[libvirt] f11 virt release notes freezing

I've been busy at my day job and not paying close enough attention to Fedora 11 docs deadlines. Apparently the window for changes essentially closes Wednesday (today).[1] The notes are essentially in place. I don't think there are any gaping holes, but I should have prompted for feedback much soon

Re: [libvirt] [Fwd: first cut public API for physical host interface configuration]

On Tue, 2009-03-31 at 10:24 +0100, Daniel P. Berrange wrote: > Simply that if the admin configures a new NIC using DHCP / IPV6 they will > wish to know what address it got after it was brought online. libvirt > does not have todo anything with these addresses, merely provide info > about them in th

Re: [libvirt] [PATCH 1/1] Add SCSI pool support.

Dave Allan wrote: Daniel P. Berrange wrote: On Mon, Mar 30, 2009 at 05:50:38PM -0400, Dave Allan wrote: Daniel P. Berrange wrote: This works now on RHEL-5, F9 and F10. I still had to disable this code if (STREQLEN(devpath, vol->target.path, PATH_MAX)) { VIR_DEBUG(_("No stable path

Re: [libvirt] PATCH: Fix XM driver to extract vifname= parameter

On Tue, Mar 31, 2009 at 04:58:42PM +0100, Daniel P. Berrange wrote: > The XM driver is not currently handling the vifname= parameter for > network devices in /etc/xen config files. This patch fixes that > ommision Okay, looks fine to me, thanks ! Daniel -- Daniel Veillard | libxml Gnome

Re: [libvirt] PATCH: Fix error reporting for guest with missing network

On Tue, Mar 31, 2009 at 04:37:36PM +0100, Daniel P. Berrange wrote: > If you attempt to create a Xen guest using a virutal network that does not > exist, or is not running you get a cryptic message > > # virsh create rhel5pv.xml > error: Failed to create domain from rhel5pv.xml > error: XML descri

Re: [libvirt] PATCH: Fix error report for unknown node devices

On Tue, Mar 31, 2009 at 03:20:40PM +0100, Daniel P. Berrange wrote: > When requesting a non-existant device you currently get > > # virsh nodedev-dumpxml foo > error: Could not find matching device 'foo' > error: invalid node device pointer in no node device with matching name > > > This patch f

Re: [libvirt] PATCH: Update XenD for new location of (localtime 1) SEXPR

On Tue, Mar 31, 2009 at 02:33:46PM +0100, Daniel P. Berrange wrote: > Newer versions of XenD have the (localtime 1) bit of the SEXPR > moved outside the (image ...) bit, and made available for both PV > and HVM guests. This patch updates our code to cope with this. For > compatability we also stil

Re: [libvirt] question about qemudDomainMigratePrepare2 of qemu_driver.c from libvirt-0.6.1

On Tue, Mar 31, 2009 at 05:29:09PM +0200, Chris Lalancette wrote: > Daniel P. Berrange wrote: > > On Tue, Mar 31, 2009 at 10:07:31AM +0200, gerrit.slo...@drv-bund.de wrote: > >> Hello > >> > >> I am just curious if the checking of the connection string in qemu_driver.c > >> from libvirt-0.6.1 in qe

Re: [libvirt] PATCH: Add --tree flag to virsh nodedev-list

On Tue, Mar 31, 2009 at 02:58:37PM +0100, Daniel P. Berrange wrote: > On Tue, Mar 31, 2009 at 12:09:21PM +0200, Daniel Veillard wrote: > > > > > +#define MAX_INDENT 100 > > > + > > > +static void > > > +cmdNodeListDevicesPrint(vshControl *ctl, > > > +char **devices, > > > +

Re: [libvirt] PATCH: Fix misc memory alloc/free/usage bugs

On Tue, Mar 31, 2009 at 02:29:02PM +0100, Daniel P. Berrange wrote: > On Tue, Mar 31, 2009 at 12:04:00PM +0200, Daniel Veillard wrote: > Actually, the buffer is the exact right size, because that > 'buf.st_size' was filled by a lstat() call, and the POSIX > semantics for lstat() are that 'st_size'

[libvirt] How to prevent libvirt from adding iptables rules?

Hi, I'm new to libvirt but not a complete neophite. I'm using libvirt and kvm in ubuntu with "vmbuilder". I'm creating a couple of VMs inside a host that is directly connected to internet with a public routeable address. Since I only have one public address, I won't use bridging. I'm using

Re: [libvirt] [RFC] Power Hypervisor Libvirt support

Hello Daniel, Look, I don't want to make big deal about this, I'll just answer these points below with some arguments they gave me. If we find out libssh2 is better, I' can change it, no problem at all. > > In fact I use libssh and not > > libssh2. I choos

Re: [libvirt] [PATCH 1/1] Add SCSI pool support.

