October 2009 - Devel - libvirt List Archives

[libvirt] libvirt update problem
by Gerry Reno 22 Oct '09

22 Oct '09

# yum list 'libvirt' Loaded plugins: dellsysidplugin2, fastestmirror, installonlyn, refresh-packagekit, versionlock Loading mirror speeds from cached hostfile Excluding mirror: mirror.newnanutilities.org * fedora: mirror.hiwaay.net * updates: mirror.hiwaay.net Reading version lock configuration Installed Packages libvirt.i586 0.6.2-8.fc11 installed libvirt.x86_64 0.6.2-8.fc11 installed Available Packages libvirt.i586 0.6.2-18.fc11 updates libvirt.x86_64 0.6.2-18.fc11 updates and when I update libvirt: Transaction Check Error: file /usr/share/libvirt/schemas/domain.rng from install of libvirt-0.6.2-18.fc11.x86_64 conflicts with file from package libvirt-0.6.2-8.fc11.i586 What is wrong here?

1 1

[libvirt] cache=none
by Stefan 22 Oct '09

22 Oct '09

Hello, * Is it possible to add the "kvm -drive"-option "cache=none" to libvirtd.conf ? It's a recommendation of XFS FAQ: http://xfs.org/index.php/XFS_FAQ#Q:_Which_settings_are_best_with_virtualiza… * Is there a generic way to handle kvm options that are not supported in libvirtd.conf ? * Is it generally a bad idea to use XFS with KVM ? Thanks Stefan

2 3

[libvirt] how can dst know that migration is over?
by Dan Kenigsberg 22 Oct '09

22 Oct '09

When migrating a domain from src to dst, it first appears 'suspended' at the dst. When it's 'running', higher-level management can tell that migration is done. However, when bugzilla.redhat.com/519204 is fixed, I can no longer count on that - migrated domain may still be suspended at the destination. So how can I tell, at the destination, that migration is over? I could add higher-level communication from src to dst, but I am reluctant to add another failure mode. I'd rather have libvirt tell me, "hey migration is done". btw, does src libvirt verify with dst libvirt that migration is successful? src qemu can be oblivious of dst qemu not receiving the last packet correctly. Regards, Dan.

2 3

[libvirt] Suggest to postpone 0.7.3 by one week
by Daniel Veillard 22 Oct '09

22 Oct '09

Basically the plan was to enter freeze this week-end and release 0.7.3 on the 30th, but looking at the current logs and status on the list, this may not be appropriate, some of the patches aren't finalized and we would end up with 0.7.3 being mostly 0.7.2 with bug fixes. So I suggest to introduce a one week delay, enter the freeze at teh end of the month instead and push 0.7.3 on Nov 6th. I assume it's not a serious problem to anybody, right ? Daniel -- Daniel Veillard | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ daniel(a)veillard.com | Rpmfind RPM search engine http://rpmfind.net/ http://veillard.com/ | virtualization library http://libvirt.org/

1 0

[libvirt] [PATCH] Fix potential false-positive OOM error reporting
by Matthias Bolte 22 Oct '09

22 Oct '09

If no matching device was found (cap == NULL) then no strdup() call was made and *wwnn and *wwpn are untouched. Checking them for NULL in this situation may result in reporting an false-positive OOM error because *wwnn and *wwpn may be initialized to NULL by the caller. Only check *wwnn and *wwpn for NULL if a matching device was found (cap != NULL) and thus strdup() was called. * src/conf/node_device_conf.c: only report an OOM error if there really is one --- src/conf/node_device_conf.c | 4 +--- 1 files changed, 1 insertions(+), 3 deletions(-) diff --git a/src/conf/node_device_conf.c b/src/conf/node_device_conf.c index 77f7be3..c2c5a44 100644 --- a/src/conf/node_device_conf.c +++ b/src/conf/node_device_conf.c @@ -1243,9 +1243,7 @@ virNodeDeviceGetWWNs(virConnectPtr conn, virNodeDeviceReportError(conn, VIR_ERR_NO_SUPPORT, "%s", _("Device is not a fibre channel HBA")); ret = -1; - } - - if (*wwnn == NULL || *wwpn == NULL) { + } else if (*wwnn == NULL || *wwpn == NULL) { /* Free the other one, if allocated... */ VIR_FREE(wwnn); VIR_FREE(wwpn); -- 1.6.0.4

3 2

[libvirt] how to duplicate an vm?
by Dan Bar Dov 22 Oct '09

22 Oct '09

I am trying to duplicate an existing vm. While the VM was down, I copied the image file, copied the xml file and manually fixed the xml file to have a different VM name, and point to the copy. The copy now works, but the original disappraeed from virt-manager. When I try to "restore" it I get: Error restoring domain '/var/lib/libvirt/images/fc11-copy.img': operation failed: image magic is incorrect What am I doing wrong/how should I replicate a VM? Thanks, Dan

2 2

[libvirt] [PATCH] consolidate virXPathNodeSet()
by Daniel Veillard 22 Oct '09

22 Oct '09

While reviewing Rich last patch I was afraid that virXPathNodeSet() could return -1 when doing an evaluation failure. It is based on xmlXPathEval() from libxml2 whose behaviour is to possibly return NULL on XPath evaluation failure independantly of the kind of failure: http://xmlsoft.org/html/libxml-xpath.html#xmlXPathEval I suggest to apply this patch to make sure we always return 0 unless the returned XPath type is of the wrong type (meaning the query passed didn't evaluate to a node set and code must be fixed) Daniel -- Daniel Veillard | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ daniel(a)veillard.com | Rpmfind RPM search engine http://rpmfind.net/ http://veillard.com/ | virtualization library http://libvirt.org/

2 2

[libvirt] Possible qemu GetVcpus bug?
by Cole Robinson 21 Oct '09

21 Oct '09

Playing with the GetVcpus API call (and virsh vcpuinfo), there is a slight inconsistency that may be a bug. In qemudDomainGetVcpus, we have: if (nodeGetInfo(dom->conn, &nodeinfo) < 0) goto cleanup; maxcpu = maplen * 8; if (maxcpu > nodeinfo.cpus) maxcpu = nodeinfo.cpus; We clamp the amount of cpus we report affinity against to nodeinfo.cpus. virsh however expects this to be VIR_NODEINFO_MAXCPUS, and makes the cpumap appropriately long (and uses this value when pretty printing): cpumaplen = VIR_CPU_MAPLEN(VIR_NODEINFO_MAXCPUS(nodeinfo)); Is there a bug here? Should qemu be using VIR_NODEINFO_MAXCPUS, or is there some explicit reason to use nodeinfo.cpus? Thanks, Cole

1 0

[libvirt] RHEL 5 with Xen - changed IPs and now weirdness
by Denny Snyder 21 Oct '09

21 Oct '09

3 5

[libvirt] Xenner status
by Gerry Reno 21 Oct '09

21 Oct '09

Can someone tell me if RHEL-6 is planned to have support for Xenner? Will this be built into qemu? Otherwise what are options for using Xenner with RHEL-5? We want to run both KVM and Xen guests managed by KVM hypervisor.

2 2