Re: The issues about architecture of the COLO checkpoint

[Dr. David Alan Gilbert]

* Lukas Straub (address@hidden) wrote:
> On Wed, 4 Mar 2020 07:44:11 +0000
> "Zhang, Chen" <address@hidden> wrote:
> 
> > > -----Original Message-----
> > > From: Daniel Cho <address@hidden>
> > > Sent: Monday, February 24, 2020 3:15 PM
> > > To: Zhang, Chen <address@hidden>
> > > Cc: Dr. David Alan Gilbert <address@hidden>; Zhanghailiang
> > > <address@hidden>; address@hidden; Jason
> > > Wang <address@hidden>
> > > Subject: Re: The issues about architecture of the COLO checkpoint
> > >
> > > Hi Zhang,
> > >
> > > Thanks for your help.
> > > However, did you occur the error which the function qemu_hexdump in
> > > colo-compare.c will crash the qemu process while doing operation with
> > > network?
> > >
> >
> > No, qemu_hexdump looks no relationship with network...
> > Do you means it will crashed in qemu_hexdump sometimes?
> 
> Jeah, I hit that bug too, but it was fixed with
> 1e907a32b77e5d418538453df5945242e43224fa "COLO-compare: Fix incorrect `if` 
> logic"
> in qemu 4.2.

Is anyone running any CI stuff for COLO?  I'm not, but bugs like that
make it clear we probably should be;  it's not trivial to setup, but it
should be doable in a nest setup these days.

Dave

> Regards,
> Lukas Straub
> 
> > > We are working on VM fault tolerance study and COLO function evalutation
> > > first. Currently we did not have a confirmed plan on it.
> >
> > OK, keep connection.
> >
> > Thanks
> > Zhang Chen
> 
--
Dr. David Alan Gilbert / address@hidden / Manchester, UK