[Gluster-devel] Glusterfs 3.7.11 with LibGFApi in Qemu on Ubuntu Xenial does not work

André Bauer abauer at magix.net
Wed Jun 15 09:11:13 UTC 2016

Hi Lists,

i just updated on of my Ubuntu KVM Servers from 14.04 (Trusty) to 16.06

I use the Glusterfs packages from the officail Ubuntu PPA and my own
Qemu packages (
https://launchpad.net/~monotek/+archive/ubuntu/qemu-glusterfs-3.7 )
which have libgfapi enabled.

On Ubuntu 14.04 everything is working fine. I only had to add the
following lines to the Apparmor config in
/etc/apparmor.d/abstractions/libvirt-qemu to get it work:

# for glusterfs
/proc/sys/net/ipv4/ip_local_reserved_ports r,
/usr/lib/@{multiarch}/glusterfs/**.so mr,
/tmp/** rw,

In Ubuntu 16.04 i'm not able to start the my VMs via libvirt or to
create new images via qemu-img using libgfapi.

Mounting the volume via fuse does work without problems.


qemu-img create gluster://storage.mydomain/vmimages/kvm2test.img 1G
Formatting 'gluster://storage.intdmz.h1.mdd/vmimages/kvm2test.img',
fmt=raw size=1073741824
[2016-06-15 08:15:26.710665] E [MSGID: 108006]
[afr-common.c:4046:afr_notify] 0-vmimages-replicate-0: All subvolumes
are down. Going offline until atleast one of them comes back up.
[2016-06-15 08:15:26.710736] E [MSGID: 108006]
[afr-common.c:4046:afr_notify] 0-vmimages-replicate-1: All subvolumes
are down. Going offline until atleast one of them comes back up.

Libvirtd log:

[2016-06-13 16:53:57.055113] E [MSGID: 104007]
[glfs-mgmt.c:637:glfs_mgmt_getspec_cbk] 0-glfs-mgmt: failed to fetch
volume file (key:vmimages) [Invalid argument]
[2016-06-13 16:53:57.055196] E [MSGID: 104024]
[glfs-mgmt.c:738:mgmt_rpc_notify] 0-glfs-mgmt: failed to connect with
remote-host: storage.intdmz.h1.mdd (Permission denied) [Permission denied]
2016-06-13T16:53:58.049945Z qemu-system-x86_64: -drive
Gluster connection failed for server=storage.intdmz.h1.mdd port=0
volume=vmimages image=checkbox.qcow2 transport=tcp: Permission denied

I don't see anything in the apparmor logs when setting everything to
complain or audit.

It also seems GlusterFS servers don't get any request because brick logs
are not complaining anything.

Any hints?

André Bauer

More information about the Gluster-devel mailing list