As for the nova scheduler for scheduler multiple attempts, If with certain host deployment attempt failed raise with detail exceptions, nova scheduler will choose other host to retry.
But after all attempts are tried. it will raise a Generic NoValidHost exception without a proper message. It will make nova show <instance> will not display useful information straightforward to end users.
So it's suggested to wrap the NoValidHost exception message with last attempt failure exception detail trace.
For example,
When using nova vmware driver to spawn a VM with the disk larger than the datastore upper limit, it will raise an exception
for DatastoreNotFound exception with detail, but after scheduler retries, it will got lost from nova show. So it would be friendly to have operators to view such error directly from the nova show instead for digging into the scheduler log.
filter_scheduler.py
schedule_run_instance
for num, instance_uuid in enumerate(instance_uuids):
request_spec['instance_properties']['launch_index'] = num
try:
try:
weighed_host = weighed_hosts.pop(0)
LOG.info(_("Choosing host %(weighed_host)s "
"for instance %(instance_uuid)s"),
{'weighed_host': weighed_host,
'instance_uuid': instance_uuid})
except IndexError:
raise exception.NoValidHost(reason="")
Fix proposed to branch: master
Review: https:/