Daniel P. Berrange wrote: On Mon, Mar 30, 2009 at 05:50:38PM -0400, Dave Allan wrote: Daniel P. Berrange wrote: This works now on RHEL-5, F9 and F10. I still had to disable this code if (STREQLEN(devpath, vol->target.path, PATH_MAX)) { VIR_DEBUG(_("No stable path found for '%s' in '

[libvirt] PATCH: Fix XM driver to extract vifname= parameter

The XM driver is not currently handling the vifname= parameter for network devices in /etc/xen config files. This patch fixes that ommision Daniel diff -r e396919fa59e src/xm_internal.c --- a/src/xm_internal.c Tue Mar 31 16:35:42 2009 +0100 +++ b/src/xm_internal.c Tue Mar 31 16:58:11 2009 +0100

Re: [libvirt] [PATCH 2/2]: Remove getURI internal driver call

Daniel P. Berrange wrote: > On Tue, Mar 31, 2009 at 11:41:58AM +0200, Daniel Veillard wrote: >> On Mon, Mar 30, 2009 at 02:10:13PM +0200, Chris Lalancette wrote: >>> This patch removes the internal getURI() driver callback. It was *never* >>> overridden by any of the drivers, so it seems kind of p

Re: [libvirt] [PATCH 1/2]: Convert internal drivers to old style struct initializers

Daniel P. Berrange wrote: > On Tue, Mar 31, 2009 at 11:40:32AM +0200, Daniel Veillard wrote: >> On Mon, Mar 30, 2009 at 02:10:05PM +0200, Chris Lalancette wrote: >>> This patch converts the internal drivers to use the old-style struct >>> initializers, which kind of gives us a self-documenting TODO

[libvirt] PATCH: Fix error reporting for guest with missing network

If you attempt to create a Xen guest using a virutal network that does not exist, or is not running you get a cryptic message # virsh create rhel5pv.xml error: Failed to create domain from rhel5pv.xml error: XML description for failed to build sexpr is not well formed or invalid This is because t

Re: [libvirt] question about qemudDomainMigratePrepare2 of qemu_driver.c from libvirt-0.6.1

Daniel P. Berrange wrote: > On Tue, Mar 31, 2009 at 10:07:31AM +0200, gerrit.slo...@drv-bund.de wrote: >> Hello >> >> I am just curious if the checking of the connection string in qemu_driver.c >> from libvirt-0.6.1 in qemudDomainMigratePrepare2 is faulty. >> The source is: >> >> [...] >> } els

[libvirt] PATCH: Fix error report for unknown node devices

When requesting a non-existant device you currently get # virsh nodedev-dumpxml foo error: Could not find matching device 'foo' error: invalid node device pointer in no node device with matching name This patch fixes that to give # virsh nodedev-dumpxml foo error: Could not find matching device

Re: [libvirt] question about qemudDomainMigratePrepare2 of qemu_driver.c from libvirt-0.6.1

On Tue, Mar 31, 2009 at 10:07:31AM +0200, gerrit.slo...@drv-bund.de wrote: > Hello > > I am just curious if the checking of the connection string in qemu_driver.c > from libvirt-0.6.1 in qemudDomainMigratePrepare2 is faulty. > The source is: > > [...] > } else { > /* Check the URI sta

Re: [libvirt] PATCH: Add --tree flag to virsh nodedev-list

On Tue, Mar 31, 2009 at 12:09:21PM +0200, Daniel Veillard wrote: > > > +#define MAX_INDENT 100 > > + > > +static void > > +cmdNodeListDevicesPrint(vshControl *ctl, > > +char **devices, > > +char **parents, > > +int num_devices

[libvirt] PATCH: Update XenD for new location of (localtime 1) SEXPR

Newer versions of XenD have the (localtime 1) bit of the SEXPR moved outside the (image ...) bit, and made available for both PV and HVM guests. This patch updates our code to cope with this. For compatability we also still put the (localtime 1) bit inside the HVM (image ...) section too - its har

Re: [libvirt] PATCH: Fix misc memory alloc/free/usage bugs

On Tue, Mar 31, 2009 at 12:04:00PM +0200, Daniel Veillard wrote: > > > +++ b/src/security_selinux.c    Mon Mar 30 14:37:45 2009 +0100 > > > @@ -303,11 +303,13 @@ SELinuxRestoreSecurityImageLabel(virConn > > >         return -1; > > > > > >     if (S_ISLNK(buf.st_mode)) { > > > +        int n; > > >

