help='Perform copy even if the object appears to exist at the remote destination.')
copy_opts.add_argument(
'--src', dest='source_arvados',
- help='The name of the source Arvados instance (required) - points at an Arvados config file. May be either a pathname to a config file, or (for example) "foo" as shorthand for $HOME/.config/arvados/foo.conf.')
+ help='The cluster id of the source Arvados instance. May be either a pathname to a config file, or (for example) "foo" as shorthand for $HOME/.config/arvados/foo.conf. If not provided, will be inferred from the UUID of the object being copied.')
copy_opts.add_argument(
'--dst', dest='destination_arvados',
- help='The name of the destination Arvados instance (required) - points at an Arvados config file. May be either a pathname to a config file, or (for example) "foo" as shorthand for $HOME/.config/arvados/foo.conf.')
+ help='The name of the destination Arvados instance (required). May be either a pathname to a config file, or (for example) "foo" as shorthand for $HOME/.config/arvados/foo.conf. If not provided, will use ARVADOS_API_HOST from environment.')
copy_opts.add_argument(
'--recursive', dest='recursive', action='store_true',
help='Recursively copy any dependencies for this object, and subprojects. (default)')
copy_opts.add_argument(
'--project-uuid', dest='project_uuid',
help='The UUID of the project at the destination to which the collection or workflow should be copied.')
+ copy_opts.add_argument(
+ '--storage-classes', dest='storage_classes',
+ help='Comma separated list of storage classes to be used when saving data to the destinaton Arvados instance.')
copy_opts.add_argument(
'object_uuid',
copy_opts.set_defaults(recursive=True)
parser = argparse.ArgumentParser(
- description='Copy a workflow or collection from one Arvados instance to another.',
+ description='Copy a workflow, collection or project from one Arvados instance to another. On success, the uuid of the copied object is printed to stdout.',
parents=[copy_opts, arv_cmd.retry_opt])
args = parser.parse_args()
+ if args.storage_classes:
+ args.storage_classes = [x for x in args.storage_classes.strip().replace(' ', '').split(',') if x]
+
if args.verbose:
logger.setLevel(logging.DEBUG)
else:
args.source_arvados = args.object_uuid[:5]
# Create API clients for the source and destination instances
- src_arv = api_for_instance(args.source_arvados)
- dst_arv = api_for_instance(args.destination_arvados)
+ src_arv = api_for_instance(args.source_arvados, args.retries)
+ dst_arv = api_for_instance(args.destination_arvados, args.retries)
if not args.project_uuid:
args.project_uuid = dst_arv.users().current().execute(num_retries=args.retries)["uuid"]
logger.error("API server returned an error result: {}".format(result))
exit(1)
- logger.info("")
+ print(result['uuid'])
+
+ if result.get('partial_error'):
+ logger.warning("Warning: created copy with uuid {} but failed to copy some items: {}".format(result['uuid'], result['partial_error']))
+ exit(1)
+
logger.info("Success: created copy with uuid {}".format(result['uuid']))
exit(0)
# Otherwise, it is presumed to be the name of a file in
# $HOME/.config/arvados/instance_name.conf
#
-def api_for_instance(instance_name):
+def api_for_instance(instance_name, num_retries):
if not instance_name:
# Use environment
return arvados.api('v1', model=OrderedJsonModel())
host=cfg['ARVADOS_API_HOST'],
token=cfg['ARVADOS_API_TOKEN'],
insecure=api_is_insecure,
- model=OrderedJsonModel())
+ model=OrderedJsonModel(),
+ num_retries=num_retries,
+ )
else:
abort('need ARVADOS_API_HOST and ARVADOS_API_TOKEN for {}'.format(instance_name))
return client
# fetch the workflow from the source instance
wf = src.workflows().get(uuid=wf_uuid).execute(num_retries=args.retries)
+ if not wf["definition"]:
+ logger.warning("Workflow object {} has an empty or null definition, it won't do anything.".format(wf_uuid))
+
# copy collections and docker images
- if args.recursive:
+ if args.recursive and wf["definition"]:
wf_def = yaml.safe_load(wf["definition"])
if wf_def is not None:
locations = []
if not body["name"]:
body['name'] = "copied from " + collection_uuid
+ if args.storage_classes:
+ body['storage_classes_desired'] = args.storage_classes
+
body['owner_uuid'] = args.project_uuid
dst_collection = dst.collections().create(body=body, ensure_unique_name=True).execute(num_retries=args.retries)
if progress_writer:
progress_writer.report(obj_uuid, bytes_written, bytes_expected)
data = src_keep.get(word)
- dst_locator = dst_keep.put(data)
+ dst_locator = dst_keep.put(data, classes=(args.storage_classes or []))
dst_locators[blockhash] = dst_locator
bytes_written += loc.size
dst_manifest.write(' ')
logger.debug('Copying %s to %s', obj_uuid, project_record["uuid"])
+
+ partial_error = ""
+
# Copy collections
- copy_collections([col["uuid"] for col in arvados.util.list_all(src.collections().list, filters=[["owner_uuid", "=", obj_uuid]])],
- src, dst, args)
+ try:
+ copy_collections([col["uuid"] for col in arvados.util.list_all(src.collections().list, filters=[["owner_uuid", "=", obj_uuid]])],
+ src, dst, args)
+ except Exception as e:
+ partial_error += "\n" + str(e)
# Copy workflows
for w in arvados.util.list_all(src.workflows().list, filters=[["owner_uuid", "=", obj_uuid]]):
- copy_workflow(w["uuid"], src, dst, args)
+ try:
+ copy_workflow(w["uuid"], src, dst, args)
+ except Exception as e:
+ partial_error += "\n" + "Error while copying %s: %s" % (w["uuid"], e)
if args.recursive:
for g in arvados.util.list_all(src.groups().list, filters=[["owner_uuid", "=", obj_uuid]]):
- copy_project(g["uuid"], src, dst, project_record["uuid"], args)
+ try:
+ copy_project(g["uuid"], src, dst, project_record["uuid"], args)
+ except Exception as e:
+ partial_error += "\n" + "Error while copying %s: %s" % (g["uuid"], e)
+
+ project_record["partial_error"] = partial_error
return project_record