We have the same issue, after a Karmic to Lucid upgrade we have the
same behavior with bacula-sd, libvirt-bin, nagios-nrpe-server, ntp and
snmp, the same for some static routes added through rc.local. And also
we have serious issues with VM's under KVM, the main problem seems to
be something about AppArmor.
We have not investigate the issue far from this point because it is a
On Fri, May 21, 2010 at 3:17 AM, Leander Janssen <firstname.lastname@example.org> wrote:
> Hi All,
> After upgrading my server from Karmic to Lucid a couple of services
> don't bind to their specified ip socket anymore during the boot process.
> The services that are affected are ssh, powerdns, iscsitarget, smb.
> When checking the log files the services complain that the ip socket
> they want to bind to does not exist.
> What these services all have in common is that I've configured
> a specific socket, instead of running the service on all interfaces and
> ip addresses (0.0.0.0 or ::
> Another detail is that I'm using multiple ip addresses on the same
> network interface. And that I'm using a bridge configuration to host
> some kvm guests. I've attached a copy of my /etc/network/interfaces.
> When I restart the services after boot they attach nicely to their
> configured socket. But this is a bit problematic on a headless server
> that is running in a remote datacenter.
> I assume some sort of timing issue with upstart where the network
> interface configration is not yet fully completed when other services
> are being started.
> Does anybody has an idea how I can further troubleshoot this issue?
> Kind Regards,
> Leander Janssen
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.10 (GNU/Linux)
> 6gNtDCjbP7etJCCS7DUTrmeGNgoeFqkuMxYggWFnACftrhGmeB KWaRBAtoPgBM6D
> hrDJhvSfF1H0FhdndbhwkmD+cRuJMA+O+eYFUgd5olZlN0Iimh bp2O05jGNRvO7/
> -----END PGP SIGNATURE-----
> ubuntu-server mailing list
> More info: https://wiki.ubuntu.com/ServerTeam
ubuntu-server mailing list
More info: https://wiki.ubuntu.com/ServerTeam