the instance boot-from-volume evacuate failed
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
OpenStack Compute (nova) |
Confirmed
|
Undecided
|
Ding Xiao |
Bug Description
I tested the "nova evacuate" command and found a problem.
My operation steps are as below:
1.I created a volume on Cinder while cinder using lvm driver, and then launched a instance from the volume on host A.
2.When host A is broken down(or the compute service down), I use "nova evacuate instanceid-xxx new-host-name-B" command to resume the instance.
3.The instance launched failed on host B.I checked the libvirt xml, and found the diffrence as below:
the instance xml on A:
<disk type='block' device='disk'>
<driver name='qemu' type='raw' cache='none'/>
<source dev='/dev/
<target dev='vda' bus='virtio'/>
<
<alias name='virtio-
<address type='pci' domain='0x0000' bus='0x00' slot='0x04' function='0x0'/>
</disk>
the instance xml on B:
<disk type="file" device="disk">
<driver name="qemu" type="qcow2" cache="none"/>
<source file="/
<target bus="virtio" dev="vda"/>
</disk>
Changed in nova: | |
assignee: | nobody → xiaoding (xiaoding1) |
assignee: | xiaoding (xiaoding1) → nobody |
assignee: | nobody → Huawei OpenStackers (huawei-openstack) |
assignee: | Huawei OpenStackers (huawei-openstack) → nobody |
assignee: | nobody → xiaoding (xiaoding1) |
status: | New → Incomplete |
status: | Incomplete → New |
tags: | added: volumes |
description: | updated |
Changed in nova: | |
status: | New → Confirmed |
The flow is broken also when using a 3par SAN over FC.