John Ferlan wrote:
On 10/29/2014 02:20 PM, Roman Bogorodskiy wrote:
Currently, build fails on FreeBSD because its struct ifreq does not
have ifr_hwaddr member. In order to fix that, check if this member
is present, otherwise fall back to the stub version of the
virNetDev{Add,Del}Multi
John Ferlan wrote:
On 10/29/2014 02:20 PM, Roman Bogorodskiy wrote:
The virGetSCSIHostNumber function return type is int, however
its stubbed version returns NULL. That results in a build fail
on systems that uses the stubbed version. Fix by using a proper
return type.
---
In subject: I've added virsh: designator to make clear which part of
the code the patch touches.
On 10/30/14 03:35, Luyao Huang wrote:
After use cidr_format in function virAsprintf and vshPrintExtra, need free
cidr_format.
And clarified this sentence a bit.
Fix the following memory leak
On 10/30/14 03:35, Luyao Huang wrote:
After use cidr_format in function virAsprintf and vshPrintExtra, need free
cidr_format.
Fix the following memory leak from valgrind, like:
18 bytes in 1 blocks are definitely lost in loss record 41 of 192
at 0x4C29BBD: malloc (in
Conrad Meyer wrote:
This enables booting interactive GRUB menus (e.g. install CDs) with
libvirt-bhyve.
Caveat: A terminal other than the '--console' option to 'virsh start'
(e.g. 'cu -l /dev/nmdm0B -s 115200') must be used to connect to
grub-bhyve because the bhyve loader path is
On 10/30/2014 02:27 PM, Peter Krempa wrote:
On 10/30/14 03:35, Luyao Huang wrote:
After use cidr_format in function virAsprintf and vshPrintExtra, need free
cidr_format.
Fix the following memory leak from valgrind, like:
18 bytes in 1 blocks are definitely lost in loss record 41 of 192
Particularly in qemuBuildNumaArgStr(), there was a need for the advice
due to memory backing, which needs to know the nodeset it will be pinned
to. With newer qemu this caused the following error when starting
domain:
error: internal error: Advice from numad is needed in case of
automatic
On Thu, Oct 30, 2014 at 02:23:00AM +, Chen, Fan wrote:
On Wed, 2014-10-29 at 14:20 +0100, Martin Kletzander wrote:
On Wed, Oct 29, 2014 at 08:33:32PM +0800, Chen Fan wrote:
diff --git a/src/util/virnuma.c b/src/util/virnuma.c
@@ -373,6 +400,12 @@ virNumaGetNodeCPUs(int node
On Wed, Oct 29, 2014 at 04:18:42PM -0600, Eric Blake wrote:
On 10/29/2014 10:35 AM, Martin Kletzander wrote:
It makes sense for none of the callers to have negative value as an
output and, fortunately, if anyone tried defining domain with negative
memory or any other value parsed by
We were missing check for the fact that the storage driver was found and
in case there is no vbox storage driver available, daemon raised the
following error each start:
error : virRegisterStorageDriver:592 : driver in
virRegisterStorageDriver must not be NULL
Fixing this makes the condition
On 10/30/14 10:03, Martin Kletzander wrote:
We were missing check for the fact that the storage driver was found and
in case there is no vbox storage driver available, daemon raised the
following error each start:
error : virRegisterStorageDriver:592 : driver in
virRegisterStorageDriver
On 10/12/14 14:12, Martin Kletzander wrote:
This function is used to cleanup a pidfile doing whatever it takes, even
killing the owning process.
Signed-off-by: Martin Kletzander mklet...@redhat.com
---
v2:
- Don't use /proc, but simply just try to acquire the pidfile.
-
On 10/12/14 14:12, Martin Kletzander wrote:
When daemon is killed right in the middle of probing a qemu binary for
its capabilities, the VM is left running. Next time the daemon is
s/VM/qemu process/ ? The qemu isn't running anything so it might confuse
someone.
starting, it cannot start
On 10/29/14 14:31, Serge Hallyn wrote:
Quoting Cédric Bosdonnat (cbosdon...@suse.com):
Without this patch, kvm and kqemu domains confined with apparmor can't start
due to virt-aa-helper not finding TEMPLATE.kvm or TEMPLATE.kqemu. This patch
points all kvm-related drivers to TEMPLATE.qemu.
On 10/30/14 10:47, Peter Krempa wrote:
On 10/29/14 14:31, Serge Hallyn wrote:
Quoting Cédric Bosdonnat (cbosdon...@suse.com):
Without this patch, kvm and kqemu domains confined with apparmor can't start
due to virt-aa-helper not finding TEMPLATE.kvm or TEMPLATE.kqemu. This patch
points all
On 10/30/14 01:15, Lyssa P. Livingston wrote:
I have a file with an ceph-based backing store. I run this command:
nova boot vm-clone
where vm-clone is the rbd-backed file. It generates this command:
qemu-img create -f qcow2 -b
On Sat, Oct 25, 2014 at 12:43:35PM +0200, Wouter Verhelst wrote:
I haven't seen a reply to this anymore. Do people still have comments?
