If device_model is set to '', xend will fail to execute the device
model. If None, xend will detect it and set a sensible qemu-dm path
instead.
Signed-off-by: Yosuke Iwamatsu <y-iwamatsu@ab.jp.nec.com>
I defined a cdrom to a guest domain as follows.
",hdc:cdrom,r"
When the guest domain was created by using xm create command, uname
of the cdrom was ''. Then creating the guest domain succeeded.
[2008-06-03 15:23:41 xend.XendDomainInfo 25956] DEBUG
(XendDomainInfo:200)
XendDomainInfo.create(['vm', ['name', 'rhel5.2ga_hvm_lvol1'],
<<snip>>
['device', ['vbd', ['uname', ''], ['dev', 'hdc:cdrom'], ['mode',
'r']]],
<<snip>>
But, when the guest domain was created by using virsh start command,
uname of the cdrom was None. Then creating the guest domain failed.
[2008-06-03 15:22:10 xend.XendDomainInfo 25559] DEBUG
(XendDomainInfo:200)
XendDomainInfo.create(['vm', ['name', 'rhel5.2ga_hvm_lvol1'],
<<snip>>
['device', ['vbd', ['dev', 'hdc:cdrom'], ['mode', 'r']]],
<<snip>>
This patch fixes to be able to create guest domains by both xm create
command and virsh start command.
FYI:
"if uname is None:" --> "if uname == '':" --> "if not uname:"
till CS#17625 current by this patch
Signed-off-by: Masaki Kanno <kanno.masaki@jp.fujitsu.com>
Domctl interface is extended to allow libxc retrieve device group
information from hypervisor. Vendor-specific iommu_ops is also
extended by adding a new operation "get_device_group_id()", which is
currently a null pointer but could be implemented later for vt-d.
Error will be raised from tools side when user trying to assign PCI
device with a sibling device being driven by dom0. User will keep
being prompted until he has hidden the entire device group (at least,
the sibling devices driven by dom0) in dom0 kernel
parameter. Hopefully this framework could be flexible enough to
support both amd iommu and vt-d.
The following 2 cases are not covered by this patch, but should be
easy to handle.
* Checking for hot-plug devices (maybe we can delay calling
ImageHandler.signalDeviceModel() until all checks are done?)
* Checking for splitted device group between different passthru
domains
Signed-off-by: Wei Wang <wei.wang2@amd.com>
This patch fix xend as when fatal error happened (e.g. qemu-dm process
was killed) log error message then mark that domain as crashed, do
what specified on crashed in the domain config file. Added some code in
xend to check those crashed hvm DM status each 30 seconds.
Signed-off-by: Xiaowei Hu <xiaowei.hu@oracle.com>
This patch removes unused resource parameter and fixes the default
value for domain_migrate. Also removes tab indent in XendOptions.py.
Signed-off-by: Zhigang Wang <zhigang.x.wang@oracle.com>
Added more informative error codes for device assignment domctls.
Introduced python bindings for device assignment and enable xend to
also perform IOMMU device assignment when assigning PCI devices.
Signed-off-by: Espen Skoglund <espen.skoglund@netronome.com>
This patch adds -s --ssl option to xm migrate. It will override
xend-relocation-ssl setting in /etc/xen/xend-config.sxp.
When mix deploy xen 3.2 and xen 3.3 servers, it's convenient to have a
command line option rather than modify /etc/xen/xend-config.sxp every
time.
Signed-off-by: Zhigang Wang <zhigang.x.wang@oracle.com>
This patch makes ssl relocation server listen on 8003 if enabled.
Whether to start ssl relocation server now controlled by
xend-relocation-ssl-server. So ssl and non-ssl relocation server can
run simultaneously. You can also only start ssl server or only start
non-ssl relocation server.
When mix deploy xen 3.2 server (has no ssl support) and 3.3 servers,
start ssl and non-ssl relocation server simultaneously can keep
backward compatibility.
It's also more reasonable to have separate ports for ssl and non-ssl.
In this patch, also renames xend-relocation-tls to xend-relocation-ssl.
Signed-off-by: Zhigang Wang <zhigang.x.wang@oracle.com>
The automatic NUMA placement code (find_relaxed_node) in
XendDomainInfo.py contains several bugs:
1. The domain just created shall not be considered in the load
calculation.
