Skip to content

tickets/DM-55285 - rucio-register performance when file is already register with Rucio #26

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
srp3rd merged 1 commit into from
Jul 6, 2026
Merged

tickets/DM-55285 - rucio-register performance when file is already register with Rucio #26

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
49 changes: 10 additions & 39 deletions python/lsst/rucio/register/rucio_interface.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -236,30 +236,6 @@ def _add_replicas(self, bundles: list[ResourceBundle]) -> None:
else:
raise Exception(f"Tried {max_retries} times and couldn't add_replicas")

def _add_file_to_dataset_with_retries(self, dataset_id, did):
retries = 0
max_retries = 5
while True:
try:
self.did_client.add_files_to_dataset(
scope=self.scope, name=dataset_id, files=[did], rse=self.rse
)
break
except rucio.common.exception.FileAlreadyExists:
if "pfn" in did:
logger.debug("file %s already registered in dataset %s", did["pfn"], dataset_id)
return # we can return, because it's already in the dataset
except rucio.common.exception.RucioException:
retries += 1
if retries < max_retries:
seconds = random.randint(10, 20)
logger.debug("failed to register one did to %s; sleeping %d seconds", dataset_id, seconds)
time.sleep(seconds)
self.did_client = DIDClient() # XXX not sure we need to do this.
else:
# we tried max_retries times, and failed, so we'll bail out
raise Exception(f"Couldn't add {did['pfn']} to dataset {dataset_id}")

def _add_files_to_dataset(self, dataset_id: str, dids: list[dict]) -> None:
"""Attach a list of files specified by Rucio DIDs to a Rucio dataset.

Expand All @@ -276,23 +252,18 @@ def _add_files_to_dataset(self, dataset_id: str, dids: list[dict]) -> None:
max_retries = 5
while True:
try:
self.did_client.add_files_to_dataset(
scope=self.scope,
name=dataset_id,
files=dids,
rse=self.rse,
self.did_client.add_files_to_datasets(
attachments=[
{
"scope": self.scope,
"name": dataset_id,
"dids": dids,
"rse": self.rse,
}
],
ignore_duplicate=True,
)
return
except rucio.common.exception.FileAlreadyExists:
# At least one already is in the dataset.
# This shouldn't happen, but if it does,
# we have to retry each individually.
for did in dids:
self._add_file_to_dataset_with_retries(
dataset_id=dataset_id,
did=did,
)
return
except rucio.common.exception.DataIdentifierNotFound as e:
raise e
except rucio.common.exception.RucioException:
Expand Down
49 changes: 3 additions & 46 deletions tests/test_interface.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -28,8 +28,6 @@
from rucio.client.didclient import DIDClient
from rucio.client.replicaclient import ReplicaClient
from rucio.common.exception import (
DataIdentifierAlreadyExists,
DataIdentifierNotFound,
FileAlreadyExists,
RucioException,
)
Expand Down Expand Up @@ -71,12 +69,14 @@ def setUp(self):
self.dc_init = patch.object(DIDClient, "__init__", return_value=None)
self.rc_add_replicas = patch.object(ReplicaClient, "add_replicas", return_value=None)
self.dc_attach_dids = patch.object(DIDClient, "attach_dids", return_value=None)
self.dc_attach_dids_to_dids = patch.object(DIDClient, "attach_dids_to_dids", return_value=None)
self.rand = patch("random.randint", return_value=1)

self.mock_rc_init = self.rc_init.start()
self.mock_dc_init = self.dc_init.start()
self.mock_rc_add_replicas = self.rc_add_replicas.start()
self.mock_dc_attach_dids = self.dc_attach_dids.start()
self.mock_dc_attach_dids_to_dids = self.dc_attach_dids_to_dids.start()
self.mock_rand = self.rand.start()

rucio_rse = "DRR1"
Expand Down Expand Up @@ -154,53 +154,10 @@ def testException1TestCase(self, MC1):
with self.assertRaises(Exception):
self.common()

@patch.object(DIDClient, "add_files_to_dataset", side_effect=RucioException("failed"))
def testException2TestCase(self, MC1):
with self.assertRaises(Exception):
self.common()

@patch.object(DIDClient, "add_files_to_dataset", side_effect=FileAlreadyExists("failed"))
def testException3TestCase(self, MC1):
def testException2TestCase(self, MC1):
self.common()

@patch.object(DIDClient, "add_dataset", return_value=None)
@patch.object(DIDClient, "add_files_to_dataset", side_effect=DataIdentifierNotFound("failed"))
def testException4TestCase(self, MC1, MC2):
with self.assertRaises(Exception):
self.common()

@patch.object(DIDClient, "add_files_to_dataset", side_effect=RucioException("failed"))
def testException5TestCase(self, MC1):
with self.assertRaises(Exception):
self.common()

@patch.object(DIDClient, "add_dataset", side_effect=DataIdentifierAlreadyExists("failed"))
@patch.object(DIDClient, "add_files_to_dataset", side_effect=DataIdentifierNotFound("failed"))
def testException6TestCase(self, MC1, MC2):
with self.assertRaises(Exception):
self.common()

@patch.object(DIDClient, "add_files_to_dataset", side_effect=RucioException("failed"))
def testException7TestCase(self, MC1):
with self.assertRaises(Exception):
self.common()

@patch.object(DIDClient, "add_dataset", side_effect=RucioException("failed"))
@patch.object(DIDClient, "add_files_to_dataset", side_effect=DataIdentifierNotFound("failed"))
def testException8TestCase(self, MC1, MC2):
with self.assertRaises(Exception):
self.common()

@patch.object(DIDClient, "add_files_to_dataset", side_effect=RucioException("failed"))
def testException9Case(self, MC1):
rucio_rse = "DRR1"
scope = "test"
dtn_url = "root://xrd1:1094//rucio"

ri = RucioInterface(self.butler, rucio_rse, scope, self.rse_root, dtn_url, DataType.DATA_PRODUCT)
with self.assertRaises(Exception):
ri._add_file_to_dataset_with_retries(None, None)

def tearDown(self):
patch.stopall()
shutil.rmtree(self.butler_repo, ignore_errors=True)
Expand Down
Loading