In various places in documentation and code, IP addresses are provided
as examples, defaults, or dummy configuration. In general the
specific IP addresses used in Xen are not always appropriate. (For
example, 1.2.3.4 is used in a few places!)
The following addresses should be used:
* For examples and documentation, 192.0.2.0/24. (See RFC3330.)
* For defaults for private networks, a random network from RFC1918.
I have randomly selected 172.30.206.0/24 for this purpose and
documented this in at the only registry I know of,
www.ucam.org/cam-grin. This network should henceforth be used for
default configurations of local bridges, test networks, etc. in
Xen tools.
The following addresses should NOT be used:
* 10.0.*.*, 10.1.*.*, 192.168.0.*, 192.168.1.*, etc. Using these
addresses gives greatly increased likelihood of collision, as
ignorant network administrators and reckless middlebox vendors
often pick networks from the bottom of 10/8 and 192.168/16.
* 169.254.*.*. These are reserved for zeroconf (ad-hoc networking)
and should not be used for Xen private networks, bridges, etc.,
etc. Use of these addresses by Xen scripts causes trouble on hosts
(eg laptops) which find themselves in ad-hoc networking
environments. I think this is not hypothetical (!) since at least
one Linux distribution have specific code to detect this case and
cause Xen startup to fail iff the host already has an external
zeroconf address.
* 1.2.3.4. WTF !?
I have also used 127.0.255.255 in one place where apparently a dummy
address is needed (some Linux kernels won't accept a lack of an NFS
server address). If 127.0.255.255 is mistakenly used it is unlikely
to do any damage to real traffic even if it does escape into the
network at large.
Signed-off-by: Ian Jackson <ian.jackson@eu.citrix.com>
This fixes two problems related to the SR class:
- get_PBDs() was not implemented
- get_physical_size did not work
Signed-off-by: Stefan Berger <stefanb@us.ibm.com>
xend can get into a situation where two processes are attempting to
interact with the xenstore socket, with disastrous results. Fix the
two bad users of xstransact, add a big warning, and fix the destructor
so future mistakes will be detected earlier.
Signed-off-by: John Levon <john.levon@sun.com>
Xend doesn't correctly work after restart, when there is a domU which
owns a pci device (driver domain). This patch fixes the problem.
Signed-off-by: Yosuke Iwamatsu <y-iwamatsu@ab.jp.nec.com>
With current XenD 3.0.4 or later try the following:
brctl addbr demo
ifconfig demo up
/etc/init.d/xend start
/etc/init.d/xend stop
ifconfig demo down
brctl delbr demo
Now, start XenD again....
/etc/init.d/xend start
And watch in horror as it re-creates your 'demo' bridge.
The problem is that the 'XendNetwork' class does not distinguish
between bridge devices that it is managing (ie those created via
XenAPI) and those which it does not manage (ie those created by OS
distro init scripts, or by apps like libvirt).
While initially I thought I could just make XenD ignore
externally-managed bridges completely, it seems to needs to know about
them otherwise it can't hook up guest VIFs to them correctly. So the
attached patch adds a 'managed' flag to the XendNetwork
class. Externally managed bridges have this set to False. At startup
XenD will now only re-create bridge devices which have the 'managed'
flag set to 'True' - ie those created via XenAPI.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Instead of starting a thread per script, run a single thread and send
orders to it. This serializes the execution of the scripts.
Signed-off-by: Stefan Berger <stefanB@us.ibm.com>
I am implementing some missing interface functions in the XSM dummy
module. Also deactivating some debugging output.
Signed-off-by: Stefan Berger <stefanb@us.ibm.com>
When unmanaged domains with on_xend_stop="suspend" exist, managed
domains are not suspended by xend stop command. The processing of
cleanup_domains() is aborted by exception. This patch keeps suspending
managed domains even if unmanaged domains exist.
Signed-off-by: Masaki Kanno <kanno.masaki@jp.fujitsu.com>
Distutils won't install xsview unless it has an __init__.py. This
patch adds an empty one, which is all that is required.
Signed-off-by: Mark Williamson <mark.williamson@cl.cam.ac.uk>
This makes no difference to the visible behaviour of "xm top" but
avoids the whole of xm from quitting if you're in "xm shell". This
provides the more intuitive behaviour of returning to the xm shell
interface after leaving xm top.
Signed-off-by: Mark Williamson <mark.williamson@cl.cam.ac.uk>
Some relics in the control tools still refer to the USB paravirt
driver. Remove this as it's dead code and is in the history for
anyone really interested. The Qemu device model has its own USB
configuration, which should still work.
Signed-off-by: Mark Williamson <mark.williamson@cl.cam.ac.uk>
This patch extends the Xen-API and the legacy XML-RPC interface with a
function to reset the policy on the system (through an update with the
default policy). I adapted the code in 'xm resetpolicy' to use this
now.
This patch also extends libxen and the documentation to reflect the
new function.
