hi,
If you already have some VMs running, try to do the same command as root.
It's certainly a right problem.
Regards
Florian CHAZAL
2010/6/1 marwen marwen marwen.e...@gmail.com
hello
I'm currently using libvirt and opennebula. I have installed the libvirt
with the driver of opennebule
On Wed, Jun 02, 2010 at 04:30:02PM -0300, Luiz Capitulino wrote:
On Wed, 02 Jun 2010 02:49:08 -0300
Eduardo Otubo ot...@linux.vnet.ibm.com wrote:
Last year, Daniel Veillard was invited to talk about Libvirt. I was
excited to see his lecture but ended up he couldn't come. This year, I
On Wed, Jun 02, 2010 at 03:53:12PM -0600, Eric Blake wrote:
On Fedora 13 with sufficient mingw32-* packages installed, running
./autobuild.sh failed to cross-compile to mingw because
mingw32-pthreads installed a broken pthread.h. With that
issue fixed, the build still failed due to use of
2010/6/3 florian chazal floriancha...@gmail.com:
hi,
If you already have some VMs running, try to do the same command as root.
It's certainly a right problem.
Regards
Florian CHAZAL
I think you're referring to the difference between connecting to
qemu:///session and qemu:///system.
The
The domain parsing code would auto-add a virtio serial controller
if it saw any virtio serial channel defined. Unfortunately it
always added a controller with index=0, even if the channel address
specified an index != 0. It only added one controller, even if
multiple controllers were referenced by
To ensure that the device addressing scheme is stable across
hotplug/unplug, all virtio serial channels needs to have an
associated port number in their address. This is then specified
to QEMU using the nr=NNN parameter
* src/conf/domain_conf.c, src/conf/domain_conf.h: Parsing
for port number
2010/6/2 Eduardo Otubo ot...@linux.vnet.ibm.com:
Now sending patches using format-patch and send-email, pretty
interesting! This is the final version, fixing the style according
to the new HACKING file.
---
src/phyp/phyp_driver.c | 34 ++
1 files changed,
When libvirtd exits it is leaving UNIX domain sockets on
the filesystem. These need to be removed.
The qemudInitPaths() method has signficant code churn to
switch from using a pre-allocated buffer on the stack, to
dynamically allocating on the heap.
* daemon/libvirtd.c, daemon/libvirtd.h: Store
On 06/02/2010 08:09 PM, David Allan wrote:
---
src/conf/node_device_conf.c |4 ++--
src/node_device/node_device_driver.c |3 ++-
2 files changed, 4 insertions(+), 3 deletions(-)
static int
nodeDeviceDestroy(virNodeDevicePtr dev)
{
-int ret = 0;
+int ret =
On 06/02/2010 10:09 PM, David Allan wrote:
---
src/conf/node_device_conf.c |4 ++--
src/node_device/node_device_driver.c |3 ++-
2 files changed, 4 insertions(+), 3 deletions(-)
diff --git a/src/conf/node_device_conf.c b/src/conf/node_device_conf.c
index 7f2dac8..6583570
On 06/02/2010 10:04 PM, David Allan wrote:
* It appears that the udev event for HBA creation arrives before the
associated sysfs data is fully populated, resulting in bogus data
for the nodedev entry until the entry is refreshed. This problem is
particularly troublesome when creating
On 06/03/2010 04:08 AM, Daniel P. Berrange wrote:
On Wed, Jun 02, 2010 at 03:53:12PM -0600, Eric Blake wrote:
On Fedora 13 with sufficient mingw32-* packages installed, running
./autobuild.sh failed to cross-compile to mingw because
mingw32-pthreads installed a broken pthread.h. With that
On 06/03/2010 12:04 AM, David Allan wrote:
* It appears that the udev event for HBA creation arrives before the
associated sysfs data is fully populated, resulting in bogus data
for the nodedev entry until the entry is refreshed. This problem is
particularly troublesome when creating
On 06/03/2010 08:54 AM, Daniel P. Berrange wrote:
The domain parsing code would auto-add a virtio serial controller
if it saw any virtio serial channel defined. Unfortunately it
always added a controller with index=0, even if the channel address
specified an index != 0. It only added one
On 06/03/2010 08:54 AM, Daniel P. Berrange wrote:
To ensure that the device addressing scheme is stable across
hotplug/unplug, all virtio serial channels needs to have an
associated port number in their address. This is then specified
to QEMU using the nr=NNN parameter
*
QEMU upstream decided against adding a 'reason' field to
the block IO event in QMP. Disable this code to remove a
annoying warning message. It will be renabled when the
error string reason is re-introduced in QEMU
---
src/qemu/qemu_monitor_json.c |4
1 files changed, 4 insertions(+), 0
On Thu, Jun 03, 2010 at 07:58:43AM -0600, Eric Blake wrote:
On 06/03/2010 04:08 AM, Daniel P. Berrange wrote:
On Wed, Jun 02, 2010 at 03:53:12PM -0600, Eric Blake wrote:
On Fedora 13 with sufficient mingw32-* packages installed, running
./autobuild.sh failed to cross-compile to mingw
First, thanks for the efforts (sometimes, as open source developers, we
get too caught up in our own work, and forget to acknowledge the work of
others; at which point our reviews come across as overly-critical, even
though that is not the intent).
Thank you for the fast answer! I preferred
On 06/03/2010 08:06 AM, Daniel P. Berrange wrote:
QEMU upstream decided against adding a 'reason' field to
the block IO event in QMP. Disable this code to remove a
annoying warning message. It will be renabled when the
error string reason is re-introduced in QEMU
ACK.
