On 06.12.2013 00:52, Eric Blake wrote:
A couple of fixes for bootstrap issues reported on IRC:
- on some older glibc systems, ./configure could deadlock due to
a glibc malloc bug
- on FreeBSD systems, a broken autom4te coupled with gettext
0.18.3 prevents bootstrap; we can't work around it,
ping
-Original Message-
From: libvir-list-boun...@redhat.com
[mailto:libvir-list-boun...@redhat.com]
On Behalf Of Chen Hanxiao
Sent: Tuesday, December 03, 2013 4:37 PM
To: libvir-list@redhat.com
Subject: Re: [libvirt] [PATCH]lxc: remove redundant mount operation
ping
On 12/05/13 20:25, Eric Blake wrote:
On 12/05/2013 09:30 AM, Peter Krempa wrote:
On 12/05/13 17:17, Peter Krempa wrote:
When changing the parsing and formatting functions in commit
43f2ccdc73090bd03f64de4d58d46ffa0134d705 I forgot to update the qemu
disk alingnment function for snapshots that
On 12/05/13 20:26, Eric Blake wrote:
On 12/05/2013 10:36 AM, Peter Krempa wrote:
https://bugzilla.redhat.com/show_bug.cgi?id=1035118
When outputting the XML for the RNG device, the code didn't format the
PCI address info. Additionally the schema wasn't expecting the info
although it was
On 12/05/13 20:25, Eric Blake wrote:
On 12/05/2013 09:50 AM, Peter Krempa wrote:
https://bugzilla.redhat.com/show_bug.cgi?id=1035108
When attempting to enable more vCPUs in the guest than is currently
enabled in the guest but less than the maximum count for the VM we
currently reported an
Commit 1c8ce29 added a way to unset a GVirConfigDomainClock set on a
GVirConfigDomain by calling gvir_config_domain_set_clock() with a NULL
argument.
However, a typo in a g_return_if_fail precondition prevents this code from
ever being called: g_return_if_fail(clock != NULL) is used while it
Hey,
I wanted to add support in libvirt-gconfig for the 'discard' attribute of
the disk driver node. If I follow the way the API is currently done, it
would be an additional method to GVirConfigDomainDisk. However, there are
quite a few attributes attached to the disk driver node, so I felt it
This class wraps creation of configuration data for the driver part of a
domain disk device. The methods needed for this are currently part of
GVirConfigDomainDisk, but since the disk driver is getting more and more
attributes, it's better to move such configuration to a dedicated class to
avoid a
---
libvirt-gconfig/tests/test-domain-create.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/libvirt-gconfig/tests/test-domain-create.c
b/libvirt-gconfig/tests/test-domain-create.c
index b5c2cf3..e0d6c00 100644
--- a/libvirt-gconfig/tests/test-domain-create.c
+++
---
libvirt-gconfig/libvirt-gconfig-domain-disk.c | 42 +++
libvirt-gconfig/libvirt-gconfig-domain-disk.h | 58 +++
libvirt-gconfig/libvirt-gconfig.sym | 3 ++
3 files changed, 77 insertions(+), 26 deletions(-)
diff --git
They are replaced by equivalent methods in GVirConfigDomainDiskDriver.
Initially, we had only one or two attributes to set on the 'driver' child
of the 'disk' node. Nowadays, we more than 5 attributes to set on this
node, mapping it as a separate object is more consistent with the rest of
Hey,
This series is based of the libosinfo work danpb did recently to reuse
libvirt scripts to validate the symbol files. These make sure the exported
symbols match what is listed in the sym file, and that these files are
alphabetically
ordered. This caught 2 minor errors in the sym files we
This is in preparation to enforcing this sorting during make check.
The script only checks for alphabetical sorting, which leads to unexpected
sorting between classes sharing a common prefix (eg the correct order is:
gvir_config_domain_get_foo;
gvir_config_domain_interface_set_foo;
gvir_connection_restore_domain_from_file_finish was not properly exported
because of a typo in the symbol name.
This commit also removes a newer occurrence of this symbol in the file
(without the typo). This can be done without breaking ABI as we haven't
made a release since this duplicate symbol
Import two test scripts from libvirt code which validate that
all symbols in libvirt-glib symbol files exist in the ELF binary, and also
validate the alphabetical sorting. These are hooked up to run with 'make
check'.