2. Nodes with too less memory for the domain must be considered during
the load calculation phase and abandoned later, currently these nodes
win with a load of 0.
3. CPUs which are offline shall not be considered in load calculation.
4. (Integer-)Dividing the load value by the number of cores per node
may conceal subtle load differences resulting in a non-optimal node to
be chosen.
Signed-off-by: Andre Przywara <andre.przywara@amd.com>
The device configurations of suspended virtual machines are
inaccessible even though for example the UUIDs of the VIFs and VBDs
are known and returned when VM.get_VIFs()/get_VBDs() is called, though
when trying to access their records for example, then handle is
reported as being invalid. This patch fixes this.
Signed-off-by: Stefan Berger <stefanb@us.ibm.com>
A domain fails to restore on IA64 since cset 17524:18727843db60.
This patch fixes it temporarily. Guest CPUID configuration for IA64
might be implemented in the future.
Signed-off-by: Kouya Shimura <kouya@jp.fujitsu.com>
The protocol field of the blkif interface is correct at startup for a
guest of a different mode from dom0 (eg. 32-bit dom0, 64-bit guest).
However, this property is not persisted on save, so a later restore
(or migrate) will setup the block interface with the wrong mode.
Signed-off-by: Herbert van den Bergh <herbert.van.den.bergh@oracle.com>
Signed-off-by: Kurt Hackel <kurt.hackel@oracle.com>
This is because function "uname = config.get('uname', '')" returns '',
rather than None, if uname is not in config variable.
Signed-off-by: Xu Dongxiao <dongxiao.xu@intel.com>
If we define wrong values to a disk parameter in domain configuration
files, we get an error message or a guest OS panic.
1. If we define a wrong disk type, xm create command error occurs
about 100 seconds later.
e.g. disk=['xyz:/xen/root-vm1.img,hda1,w']
2. If we forget a disk type, a guest OS panic occurs.
e.g. disk=['/xen/root-vm1.img,hda1,w']
Signed-off-by: Masaki Kanno <kanno.masaki@jp.fujitsu.com>
This has been broken since cset 16964:5d84464dc1fc
Also deal better with very early errors (close sender side socket)
Signed-off-by: Steven Hand <steven.hand@cl.cam.ac.uk>
If we define the form of list of string to "cpus" parameter,
different CPU affinities are set for each VCPU as follows.
# grep cpus /etc/xen/vm1
cpus = ["2", "3"]
vcpus = 2
# xm create vm1
Using config file "/etc/xen/vm1".
Started domain vm1
# xm vcpu-list vm1
Name ID VCPU CPU State Time(s)
CPU Affinity
vm1 1 0 2 r-- 3.5 2
vm1 1 1 3 -b- 3.2 3
If we define the form of string to "cpus" parameter as before,
a same CPU affinity is set for each VCPU as follows.
# grep cpus /etc/xen/vm2
cpus = "2,3"
vcpus = 2
# xm create vm2
Using config file "/etc/xen/vm2".
Started domain vm2
# xm vcpu-list vm2
Name ID VCPU CPU State Time(s)
CPU Affinity
vm2 2 0 3 -b- 3.0
2-3
vm2 2 1 2 r-- 2.6
2-3
Signed-off-by: Masaki Kanno <kanno.masaki@jp.fujitsu.com>
I am refactoring the security.on() call to return the actual type of
the security module that is found to be enabled rather than just
returning True or False.
Signed-off-by: Stefan Berger <stefanb@us.ibm.com>
Solaris uses a simple indexing scheme for block devices. Parts of xend
translate them as hexadecimal (such as block-attach), and decimal, or
unconverted, elsewhere (such as block-detach). Harmonise these
interfaces by allowing decimal specifications.
Also allow Solaris-style block device names.
Signed-off-by: John Levon <john.levon@sun.com>
- When we overwrite a cpuid's input, apply the default policy
for the other registers.
- For the python binding, get unsigned long instead long.
- Fix the multiple inputs cpuid's configuration parsing.
Signed-off-by: Jean Guyader <jean.guyader@eu.citrix.com>
Keir Fraser