Hi Eric,
Here is my try to write a quick upstart script for virtlockd.
It should be named /etc/init/virtlockd.conf, then : ln -s /lib/init/upstart-job
/etc/init.d/virtlockd
It seems to work for me :
# virtlockd - Locking daemon for libvirt
description "virtlockd"
start on filesystem and runlevel [2345]
stop on starting rc RUNLEVEL=[016]
pre-start script
test -x /usr/sbin/virtlockd || { stop; exit 0; }
test -c /dev/null || { stop; exit 0; }
end script
script
start-stop-daemon --start --make-pidfile --pidfile /var/run/virtlockd.pid
--exec /usr/sbin/virtlockd >> /var/log/libvirt/virtlockd.log 2>&1
end script
-----Message d'origine-----
De : Eric Blake [mailto:eblake@redhat.com]
Envoyé : vendredi 11 octobre 2013 16:01
À : NEVEU Stephane; libvirt-users@redhat.com
Objet : Re: [libvirt-users] upstart script for virtlockd
On 10/11/2013 03:25 AM, NEVEU Stephane wrote:> Hi all,
>
> Trying to test libvirt 1.1.3 with virtlockd locking my qcow2 images on a
NFS storage between two kvm hosts.
> ./configure ... --with-init-script=upstart
>
> Libvirtd upstart script is actually well generated but I can't see
> nothing about virtlockd... or am I blind ? :)
It looks like virtlockd was added after Fedora had already moved over to
systemd, so no one has bothered to write an upstart init script for it yet.
Patches welcome.
> Nevertheless, running virtlockd -d && service libvirtd restart
works fine.
> Am I wrong thinking that editing qemu-lockd.conf with my needs and
qemu.conf with lock_manager="lockd" was enough ?
Yes, I think that's a reasonable workaround in the meantime.
--
Eric Blake eblake redhat com +1-919-301-3266
Libvirt virtualization library http://libvirt.org