总结热迁移失败的几种情况如下:
被热迁移的vm内存读写速度超过了内存同步的速度,让热迁移一直没办法完成内存在源节点和目的节点的同步。
We are having migration failures. The procedure was from section 3. “Migrate between hypervisors” (not a live migration). The error message is as follows:
# nova migrate --poll $u
ERROR (BadRequest): No valid host was found. No valid host found for cold
migrate (HTTP 400) (Request-ID: req-592d59db-9185-4775-b5e2-940aa657a62c)
还是按推荐的一个一个顺序热迁移。
* 0 : treated as unlimited.
* Any positive integer representing maximum concurrent builds.
"""),
# TODO(sfinucan): Add min parameter
cfg.IntOpt('max_concurrent_live_migrations',
default=1,
help="""
Maximum number of live migrations to run concurrently. This limit is enforced
to avoid outbound live migrations overwhelming the host/network and causing
failures. It is not recommended that you change this unless you are very sure
that doing so is safe and stable in your environment.
Live migration generates excessive network traffic. A live migration running at full speed can have a negative impact on the services that communicate over the same network; for example, the message queue. To set a lower maximum speed, use the following command:
# virsh migrate-setspeed domain speed_in_MBps
Alternatively, set the desired speed (again in MBps) beforehand. Example:
# openstack-config --set /etc/nova/nova.conf libvirt live_migration_bandwidth 50
libvirt改变vm 镜像的owner,除非监测到在共享存储上。libvirt有NFS, GFS2 and SMB/CIFS,但是没有Gluster。所以Gluster环境,libvirt不能检测出是在共享存储的环境,disk文件在成功迁移之后会被赋给back to root:root,然而destination host会将disk文件赋给libvirt-qemu:kvm。
存在竞合关系,所以有可能会时不时的不能再次迁移。
解决方式:
stop your guests, then stop libvirt, and edit /etc/libvirt/qemu.conf - this contains a commented out entry ‘dynamic_ownership=1’, which is the default. Change this to 0, and remove the comment. Then do a chown to libvirt-qemu:kvm for all your stopped images. Then you can start the service libvirt-bin again, and bring up the guests. Repeat on the other half of your cluster, and test a live migration.
其他解决方式:
using libgfapi, giving libvirt direct access without having to go through the filesystem. This is the preferred setup for libvirt+gluster and should also result in better I/O performance.
When migrating an instance with multipathing configured, you need to ensure consistent multipath device naming between the source and destination nodes. The migration will fail if the instance cannot resolve multipath device names in the destination node.
You can ensure consistent multipath device naming by forcing both source and destination nodes to use device WWIDs. To do this, disable user-friendly names and restart multipathd by running these commands on both source and destination nodes:
# mpathconf --enable --user_friendly_names n
# systemctl restart multipathd