>On Sat, Jul 22, 2017 at 05:45:59 -0400, Yi Wang wrote:

>> This patch fix this condition:

>>   -vm has the "auto" placement in vcpu

>>   -hot-plug memory with source node "1-3" through attach-device command

>>   -migrate the vm to a host with only 2 numa node

>> And the migration will fail with error:

>> "error: unsupported configuration: NUMA node 2 is unavailable"

>> 

>> Signed-off-by: Yi Wang <wang yi59 zte com cn>

>> Signed-off-by: Xi Xu <xu xi8 zte com cn>

>> ---

>>  src/qemu/qemu_process.c | 10 ++++++++++

>>  1 file changed, 10 insertions(+)

>> 

>> diff --git a/src/qemu/qemu_process.c b/src/qemu/qemu_process.c

>> index 7b708be..dcc564c 100644

>> --- a/src/qemu/qemu_process.c

>> +++ b/src/qemu/qemu_process.c

>> @@ -5259,6 +5259,16 @@ qemuProcessPrepareDomain(virConnectPtr conn,

>>              goto cleanup;

>>      }

>>  

>> +    VIR_DEBUG("Updating memory source nodes");

>> +    for (i = 0; i < vm->def->nmems; i++) {

>> +        virDomainMemoryDefPtr mem = vm->def->mems[i];

>> +        if (priv->autoNodeset && mem && mem->sourceNodes) {

>> +            virBitmapFree(mem->sourceNodes);

>> +            if (!(mem->sourceNodes = virBitmapNewCopy(priv->autoNodeset)))

>> +                goto cleanup;

>

>This is not correct. This code will be executed even during normal

>startup and it would remove any manual pinning the user set-up.

What you said is right, this is indeed a problem that my patch ignored.

>

>I think the problem might be that the XML retains the nodesets of

>automatically placed memory modules while formatting the migratable XML.

The problem is that the source nodemask specified in the "virsh attach-device"

XML, which doesn't exist in the destination host. For example:

    <memory model='dimm'>

        <target>

            <size unit='MiB'>1024</size>

            <node>0</node>

        </target>

        <source>

            <nodemask>1-3</nodemask>

        </source>

    </memory>

The source of the problem is that we must set source nodemask when hot-plug

memory dimm when the VM has auto placement, or it failed:

# virsh attach-device centos mem_hp.xml

error: Failed to attach device from mem_hp.xml

error: internal error: Advice from numad is needed in case of automatic numa placement

And after we set source nodemask, it might fail to migrate.

So, a better way to work around this problem is add a "auto" supported to <source>?

Any other suggestion?