The libvirt daemon was not started after installation because the
initscritp is named 'libvirtd' not like the package 'libvirt'.
The same problem appear in the uninstall.sh. The service was not
stopped.
Signed-off-by: Jonatan Schlag <jonatan.schlag@ipfire.org>
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
this id's are blacklisted in new cdc_ether module
because the r8152 module should used but the
3.14 module not know this id's.
Signed-off-by: Arne Fitzenreiter <arne_f@ipfire.org>
As a normal user, it is not possible to use qemu with KVM. This is bad
because it is better when it is possible to start the machine with a
less privileged user. To achieve this a group KVM is created and the
access to /dev/kvm is allowed for this group. So every user in this
group can use qemu with KVM.
This change is also useful for libvirt because the VMs can be started
with user nobody and group kvm.
Signed-off-by: Jonatan Schlag <jonatan.schlag@ipfire.org>
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
The if statement in line 89 and 99 are useless with the -e
conditional expression because it returns true if the path ist a
regular file or a directory.
So "/etc/init.d/ " returns true and "/etc/init.d/avahi" return also true,
but the statement should return only true if we have a regular file.
So -f if the right conditional expression, and we only try to execute
the init script if the path "/etc/init.d/${1}" points to a regular file.
Signed-off-by: Jonatan Schlag <jonatan.schlag@ipfire.org>
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
It is possible to communicate per ssh via a socket with libvirt. It is
not a good idea to do this as root, so the remote user is now
libvirt-remote. Only this user or users in the group libvirt-remote can
communicate with the socket.
The user libvirt-remote is created without a password. The users have to
set a password for this user after installation.
Signed-off-by: Jonatan Schlag <jonatan.schlag@ipfire.org>
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Coreutils 8.25 needs the automake 1.15
I have send this patch earlier.
Plaese merge automake before coreutils 8.25
Signed-off-by: Marcel Lorenz <marcel.lorenz@ipfire.org>
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
This update will pull in rpcbind as new dependency which
will automatically remove portmap when installed.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
The initscript now takes care that the squid proxy server process
is properly shut down. If that fails, it will remove the cache
index and let it be recreated at the next start. A warning is
shown to the user.
The "flush" command will now remove the entire proxy cache.
Signed-off-by: Matthias Fischer <matthias.fischer@ipfire.org>
Signed-off-by: Arne Fitzenreiter <arne.fitzenreiter@ipfire.org>
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Portmap is not maintained anymore that's why it is replaced by rpcbind.
Rpcbind provides also rpcinfo which is quite useful for debugging.
Signed-off-by: Jonatan Schlag <jonatan.schlag@ipfire.org>
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Update from 'dnsmasq 2.76test10' to '2.76test13'
Includes latest patches (001-006) from upstream.
Signed-off-by: Matthias Fischer <matthias.fischer@ipfire.org>
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
- Deleted a few absolutely unneeded tabs.
- Inserted/deleted some needed/unnecessary line breaks.
Signed-off-by: Matthias Fischer <matthias.fischer@ipfire.org>
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
the file update the bootdevice to uuid in fstab and grub1 config.
this is not needed since we use uuid at default and grub2
Signed-off-by: Arne Fitzenreiter <arne_f@ipfire.org>
The 'configure'-options were sorted (kind of) to get a better overview.
Added latest patches from upstream.
Changed '--enable-async-io=8' to '--enable-async-io=16' because of
http://www.squid-cache.org/mail-archive/squid-users/200705/0768.html :
"The default number of threads is dependent on the number of aufs
cache_dir lines, based on a reasonable estimate of how the code behaves."
Signed-off-by: Matthias Fischer <matthias.fischer@ipfire.org>
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
GRE connections were not correctly forwarded without the helper
being enabled. Choosing the wrong protocol here did not allow
us to load it properly.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