I'm planning on doing a release of nbd later this weekend, and would
like to include this (not the TLS implementation yet, but at least the
spec)
Hi Wouter,
https://bugzilla.redhat.com/show_bug.cgi?id=1155441
---
src/qemu/qemu_hotplug.c | 13 -
1 file changed, 12 insertions(+), 1 deletion(-)
diff --git a/src/qemu/qemu_hotplug.c b/src/qemu/qemu_hotplug.c
index 33241fb..d3bf392 100644
--- a/src/qemu/qemu_hotplug.c
+++
On Thu, Oct 30, 2014 at 2:29 AM, Roman Bogorodskiy
bogorods...@gmail.com wrote:
Conrad Meyer wrote:
This enables booting interactive GRUB menus (e.g. install CDs) with
libvirt-bhyve.
Caveat: A terminal other than the '--console' option to 'virsh start'
(e.g. 'cu -l /dev/nmdm0B -s 115200')
On 10/23/2014 09:46 PM, Taowei Luo wrote:
The API on IHardDiskAttachment is merged into IMediumAttachment.
So, we don't need it anymore.
---
src/vbox/vbox_storage.c | 160
src/vbox/vbox_tmpl.c | 202
On 10/30/2014 06:38 AM, Ján Tomko wrote:
https://bugzilla.redhat.com/show_bug.cgi?id=1155441
---
src/qemu/qemu_hotplug.c | 13 -
1 file changed, 12 insertions(+), 1 deletion(-)
ACK. Safe during the freeze.
diff --git a/src/qemu/qemu_hotplug.c b/src/qemu/qemu_hotplug.c
index
When the list of domains is fetched and being printed, but in the
meantime one domain was undefined before its status was fetched, the
output then includes domain with no state. With this patch, such
domain is skipped over as consecutive 'virsh list --all' (or the same
one ran a second later)
When one domain is being undefined and at the same time started, for
example, there is a possibility of a rare problem occuring.
- Thread 1 does virDomainUndefine(), has the lock, checks that the
domain is active and because it's not, calls
virDomainObjListRemove().
- Thread 2 does
First patch just fixes 'virsh list' for a rare case that is similar to
the one fixed in the third patch. Second patch is just a helper for
reproducing the race.
This was found in qemu driver but is fixed on upper level, so it
should be more general.
Martin Kletzander (3):
virsh: don't list
To reproduce the problem that this series is trying to fix, do the
following:
- Term 1:
LIBVIRT_DEBUG=2 LIBVIRT_LOG_OUTPUTS='2:stderr' daemon/libvirtd
- Term 2:
virsh undefine $dom
- In term 1 look for:
NOW!!!
- Term 3:
virsh start $dom
- Enjoy:
domain is started, but not
Patch 43b67f2e disallowed network tuning only with qemu driver, however
this patch moved the check for root privileges into
virNetDevBandwidthSetInternal wrapper function, so the call should now
fail in all possible cases. The reason to create another wrapper is that
we do execute a test suite
Since there was a valid note to patch 43b67f2e about the best spot to
check for bandwidth set call while having libvirt daemon run in session
mode, this patch reverts previous changes dealing with bandwith
(excluding NUMA!) in qemu_driver.c and qemu_command.c. There will be
another patch in the
As Laine noted correctly, we should be testing for privileges on a global scale,
when trying to set bandwidth parameters. Therefore I moved the check from
qemu_driver.c
and qemu_command.c into virnetdevbandwidth.c
Erik Skultety (2):
qemu: revert patch - bandwidth tuning in session mode
Michal Privoznik wrote:
On 11.10.2014 00:03, Jim Fehlig wrote:
V2 of a small series to improve libxl driver's support for
user-specified
config
https://www.redhat.com/archives/libvir-list/2014-September/msg01243.html
Patches 1-4 of that series have been pushed. In patch 5, Michal
The value is stored in unsigned long long, so ULLONG_MAX is the proper
upper limit to use.
Signed-off-by: Martin Kletzander mklet...@redhat.com
---
Even though this is a build-breaker (for 32bit systems
memtune-unlimited fails to parse), I'm not pushing it as one because
it feels odd that such
This enables booting interactive GRUB menus (e.g. install CDs) with
libvirt-bhyve.