This commit is based on a libosinfo patch from Daniel P. Berrange
---
This method was never exported. GVirConfigDomainCpu inherits from
GVirConfigCapabilitiesCpu so gvir_config_capabilities_cpu_get_features()
should be used to get these features (this calls the virtual method
GVirConfigCapabilities::get_features() which GVirConfigDomainCpu
implements).
---
55bfd02 added a 'ram' attribute for qxl video devices
but didn't update the ABI check.
https://bugzilla.redhat.com/show_bug.cgi?id=1035123
---
src/conf/domain_conf.c | 7 +++
1 file changed, 7 insertions(+)
diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c
index 0268847..b76cf26
On 12/06/13 15:01, Ján Tomko wrote:
55bfd02 added a 'ram' attribute for qxl video devices
but didn't update the ABI check.
https://bugzilla.redhat.com/show_bug.cgi?id=1035123
---
src/conf/domain_conf.c | 7 +++
1 file changed, 7 insertions(+)
diff --git a/src/conf/domain_conf.c
On 12/06/2013 03:10 PM, Peter Krempa wrote:
On 12/06/13 15:01, Ján Tomko wrote:
55bfd02 added a 'ram' attribute for qxl video devices
but didn't update the ABI check.
https://bugzilla.redhat.com/show_bug.cgi?id=1035123
---
src/conf/domain_conf.c | 7 +++
1 file changed, 7
In commit f386825 we began adding the option --local=/$mydomain/ to
all dnsmasq commandlines (later changed to local=/$mydomain/ when we
moved the options from the commandline to a conf file) with the stated
reason of preventing forwarding of DNS queries for names that weren't
fully qualified
A newly finished part of the large gluster snapshot series.
These patches allow to parse and format definitions of snapshot
disks including block, dir and network backing.
Peter Krempa (3):
snapshot: schema: Split out snapshot disk driver definition
snapshot: Add support for specifying
Extract the definition to a new type to allow avoiding of duplication.
---
docs/schemas/domainsnapshot.rng | 24 ++--
1 file changed, 14 insertions(+), 10 deletions(-)
diff --git a/docs/schemas/domainsnapshot.rng b/docs/schemas/domainsnapshot.rng
index 7b46df1..169fcfb 100644
Amend tests to check parsing of the various new disk types that can now
be specified.
---
tests/domainsnapshotxml2xmlin/disk_snapshot.xml | 18 ++
tests/domainsnapshotxml2xmlout/disk_snapshot.xml | 18 ++
2 files changed, 36 insertions(+)
diff --git
Add support for specifying various types when doing snapshots. This will
later allow to do snapshots on network backed volumes. Disks of type
'volume' are not supported by snapshots (yet).
---
docs/schemas/domainsnapshot.rng | 76 +++--
src/conf/snapshot_conf.c
On 06.12.2013 05:15, Wangyufei (James) wrote:
From be1ce1c794b05a27aa85d037500a821b8c514d12 Mon Sep 17 00:00:00 2001
From: Zhou Yimin zhouyi...@huawei.com
Date: Fri, 6 Dec 2013 11:38:14 +0800
Subject: [PATCH] Cgroup: Replace 'newpath' with 'newPath'
Unifying codding style, replace 'newpath'
On 06.12.2013 07:20, Chen Hanxiao wrote:
From: Chen Hanxiao chenhanx...@cn.fujitsu.com
If we failed to make path for pts,
lxcContainerMountFSDevPTS will return the value
virAsprintf returned rather than -1.
Signed-off-by: Chen Hanxiao chenhanx...@cn.fujitsu.com
---
I tested the patch, and it didn't work for me (already found the bug,
though, and a new patch will be sent shortly). The bug is related to the
cGroup code (specifically, the virDomainDiskDefForeachPath function in
conf/domain_conf.c). The entire libvirt codebase does not yet utilize the
actualtype
On 06.12.2013 00:36, Jim Fehlig wrote:
This series moves some driver-specific files from the main libvirt-daemon
package to the respective libvirt-daemon-driver-* package when building with
driver modules.
This was originally all one patch that I found difficult to review, so I tried
to
On 06.12.2013 00:36, Jim Fehlig wrote:
Only add /etc/libvirt/nwfilter to the libvirt-daemon files list
if building with nwfilter support.