Signed-off-by: Stefan Berger <stefanb@us.ibm.com>
We can give multiple domains to the following commands.
But, the help messages of the commands are explained to be able
to give a single domain.
Signed-off-by: Masaki Kanno <kanno.masaki@jp.fujitsu.com>
Recover the system if the XML representation of the currently loaded
policy is missing. Force the installation of the DEFAULT policy.
Signed-off-by: Stefan Berger <stefanb@us.ibm.com>
This patch moves the directory of files where xend is writing policies
and resource labels into to /var/lib/xend/security/policies.
Signed-off-by: Stefan Berger <stefanb@us.ibm.com>
This patch implements a (non Xen-API) legacy XML-RPC interface for the
ACM commands and funnels the calls into code introduced by the Xen-API
support for ACM security management. Since some of the functionality
has changed, also the xm applications have changed. In particular the
following old commands have been removed along with some tools the
have become obsolete now:
- loadpolicy (included in: setpolicy)
- makepolicy (included in: setpolicy)
- cfgbootpolicy (included in: setpolicy)
and the following commands been introduced:
- setpolicy
- getpolicy
- resetpolicy
All tools have been adapted to work in Xen-API and legacy XML-RPC
mode. Both modes support the same functionality.
Signed-off-by: Stefan Berger <stefanb@us.ibm.com>
Although XenAPI c-bindings support this operation on Network class,
there is no implementation in xend. This patch provides one.
Signed-off-by: Jim Fehlig <jfehlig@novell.com>
This makes use of the domain config option guest_os_type for
ia64 and removes the backing for the previous ACPI based
mechanism used previously. A user wanting optimal performance
for a specific type of OS guest running in an HVM domain should
make use of this new option. See updated xmexmaple.vti for
available options. All supported OSes should always work using
the default option or leaving the option unspecified. Originally
based on patch from Zhang Xin.
Signed-off-by: Alex Williamson <alex.williamson@hp.com>
This is an optional domain config entry that may be used by
architecture specific builder code to setup various optimizations
based on the guest OS intended to run in the domain. Based on
patch from Zhang Xin.
Signed-off-by: Alex Williamson <alex.williamson@hp.com>
Acked-by: Keir Fraser <keir.fraser@citrix.com>
This patch fixes the case where a module line is supposed to be added
to the very end of the file but the file does not end in with a new
line. Also fixes a problem that in some cases the module line would
not be properly be removed.
Signed-off-by: Stefan Berger <stefanb@us.ibm.com>
Long options of xm sched-credit command are ignored (--domain,
--weight, --cap). This patch fixes it.
Signed-off-by: Masaki Kanno <kanno.masaki@jp.fujitsu.com>
This patch adds support to XenD for configuring the previously added
TLS encryption and x509 certificate validation. At this time I have
only enabled this config to be done system-wide via
/etc/xen/xend-config.sxp. Since it requires the admin to add
certificates on the local FS, there's not much point in making it per
VM. The x509 certificates are located in /etc/xen/vnc. Since this
requires a special VNC client program (GTK-VNC,
virt-viewer/virt-manager or VeNCrypt viewer) the use of TLS is
disabled by default. Admins can enable it if they are using a suitable
client.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
This patch adds in the upstream QEMU VNC authentication code. This
spports the previous VNC password auth scheme, as well as the VeNCrypt
protocol extension. The latter allows for performing a TLS handshake,
and client verification of the server identify using x509
certificates. It is also possible for the server to request a client
certificate and validate that as a simple auth scheme. The code
depends on GNU TLS for SSL APIs, and the configure script will
auto-detect this.
The image.py code is changed to deal with the new syntax for the -vnc
option. In particular password authentication is now enabled by
giving the 'password' flag,
eg -vnc 0.0.0:1,password
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
In summary, this allows a xenstore transaction object to be passed
around the various device controllers, so that they don't have to do
lots of singleton transactions. Transactions have very heavy I/O
impact from xenstored so reducing their number is important.
When running 3 guests, this patch reduces the impact of 'xm list
--long' from 176 transactions, scaling O(n) with guests, to 26
transactions with O(1) scaling.
I have previously attempted to also address the same issue with 'xm
create' but that's much harder since the device front/back handshake
requires that XenD use a number of small transactions. So i've not
changed anything here.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
This patch triggers a script when a resource's label changes. The xend
config file should provide a variable 'resource-label-change-script'
that can then be launched.
Signed-off-by; Stefan Berger <stefanb@us.ibm.com>
In HVM config file:
timer_mode=0 # Default: virtual time is delayed when timer ticks are
# missed dur to preemption
timer_mode=1 # Virtual time always equals wall time, even while missed
# ticks are pending
From: Haitao Shan <haitao.shan@intel.com>
Signed-off-by: Keir Fraser <keir@xensource.com>
Keir Fraser
Alex Williamson
Keir Fraser