Caveat: A terminal other than the '--console' option to 'virsh start'
(e.g. 'cu -l /dev/nmdm0B -s 115200') must be used to connect to
grub-bhyve because the bhyve loader path is synchronous and must occur
before
On 10/30/2014 09:49 AM, Martin Kletzander wrote:
The value is stored in unsigned long long, so ULLONG_MAX is the proper
upper limit to use.
No, it's not.
Signed-off-by: Martin Kletzander mklet...@redhat.com
---
Even though this is a build-breaker (for 32bit systems
memtune-unlimited
Additionally, make the bootloader tag optional (for bhyveload with
custom arguments) (also, matches the actual parser).
---
docs/schemas/domaincommon.rng | 17 +++--
1 file changed, 11 insertions(+), 6 deletions(-)
diff --git a/docs/schemas/domaincommon.rng
---
src/bhyve/bhyve_capabilities.c | 37 +
src/bhyve/bhyve_capabilities.h | 3 +++
2 files changed, 40 insertions(+)
diff --git a/src/bhyve/bhyve_capabilities.c b/src/bhyve/bhyve_capabilities.c
index 132ce91..6e9a943 100644
---
We still default to bhyveloader(1) if no explicit bootloader
configuration is supplied in the domain.
If the /domain/bootloader looks like grub-bhyve and the user doesn't
supply /domain/bootloader_args, we make an intelligent guess and try
chainloading the first partition on the disk (or a CD if
Drvbhyve hardcodes bhyveload(8) as the host bootloader for guests.
bhyveload(8) loader only supports FreeBSD guests.
This patch series adds bootloader and bootloader_args handling to
bhyve_command, so libvirt can boot non-FreeBSD guests in Bhyve.
Additionally, support for grub-bhyve(1)'s
---
.../bhyvexml2argv-bhyveload-explicitargs.args | 3 +++
.../bhyvexml2argv-bhyveload-explicitargs.ldargs| 1 +
.../bhyvexml2argv-bhyveload-explicitargs.xml | 23 +
.../bhyvexml2argv-custom-loader.args | 3 +++
---
.../bhyvexml2argv-acpiapic.ldargs | 1 +
tests/bhyvexml2argvdata/bhyvexml2argv-base.ldargs | 1 +
.../bhyvexml2argvdata/bhyvexml2argv-console.ldargs | 1 +
.../bhyvexml2argv-disk-cdrom.ldargs| 1 +
.../bhyvexml2argv-disk-virtio.ldargs | 1 +
---
.../bhyvexml2argv-serial-grub-nocons.args | 4
.../bhyvexml2argv-serial-grub-nocons.devmap| 1 +
.../bhyvexml2argv-serial-grub-nocons.ldargs| 2 ++
.../bhyvexml2argv-serial-grub-nocons.xml | 26 ++
On Thu, Oct 30, 2014 at 10:42:05AM +0100, Peter Krempa wrote:
On 10/12/14 14:12, Martin Kletzander wrote:
When daemon is killed right in the middle of probing a qemu binary for
its capabilities, the VM is left running. Next time the daemon is
s/VM/qemu process/ ? The qemu isn't running
On 10/30/2014 09:55 AM, Eric Blake wrote:
On 10/30/2014 09:49 AM, Martin Kletzander wrote:
The value is stored in unsigned long long, so ULLONG_MAX is the proper
upper limit to use.
No, it's not.
Signed-off-by: Martin Kletzander mklet...@redhat.com
---
Even though this is a
---
src/lxc/lxc_container.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
Pushed as trivial.
diff --git a/src/lxc/lxc_container.c b/src/lxc/lxc_container.c
index 2af2674..f02b959 100644
--- a/src/lxc/lxc_container.c
+++ b/src/lxc/lxc_container.c
@@ -281,7 +281,7 @@ static int
On 10/30/2014 10:09 AM, Eric Blake wrote:
On 10/30/2014 09:55 AM, Eric Blake wrote:
On 10/30/2014 09:49 AM, Martin Kletzander wrote:
The value is stored in unsigned long long, so ULLONG_MAX is the proper
upper limit to use.
No, it's not.
Signed-off-by: Martin Kletzander
Jim Fehlig wrote:
Michal Privoznik wrote:
On 11.10.2014 00:03, Jim Fehlig wrote:
V2 of a small series to improve libxl driver's support for
user-specified
config
https://www.redhat.com/archives/libvir-list/2014-September/msg01243.html
Patches 1-4 of that series have been pushed.
Thanks. I’ll be looking for the patch.
Lyssa
On Oct 30, 2014, at 03:03, Peter Krempa pkre...@redhat.com wrote:
On 10/30/14 01:15, Lyssa P. Livingston wrote:
I have a file with an ceph-based backing store. I run this command:
nova boot vm-clone
On 10/30/2014 10:23 AM, Eric Blake wrote:
Or maybe the problem is that at some point we used unsigned long, and
later moved to unsigned long long, but never updated the comment? I'm
trying to investigate the history of this code...
