config 20 sriov direct bond ports，vm create failed.

[Michael liu]

nova log：

 2016-04-08 09:57:48.640 5057 INFO nova.compute.manager [req-4e1b4d70-62b6-4158-8413-3c9f226fd13b - - - - -] report alarm_instance_shutoff success

2016-04-08 09:57:48.712 5057 INFO nova.compute.manager [req-4e1b4d70-62b6-4158-8413-3c9f226fd13b - - - - -] [instance: d860169c-0dac-448f-a644-01a9b200cebe] During _sync_instance_power_state the DB power_state (1) does not match the vm_power_state from the hypervisor (4). Updating power_state in the DB to match the hypervisor.

2016-04-08 09:57:48.791 5057 WARNING nova.compute.manager [req-4e1b4d70-62b6-4158-8413-3c9f226fd13b - - - - -] [instance: d860169c-0dac-448f-a644-01a9b200cebe] Instance shutdown by itself. Calling the heal_instance_state. Current vm_state: active, current task_state: None, original DB power_state: 1, current VM power_state: 4

2016-04-08 09:57:48.892 5057 INFO nova.compute.manager [req-4e1b4d70-62b6-4158-8413-3c9f226fd13b - - - - -] alarm_notice_heal_event result:1,host_name:tfg120,instance_id:d860169c-0dac-448f-a644-01a9b200cebe,instance_name:vfnicdirect,vm_state:active,power_state:shutdown,action:start

2016-04-08 09:57:48.997 5057 INFO nova.compute.manager [req-4e1b4d70-62b6-4158-8413-3c9f226fd13b - - - - -] Refresh_instance_block_device_info:False

2016-04-08 09:57:48.998 5057 INFO nova.compute.manager [req-4e1b4d70-62b6-4158-8413-3c9f226fd13b - - - - -] [instance: d860169c-0dac-448f-a644-01a9b200cebe] Rebooting instance

2016-04-08 09:57:49.373 5057 WARNING nova.compute.manager [req-4e1b4d70-62b6-4158-8413-3c9f226fd13b - - - - -] [instance: d860169c-0dac-448f-a644-01a9b200cebe] trying to reboot a non-running instance: (state: 4 expected: 1)

2016-04-08 09:57:49.479 5057 INFO nova.virt.libvirt.driver [-] [instance: d860169c-0dac-448f-a644-01a9b200cebe] Instance destroyed successfully.

libvirtd log：

2016-04-08 02:05:05.785+0000: 4778: info : qemuDomainDestroyFlags:2227 : Log: VM: name= instance-000000b8

2016-04-08 02:05:16.156+0000: 4771: info : qemuDomainDefineXMLFlags:7576 : Creating domain 'instance-000000b8'

2016-04-08 02:05:16.158+0000: 4773: info : qemuDomainCreateWithFlags:7448 : Log: VM: name= instance-000000b8

2016-04-08 02:05:16.158+0000: 4773: info : qemuProcessStart:4412 : vm=0x7f19482fdb30 name=instance-000000b8 id=-1 asyncJob=0 migrateFrom=<null> stdin_fd=-1 stdin_path=<null> snapshot=(nil) vmop=0 flags=0x1

2016-04-08 02:05:16.169+0000: 4773: info : virNetDevReplaceNetConfig:2541 : Replace Net Config of linkdev enp132s0f0, vf 28, macaddress 00:d1:d4:00:05:03, vlanid 1250, stateDir /var/run/libvirt/hostdevmgr

2016-04-08 02:05:16.169+0000: 4773: info : virNetDevReplaceNetConfig:2566 : Replace Vf Config of enp132s0f0, vf 28, vlanid 1250, stateDir /var/run/libvirt/hostdevmgr

2016-04-08 02:05:16.169+0000: 4773: info : virNetDevReplaceVfConfig:2390 : pflinkdev enp132s0f0, vf 28,vlanid 1250

2016-04-08 02:05:16.178+0000: 4773: info : virNetDevReplaceVfConfig:2428 : save oldmac 00:d1:d4:00:05:03, oldvlanid 1250

2016-04-08 02:05:16.178+0000: 4773: info : virNetDevSetVfConfig:2196 : ifname enp132s0f0,ifindex -1,vf 28,macaddress 00:d1:d4:00:05:03, vlanid 1250

qemu log：

kvm_alloc_slot: no free slot available

2016-04-08 06:21:04.793+0000: shutting down

[Michael liu (ztehypervisor)]

The VM has been shut off by hypervisor，so nova would use heal_instance_state function to reboot the VM.

[Michael liu (ztehypervisor)]

http://patchwork.ozlabs.org/patch/293580/
In this patch，when KVM_CAP_NR_MEMSLOTS is implemented ，Convert the static slots array to
be dynamically allocated, supporting more slots when available.
But how many slots we can use，the number is limited by the host kernel？Do any one has the same problem？

[Thomas Huth (th-huth)]

Can you please provide the complete QEMU command line when reporting QEMU bugs? Just dumping a log of Nova is not very useful for debugging QEMU problems. Thanks.

[Launchpad Janitor (janitor)]

[Expired for QEMU because there has been no activity for 60 days.]