On Wed, Oct 13, 2010 at 10:18:42AM +0200, Guido Günther wrote: > Hi, > when using HAL instead of udev for nodeinfo we don't fail if HAL isn't > available. We also shouldn't fail if D-Bus is missing thenn either. > Attached patch fixes this. O.k. to apply? > Cheers, > -- Guido > >From aab94ae8d34e462cd937ada7367560389f79aa25 Mon Sep 17 00:00:00 2001 > From: =?UTF-8?q?Guido=20G=C3=BCnther?= <agx(a)sigxcpu.org&gt; > Date: Wed, 13 Oct 2010 09:07:48 +0200 > Subject: [PATCH] Don't fail on missing D-Bus > > We don't fail when we can't contact HAL so we shouldn't fail if we can't > contact D-Bus either. > --- > .gnulib | 2 +- > src/node_device/node_device_hal.c | 4 ++++ > 2 files changed, 5 insertions(+), 1 deletions(-) > > diff --git a/.gnulib b/.gnulib > index 2bb63bf..833bff3 160000 > --- a/.gnulib > +++ b/.gnulib > @@ -1 +1 @@ > -Subproject commit 2bb63bfb25474ea147ee9f1523c0337997359a4c > +Subproject commit 833bff3a531e4730779f8c7be1d9f7c471e6f221 > diff --git a/src/node_device/node_device_hal.c b/src/node_device/node_device_hal.c hum, that .gnulib change certainly should not be commited > index c2b3c8f..547e5ca 100644 > --- a/src/node_device/node_device_hal.c > +++ b/src/node_device/node_device_hal.c > @@ -724,6 +724,10 @@ static int halDeviceMonitorStartup(int privileged ATTRIBUTE_UNUSED) > dbus_conn = dbus_bus_get(DBUS_BUS_SYSTEM, &err); > if (dbus_conn == NULL) { > VIR_ERROR0(_("dbus_bus_get failed")); > + /* We don't want to show a fatal error here, > + otherwise entire libvirtd shuts down when > + D-Bus isn't running */ > + ret = 0; > goto failure; > } > dbus_connection_set_exit_on_disconnect(dbus_conn, FALSE); With your explanation, that makes sense, yes, ACK to that part