Re: [libvirt] PATCH: Clear reference after closing connection object

On Tue, Mar 31, 2009 at 01:16:31PM +0100, Daniel P. Berrange wrote: > In the python bindings, if you call close() it will close the connection > and release the handle, but it doesn't clear the python reference to the > C object - self._o. So later in garbage colection __del__ runs and tries > to f

[libvirt] question about qemudDomainMigratePrepare2 of qemu_driver.c from libvirt-0.6.1

Hello I am just curious if the checking of the connection string in qemu_driver.c from libvirt-0.6.1 in qemudDomainMigratePrepare2 is faulty. The source is: [...] } else { /* Check the URI starts with "tcp:". We will escape the * URI when passing it to the qemu monitor, so b

Re: [libvirt] PATCH: Fix duplicate methods in python binding

On Tue, Mar 31, 2009 at 01:15:14PM +0100, Daniel P. Berrange wrote: > There are three methods with the same name in the python binding > due to a generator bug. This patch fixes them, so we get methods > networkCreateXML() and storagePoolCreateXML() instad of three > called createXML() Oops :-)

Re: [libvirt] [PATCH] additional options in the spec file for storage drivers and numactl

On Mon, Mar 30, 2009 at 12:50:09AM +0900, Ryota Ozaki wrote: > Hi, > > This patch adds switches in the spec file for storage drivers and numactl > to be easy to turn on/off options of configure. Looks good, applied and commited, now I will have to update Fedora spec file accordingly. Thanks

Re: [libvirt] [PATCH] prevent an executed external process to be a zombie

On Mon, Mar 30, 2009 at 12:49:55AM +0900, Ryota Ozaki wrote: > Hi, > > This patch prevents an executed external process to be a zombie > by doing waitpid() even if virPipeReadUntilEOF() fails. Makes sense, applied and commited, thanks a lot ! Daniel -- Daniel Veillard | libxml Gnom

[libvirt] PATCH: Clear reference after closing connection object

In the python bindings, if you call close() it will close the connection and release the handle, but it doesn't clear the python reference to the C object - self._o. So later in garbage colection __del__ runs and tries to free the already free'd object Daniel diff -r a76dc8c27d72 python/generator

[libvirt] PATCH: Fix duplicate methods in python binding

There are three methods with the same name in the python binding due to a generator bug. This patch fixes them, so we get methods networkCreateXML() and storagePoolCreateXML() instad of three called createXML() Daniel diff -r e26692afef21 python/generator.py --- a/python/generator.py Tue Ma

Re: [libvirt] PATCH: Fix misc memory alloc/free/usage bugs

On Tue, Mar 31, 2009 at 11:07:07AM +0100, Daniel P. Berrange wrote: > On Tue, Mar 31, 2009 at 12:04:00PM +0200, Daniel Veillard wrote: > > On Mon, Mar 30, 2009 at 11:41:20PM +0900, Ryota Ozaki wrote: > > > Hi Daniel, > > > > > > On Mon, Mar 30, 2009 at 10:39 PM, Daniel P. Berrange > > > wrote: >

Re: [libvirt] PATCH: Add --tree flag to virsh nodedev-list

On Mon, Mar 30, 2009 at 02:44:31PM +0100, Daniel P. Berrange wrote: > > The virsh noddev-list command is used to display node devices. It just > prints out their names in a flat list. When detaching devices for purposes > of PCI passthrough though, it is important to understand what devices are >

Re: [libvirt] PATCH: Fix misc memory alloc/free/usage bugs

On Tue, Mar 31, 2009 at 12:04:00PM +0200, Daniel Veillard wrote: > On Mon, Mar 30, 2009 at 11:41:20PM +0900, Ryota Ozaki wrote: > > Hi Daniel, > > > > On Mon, Mar 30, 2009 at 10:39 PM, Daniel P. Berrange > > wrote: > > > A few more small bugs > > > > > >  * src/node_device.c: Don't strdup() a NUL

Re: [libvirt] Patch to allow setting of svirt XML.

