusp done by migration (was Re: [RFC][PATCH 1/5]Virtualization/containers: startup)

From: Sam Vilain
Date: Sun Feb 12 2006 - 18:27:46 EST


Vasily Averin wrote:
When checkpointing it is important to preserve all state. If you are
doing transparent highly available computing, you need to make sure all
system calls get the same answers in the clones. So you would need to
virtualise the entropy pool.
From my point of view it is important to preserve only all the determinated state.
Ok, lets we've checkpointed and saved current entropy pool. But we have not any
guarantee that pool will be in the same state at the moment of first access to
it after wakeuping. Because a new entropy can change it unpredictable.
Am I right?

Good point.

In general this comes under the heading of "when doing highly available
computing, ensure each of your computing replicas get exactly the same
inputs and always double-check their outputs".

There are lots of 'inputs' that affect the state of the system; when I
referred to "/dev/random and the system clock", I was figuratively
referring to these. Others might include timing of disk IO events, for
instance.

I'd rather not discuss this too much; it was intended to be a flippant
"wishful thinking" thought, and is certainly not something I have spent
the necessary time investigating to discuss well :).

Sam.
-
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/