---
libvirt.spec.in | 2 ++
1 file changed, 2 insertions(+)
diff --git a/libvirt.spec.in b/libvirt.spec.in
index eff7103..7f2a23a 100644
---
Michal Privoznik wrote:
On 06.12.2013 00:36, Jim Fehlig wrote:
Only add /etc/libvirt/nwfilter to the libvirt-daemon files list
if building with nwfilter support.
---
libvirt.spec.in | 2 ++
1 file changed, 2 insertions(+)
diff --git a/libvirt.spec.in b/libvirt.spec.in
index
Jim Fehlig wrote:
Michal Privoznik wrote:
On 06.12.2013 00:36, Jim Fehlig wrote:
Only add /etc/libvirt/nwfilter to the libvirt-daemon files list
if building with nwfilter support.
---
libvirt.spec.in | 2 ++
1 file changed, 2 insertions(+)
diff --git a/libvirt.spec.in
On 12/06/2013 10:19 AM, Jim Fehlig wrote:
I modified commit message to reflect this change. Updated patch attached.
ACK to the updated version. Thanks for tackling this series.
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
Here are updated patches that implement RBD storage pool support in libvirt.
By and large, my changes are confined to qemu_conf.c, but I did need a couple
of changes in domain_conf.c in order to get things working. I've tested these
patches on my equipment, and they worked there, but they could
While testing Peter's changes to my patches, I discovered that when
restarting libvirt with domains running, they would no longer show
up when a 'virsh list' was issued. I tracked this bug down to this
section of code within virDomainDiskDefParseXML. The symptom was
that libvirt was issuing an
This helper function pulls the host definitions from the
RBD storage pool definition, and applies it to the disk
definition.
---
src/qemu/qemu_conf.c | 39 +++
1 file changed, 39 insertions(+)
diff --git a/src/qemu/qemu_conf.c b/src/qemu/qemu_conf.c
index
virDomainDiskDefForeachPath returned 0 previously for disks of type
VIR_DOMAIN_DISK_TYPE_NETWORK. This patch extends this check to also
bypass the function for VIR_DOMAIN_DISK_TYPE_VOLUME, shoudl the new
disk variable def-actualtype contain VIR_DOMAIN_DISK_TYPE_NETWORK.
---
src/conf/domain_conf.c
This patch is updated to not contain a switch statement
on the mode. The patch as a whole allows RBD storage pool
volumes to be used within domain XML.
---
src/qemu/qemu_conf.c | 23 ++-
1 file changed, 22 insertions(+), 1 deletion(-)
diff --git a/src/qemu/qemu_conf.c
Michal Privoznik wrote:
On 06.12.2013 00:36, Jim Fehlig wrote:
This series moves some driver-specific files from the main libvirt-daemon
package to the respective libvirt-daemon-driver-* package when building with
driver modules.
This was originally all one patch that I found difficult
Unfortunately, after I got home, I discovered that this patch introduces a
bug that seems to have been previously hidden behind the one that this
fixes. Doubly unfortunate, is that domains using RBD (and probably ISCSI,
but I don't have the facilities to test this) volumes with authentication
Python 3 no longer accepts 'except Exception, e:' as valid while Python
2.4 does not accept the new syntax 'except Exception as e:' so this uses
a fall back method that is compatible with both.
---
libvirt-override-virStream.py | 3 ++-
libvirt-override.py | 8 ++--
2 files changed,
The 'print' statement no longer exists in Python 3 and now must be
called as a function. This is compatible down to Python 2.4 as we are
not using any special syntax of the function.
---
sanitytest.py | 10 +-
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/sanitytest.py
I built a test domain using the default storage pool, backed by files under
/var/lib/libvirt/images/. I can confirm that the apparent race condition
also appears there, too. So it definitely seems like using volumes in
general can cause issues with domains being restarted after a libvirt
restart.
This patch works around a race condition present during
libvirt startup. The race condition only applies when
using storage pool volumes for domain disks, and even
then, only when restarting libvirt with running domains.
The gist of the patch is simply to enter a (limited)
retry loop during
This patch works around a race condition on libvirt start. It feels a
little hackish to me, but I have a hard time letting go of problems, so
I had to at least try and implement something before going to sleep. I'd
love some comments on the workaround, and welcome any improvement
suggestions, as
44 matches
Mail list logo