--
Eric Blake
On 06/03/2010 08:02 AM, Daniel P. Berrange wrote:
So, is the libvirt patch worth installing, while we wait for the
eventual fix of dependencies to not force mingw32-pthreads on 64-bit users?
How about changing configure.ac so that it defaults to looking for win32
threads on Win32, rather
On Thu, Jun 03, 2010 at 09:53:07AM -0400, Cole Robinson wrote:
On 06/03/2010 12:04 AM, David Allan wrote:
* It appears that the udev event for HBA creation arrives before the
associated sysfs data is fully populated, resulting in bogus data
for the nodedev entry until the entry is
On Thu, Jun 03, 2010 at 09:50:26AM -0400, Cole Robinson wrote:
On 06/02/2010 10:09 PM, David Allan wrote:
---
src/conf/node_device_conf.c |4 ++--
src/node_device/node_device_driver.c |3 ++-
2 files changed, 4 insertions(+), 3 deletions(-)
diff --git
I sent this patch yesterday but have not seen it on the list. Resending ...
Regards,
Jim
From f93f0d7c39aec219a9b2de590a9d97f2fbdaf6a0 Mon Sep 17 00:00:00 2001
From: Jim Fehlig jfeh...@novell.com
Date: Wed, 2 Jun 2010 18:07:17 -0600
Subject: [PATCH] Allocate buffer to hold xend response
There
On 06/03/2010 10:54 AM, Jim Fehlig wrote:
I sent this patch yesterday but have not seen it on the list. Resending
...
Regards,
Jim
There are cases when a response from xend can exceed 4096 bytes, in
which case anything beyond 4096 is ignored. This patch changes the
current fixed-size,
2010/6/3 Eric Blake ebl...@redhat.com:
The rule of thumb for generated files:
If they are distributed, they should be generated in $(srcdir);
otherwise, they should be built by the end user in $(builddir).
Since our .xml docs are built with python, and we want them
available even to end users
Jim Fehlig wrote:
...
Subject: [PATCH] Allocate buffer to hold xend response
There are cases when a response from xend can exceed 4096 bytes, in
which case anything beyond 4096 is ignored. This patch changes the
current fixed-size, stack-allocated buffer to a dynamically allocated
buffer
Jim Meyering wrote:
Jim Fehlig wrote:
...
Subject: [PATCH] Allocate buffer to hold xend response
There are cases when a response from xend can exceed 4096 bytes, in
which case anything beyond 4096 is ignored. This patch changes the
current fixed-size, stack-allocated buffer to a dynamically
On 06/03/2010 04:51 PM, Jim Fehlig wrote:
+if (content_length 0) {
ssize_t ret;
-if ((unsigned int) content_length (n_content + 1))
-content_length = n_content - 1;
+if (VIR_ALLOC_N(*content, content_length) 0 ) {
+virReportOOMError();
This blog post [1] explains how to apply some OpenSuse patches [2]
(especially xen-domctl-ver7.patch) to the libvirt 0.8.1 Debian package
in order to get Xen 4.0 support.
On IRC jonnyt reported problems with PCI passthrough in such a setup.
Debugging and XenD hacking revealed that XenD 4.0
Pass a struct containing the parameters instead of passing each
one individually. This make future extensions a bit simpler.
---
src/esx/esx_driver.c | 44 ++---
src/esx/esx_util.c | 100 +-
src/esx/esx_util.h | 15 ++-
Allow to specify a proxy to be used by libcurl.
---
docs/drvesx.html.in | 20 -
src/esx/esx_driver.c | 12 +++---
src/esx/esx_util.c | 57 ++
src/esx/esx_util.h |4 +++
src/esx/esx_vi.c | 17 --
On Thu, Jun 03, 2010 at 03:24:33PM +1000, Justin Clift wrote:
Hi all,
Looking at the virsh man page (tools/virsh.pod), it's getting pretty
unwieldy and hard to grok for end users.
Wondering if it's worth breaking it up into sections?
Maybe something similar to how git structures their
On 06/04/2010 10:29 AM, Daniel Veillard wrote:
Honnestly, git does it that way and I hate that. The advantage of man
is that it's text and you can search for verbs or descriptions which may
correspond to the actions you're trying to accomplish. If you split it
in a zillion man pages it's a
These two patches are posted together because applying the 2nd exposes
the bug fixed by the first.
Here are the results of tests I made with various block sizes before
deciding the 1MB really was the best balance (all tests were done on a
paused 512MB domain, saving to local disk on a Lenovo T61
The pointer to the xml describing the domain is saved into an object
prior to calling VIR_REALLOC_N() to make the size of the memory it
points to a multiple of QEMU_MONITOR_MIGRATE_TO_FILE_BS. If that
operation needs to allocate new memory, the pointer that was saved is
no longer valid.
To avoid
See https://bugzilla.redhat.com/show_bug.cgi?id=599091
Saving a paused 512MB domain took 3m47s with the old block size of 512
bytes. Changing the block size to 1024*1024 decreased the time to 56
seconds. (Doubling again to 2048*1024 yielded 0 improvement; lowering
to 512k increased the save time
Eric Blake wrote:
BTW, I pushed the patch after Eric's ACK. I'll role another to address
these issues, once I get confirmation on the NUL-termination.
Looking forward to the followup, and thanks for Jim for catching
something I missed in my ACK (even if we didn't catch it in time).
Matthias Bolte wrote:
This blog post [1] explains how to apply some OpenSuse patches [2]
(especially xen-domctl-ver7.patch)
Ah yes, reminds me to upstream this patch. We had all of the cpu pools
patches (which changed domctl interface) in Xen4.0 in openSUSE build
service. These patches
38 matches
Mail list logo