Looking a bit deeper, commit 4888f0fb5 was where we changed
On 29.10.2014 16:44, Martin Kletzander wrote:
There was one message used for all parsing errors in the blkio code.
This series fixes it in both qemu and lxc drivers where the code is
used.
Martin Kletzander (2):
qemu: improve error message for invalid blkiotune settings
lxc: improve error
On 30.10.2014 07:40, Martin Kletzander wrote:
Particularly in qemuBuildNumaArgStr(), there was a need for the advice
due to memory backing, which needs to know the nodeset it will be pinned
to. With newer qemu this caused the following error when starting
domain:
error: internal error:
On 30.10.2014 16:04, Martin Kletzander wrote:
When one domain is being undefined and at the same time started, for
example, there is a possibility of a rare problem occuring.
- Thread 1 does virDomainUndefine(), has the lock, checks that the
domain is active and because it's not, calls
On 30.10.2014 16:04, Martin Kletzander wrote:
When the list of domains is fetched and being printed, but in the
meantime one domain was undefined before its status was fetched, the
output then includes domain with no state. With this patch, such
domain is skipped over as consecutive 'virsh list
On Thu, Oct 30, 2014 at 06:39:56PM +0100, Michal Privoznik wrote:
On 30.10.2014 16:04, Martin Kletzander wrote:
When one domain is being undefined and at the same time started, for
example, there is a possibility of a rare problem occuring.
- Thread 1 does virDomainUndefine(), has the
On 30.10.2014 16:14, Erik Skultety wrote:
Since there was a valid note to patch 43b67f2e about the best spot to
check for bandwidth set call while having libvirt daemon run in session
mode, this patch reverts previous changes dealing with bandwith
(excluding NUMA!) in qemu_driver.c and
On 30.10.2014 16:14, Erik Skultety wrote:
Patch 43b67f2e disallowed network tuning only with qemu driver, however
this patch moved the check for root privileges into
virNetDevBandwidthSetInternal wrapper function, so the call should now
fail in all possible cases. The reason to create another
On 10/30/2014 10:51 AM, Eric Blake wrote:
On 10/30/2014 10:23 AM, Eric Blake wrote:
Or maybe the problem is that at some point we used unsigned long, and
later moved to unsigned long long, but never updated the comment? I'm
trying to investigate the history of this code...
Or maybe the
Commit 6c9a8a4 (Oct 2014) exposed a long-standing issue on 32-bit
machines: code related to virDomainSetMemoryParameters has always
been documented as using a 64-bit limit, but it was implemented by
calling virDomainParseMemory which enforced an 'unsigned long'
limit. Since
Hi,
I'm looking into why virt-aa-helper isn't adding allow rules for
backing stores nested deeper than 1. So if I do
qemu-img create -f qcow2 l1.img 10G
qemu-img create -f qcow2 -b l1.img l2.img
and use l2.img in a domain, then virt-aa-helper will add allow
rules for the domain to access both
On 10/30/2014 02:32 PM, Serge Hallyn wrote:
Hi,
I'm looking into why virt-aa-helper isn't adding allow rules for
backing stores nested deeper than 1. So if I do
qemu-img create -f qcow2 l1.img 10G
qemu-img create -f qcow2 -b l1.img l2.img
Oops, you forgot the backing format. Without
Quoting Eric Blake (ebl...@redhat.com):
On 10/30/2014 02:32 PM, Serge Hallyn wrote:
Hi,
I'm looking into why virt-aa-helper isn't adding allow rules for
backing stores nested deeper than 1. So if I do
qemu-img create -f qcow2 l1.img 10G
qemu-img create -f qcow2 -b l1.img l2.img
On Thu, 2014-10-30 at 07:55 +0100, Martin Kletzander wrote:
On Thu, Oct 30, 2014 at 02:23:00AM +, Chen, Fan wrote:
On Wed, 2014-10-29 at 14:20 +0100, Martin Kletzander wrote:
On Wed, Oct 29, 2014 at 08:33:32PM +0800, Chen Fan wrote:
diff --git a/src/util/virnuma.c b/src/util/virnuma.c
When pass None or a empty dictionary to time, it will
report error.Allow a one-element dictionary which
contains 'seconds',setting JUST seconds will do the
sane thing of passing 0 for nseconds, instead of
erroring out.If dict have a unkown key, it will report error.
Signed-off-by: Luyao Huang
Hi,
I would like to use this command to backup my virtual machine on my
secondary server
virsh migrate --live --suspend --copy-storage-inc virtual_machine
qemu+ssh://root@secondary_server/system
But we don't have option to disable shut off the virtual machine on the
source server.
Would
61 matches
Mail list logo