<div dir="ltr">Ok, the problem was that the directory /tmp/openmpi...etc... is not present on the destination node.<div><br></div><div>Updating CRIU to 2.0 fixed the problem! :)<br></div></div><div class="gmail_extra"><br clear="all"><div><div class="gmail_signature"><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div>__</div><div><span style="font-size:12.8px">Federico Reghenzani</span></div><div><font size="1">M.Eng. Student @ Politecnico di Milano</font></div><div><span style="font-size:x-small">Computer Science and Engineering</span></div><div><br></div><div><br></div></div></div></div></div></div></div></div>
<br><div class="gmail_quote">2016-03-30 17:16 GMT+02:00 Tycho Andersen <span dir="ltr"><<a href="mailto:tycho.andersen@canonical.com" target="_blank">tycho.andersen@canonical.com</a>></span>:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div class="HOEnZb"><div class="h5">On Wed, Mar 30, 2016 at 06:15:20PM +0300, Pavel Emelyanov wrote:<br>
> On 03/30/2016 06:10 PM, Tycho Andersen wrote:<br>
> > On Wed, Mar 30, 2016 at 06:04:04PM +0300, Pavel Emelyanov wrote:<br>
> >> On 03/30/2016 05:59 PM, Tycho Andersen wrote:<br>
> >>> On Wed, Mar 30, 2016 at 05:50:32PM +0300, Pavel Emelyanov wrote:<br>
> >>>> On 03/30/2016 05:46 PM, Tycho Andersen wrote:<br>
> >>>>> On Wed, Mar 30, 2016 at 05:26:15PM +0300, Pavel Emelyanov wrote:<br>
> >>>>>> On 03/30/2016 03:33 PM, Federico Reghenzani wrote:<br>
> >>>>>>> Hi all!<br>
> >>>>>>><br>
> >>>>>>> We have problem restoring Open MPI daemons with child processes that uses shared memory:<br>
> >>>>>>><br>
> >>>>>>> (00.022447) 255: Opening ghost file 0x3 for tmp/openmpi-sessions-root@roaster-vm3_0/60995/1/shared_mem_pool.roaster-vm3.1<br>
> >>>>>>> (00.022479) 255: Error (files-reg.c:139): Can't open ghost file //tmp/openmpi-sessions-root@roaster-vm3_0/60995/1/shared_mem_pool.roaster-vm3.1.cr.3.ghost: No such file or directory<br>
> >>>>>><br>
> >>>>>> Can you check whether the \dirname of this path exists?<br>
> >>>>>> I mean this -- //tmp/openmpi-sessions-root@roaster-vm3_0/60995/1/<br>
> >>>>>><br>
> >>>>>> Presumably this is the case when not only the file was removed, but also<br>
> >>>>>> some dir components. And we've fixed it only in 2.0.<br>
> >>>>><br>
> >>>>> The users I had report it were using post 2.0, so there's some other<br>
> >>>>> bug here. I'm trying to reproduce now again, but not having any luck<br>
> >>>>> :(<br>
> >>>><br>
> >>>> Do they also see the ENOENT errno from the open(O_CREAT) call?<br>
> >>><br>
> >>> Which call do you mean here? I don't see anything close to this that<br>
> >>> does an O_CREAT on dump.<br>
> >><br>
> >> Yes, because the error is on restore :) For 1.8 this is files-reg.c create_ghost()'s<br>
> ><br>
> > Oh, derp, I misread the error. I've seen failures reported on dump<br>
> > here:<br>
> ><br>
> > <a href="https://github.com/xemul/criu/blob/master/criu/files-reg.c#L628" rel="noreferrer" target="_blank">https://github.com/xemul/criu/blob/master/criu/files-reg.c#L628</a><br>
><br>
> Ouch! And what was the errno?! I can hardly imagine the reason for a regular<br>
> file open failure via proc :(<br>
<br>
</div></div>ENOENT, yeah, I was confused as well, and haven't been able to<br>
reproduce it either :(<br>
<span class="HOEnZb"><font color="#888888"><br>
Tycho<br>
</font></span></blockquote></div><br></div>