On Mon, Mar 30, 2009 at 09:15:33AM -0400, Daniel J Walsh wrote: > On 03/28/2009 07:51 AM, Daniel P. Berrange wrote: > >On Thu, Mar 26, 2009 at 11:51:21AM -0400, Daniel J Walsh wrote: > >>This patch fixes the seclabel handling in domain_conf.c to allow > >>virt-manager to set the seclabel model, typ

Re: [libvirt] PATCH: Fix misc memory alloc/free/usage bugs

On Mon, Mar 30, 2009 at 11:41:20PM +0900, Ryota Ozaki wrote: > Hi Daniel, > > On Mon, Mar 30, 2009 at 10:39 PM, Daniel P. Berrange > wrote: > > A few more small bugs > > > >  * src/node_device.c: Don't strdup() a NULL device parent > >  * src/qemu_conf.c: Don't try to run access() on a NULL binar

Re: [libvirt] [PATCH 1/1] Add SCSI pool support.

On Mon, Mar 30, 2009 at 05:50:38PM -0400, Dave Allan wrote: > Daniel P. Berrange wrote: > > > >This works now on RHEL-5, F9 and F10. > > > >I still had to disable this code > > > >if (STREQLEN(devpath, vol->target.path, PATH_MAX)) { > >VIR_DEBUG(_("No stable path found for '%s' in '%s'"

Re: [libvirt] [PATCH 2/2]: Remove getURI internal driver call

On Tue, Mar 31, 2009 at 11:41:58AM +0200, Daniel Veillard wrote: > On Mon, Mar 30, 2009 at 02:10:13PM +0200, Chris Lalancette wrote: > > This patch removes the internal getURI() driver callback. It was *never* > > overridden by any of the drivers, so it seems kind of pointless (I even > > checked

Re: [libvirt] Upcoming 0.6.2 release

On Fri, 2009-03-27 at 18:49 +0100, Daniel Veillard wrote: > I would suggest the following: > - I post every week, say at the end of the week when I think the > next releases will be likely to occur, trying to refine as we get > closer > - We try to not make any release if the

Re: [libvirt] [PATCH 1/2]: Convert internal drivers to old style struct initializers

On Tue, Mar 31, 2009 at 11:40:32AM +0200, Daniel Veillard wrote: > On Mon, Mar 30, 2009 at 02:10:05PM +0200, Chris Lalancette wrote: > > This patch converts the internal drivers to use the old-style struct > > initializers, which kind of gives us a self-documenting TODO list for each > > of > > th

Re: [libvirt] [PATCH 2/2]: Remove getURI internal driver call

On Mon, Mar 30, 2009 at 02:10:13PM +0200, Chris Lalancette wrote: > This patch removes the internal getURI() driver callback. It was *never* > overridden by any of the drivers, so it seems kind of pointless (I even > checked > the "in-progress" VirtualBox, OpenNebula, and Power Hypervisor drivers

Re: [libvirt] [PATCH 1/2]: Convert internal drivers to old style struct initializers

On Mon, Mar 30, 2009 at 02:10:05PM +0200, Chris Lalancette wrote: > This patch converts the internal drivers to use the old-style struct > initializers, which kind of gives us a self-documenting TODO list for each of > the drivers. It also fixes up some bad comments in the internal driver list. >

Re: [libvirt] PATCH: Xen PCI device passthrough

On Tue, Mar 31, 2009 at 10:10:14AM +0200, Gerd Hoffmann wrote: > Hi, > > > +static const char * > > +pciFindStubDriver(virConnectPtr conn) > > +{ > > +char drvpath[PATH_MAX]; > > +int probed = 0; > > + > > +recheck: > > +pciDriverDir(drvpath, sizeof(drvpath), "pci-stub"); > > +if

Re: [libvirt] [Fwd: first cut public API for physical host interface configuration]

On Mon, Mar 30, 2009 at 02:45:51PM -0700, David Lutterkort wrote: > On Mon, 2009-03-30 at 12:40 +0100, Daniel P. Berrange wrote: > > Well you'll need to be able to query live interfaces anyway, because > > you'll need to be able to report IPv4 addresses assigned by DHCP and > > report IPv6 address

Re: [libvirt] PATCH: Misc error handling / OOM fixes

On Mon, Mar 30, 2009 at 12:52:48PM +0100, Daniel P. Berrange wrote: > I ran the OOM checking test suite again and uncovered a handful of > pretty minor bugs - no crashers in real code - just forgetting to > report OOM correctly > > * src/datatypes.c: Convert to use the virReportErrorHelper macro

Re: [libvirt] PATCH: Xen PCI device passthrough

Hi, > +static const char * > +pciFindStubDriver(virConnectPtr conn) > +{ > +char drvpath[PATH_MAX]; > +int probed = 0; > + > +recheck: > +pciDriverDir(drvpath, sizeof(drvpath), "pci-stub"); > +if (virFileExists(drvpath)) > +return "pci-stub"; > +pciDriverDir(drvpath,