Re: [Qemu-devel] [PATCH RFC 0/5] disk deadlines

[Stefan Hajnoczi <stefanha@redhat.com>]

On Tue, Sep 08, 2015 at 11:33:09AM +0200, Paolo Bonzini wrote:
> 
> 
> On 08/09/2015 10:00, Denis V. Lunev wrote:
> > How the given solution works?
> > 
> > If disk-deadlines option is enabled for a drive, one controls time completion
> > of this drive's requests. The method is as follows (further assume that this
> > option is enabled).
> > 
> > Every drive has its own red-black tree for keeping its requests.
> > Expiration time of the request is a key, cookie (as id of request) is an
> > appropriate node. Assume that every requests has 8 seconds to be completed.
> > If request was not accomplished in time for some reasons (server crash or smth
> > else), timer of this drive is fired and an appropriate callback requests to
> > stop Virtial Machine (VM).
> > 
> > VM remains stopped until all requests from the disk which caused VM's stopping
> > are completed. Furthermore, if there is another disks with 'disk-deadlines=on'
> > whose requests are waiting to be completed, do not start VM : wait completion
> > of all "late" requests from all disks.
> > 
> > Furthermore, all requests which caused VM stopping (or those that just were not
> > completed in time) could be printed using "info disk-deadlines" qemu monitor
> > option as follows:
> 
> This topic has come up several times in the past.
> 
> I agree that the current behavior is not great, but I am not sure that
> timeouts are safe.  For example, how is disk-deadlines=on different from
> NFS soft mounts?  The NFS man page says
> 
>      NB: A so-called "soft" timeout can cause silent data corruption in
>      certain cases.  As such, use the soft option only when client
>      responsiveness is more important than data integrity.  Using NFS
>      over TCP or increasing the value of the retrans option may
>      mitigate some of the risks of using the soft option.
> 
> Note how it only says "mitigate", not solve.

The risky part of "soft" mounts is probably that the client doesn't know
whether or not the request completed.  So it doesn't know the state of
the data on the server after a write request.  This is the classic
Byzantine fault tolerance problem in distributed systems.

This patch series pauses the guest like rerror=stop.  Therefore it's
different from NFS "soft" mounts, which are like rerror=report.

Guests running without this patch series may suffer from the NFS "soft"
mounts problem when they time out and give up on the I/O request just as
it actually completes on the server, leaving the data in a different
state than expected.

This patch series solves that problem by pausing the guest.  Action can
be taken on the host to bring storage back and resume (similar to
ENOSPC).

In order for this to work well, QEMU's timeout value must be shorter
than the guest's own timeout value.

Stefan