[CRIU] Ghost file: no such file or directory

[Tycho Andersen]

On Wed, Mar 30, 2016 at 02:45:01PM +0200, Federico Reghenzani wrote:
> Also, I think that ghost-file-3.img is suspicious because is too big:
> ~140MB for a process that consumes ~4MB of ram, and the application does
> nothing with shared memory (it only initializes the open mpi module). Also
> the /tmp/openmpi-sessions-root at roaster-vm3_0 directory has a size of only
> 256KB.

But the ghost file means a file that has been deleted, so you won't
see it if you do du of that directory, and it won't be in the RAM of
the container.

Anyway, w.r.t. the core issue of not being able to open the ghost
file, I have had users report this to me as well, but have not been
able to reproduce it myself.

Tycho

> __
> Federico Reghenzani
> M.Eng. Student @ Politecnico di Milano
> Computer Science and Engineering
> 
> 
> 
> 2016-03-30 14:33 GMT+02:00 Federico Reghenzani <
> federico1.reghenzani at mail.polimi.it>:
> 
> > Hi all!
> >
> > We have problem restoring Open MPI daemons with child processes that uses
> > shared memory:
> >
> > (00.022447)    255: Opening ghost file 0x3 for
> > tmp/openmpi-sessions-root at roaster-vm3_0
> > /60995/1/shared_mem_pool.roaster-vm3.1
> > (00.022479)    255: Error (files-reg.c:139): Can't open ghost file
> > //tmp/openmpi-sessions-root at roaster-vm3_0/60995/1/shared_mem_pool.roaster-vm3.1.cr.3.ghost:
> > No such file or directory
> > (00.022558)  `- X [/dev/shm/open_mpi.0000.cr.1.ghost] ghost
> > (00.022570)  `- X [/dev/shm/open_mpi.0001.cr.2.ghost] ghost
> > (00.022586) Error (files-reg.c:445):  `- XFail
> > [/tmp/openmpi-sessions-root at roaster-vm3_0/60995/1/shared_mem_pool.roaster-vm3.1.cr.3.ghost]
> > ghost: No such file or directory
> > (00.022595) Error (files-reg.c:445):  `- XFail
> > [/tmp/openmpi-sessions-root at roaster-vm3_0/60995/1/shared_mem_btl_module.roaster-vm3.1.cr.4.ghost]
> > ghost: No such file or directory
> > (00.022613) Error (files-reg.c:445):  `- XFail
> > [/dev/shm/open_mpi.0000.cr.5.ghost] ghost: No such file or directory
> > (00.022622) Error (cr-restore.c:1995): Restoring FAILED.
> >
> > Full criu logs attached.
> >
> > We are migrating to another machine, in unshared context (CLONE_NEWPID |
> > CLONE_NEWNS), CRIU version 1.8.
> >
> >
> >
> > Any suggestions?
> > Cheers,
> > Federico
> > __
> > Federico Reghenzani
> > M.Eng. Student @ Politecnico di Milano
> > Computer Science and Engineering
> >
> >
> >
> >

> _______________________________________________
> CRIU mailing list
> CRIU at openvz.org
> https://lists.openvz.org/mailman/listinfo/criu