Skip to content
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

fix: Make the TDS size-sub-failure test more deterministic #6588

Open
wants to merge 1 commit into
base: main
Choose a base branch
from
Open

fix: Make the TDS size-sub-failure test more deterministic #6588

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
17 changes: 14 additions & 3 deletions py/server/tests/test_table_data_service.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -53,6 +53,7 @@ def __init__(self, gen_pa_table: Generator[pa.Table, None, None], pt_schema: pa.
self.partitions_size_subscriptions: Dict[TableLocationKey, bool] = {}
self.existing_partitions_called: int = 0
self.partition_size_called: int = 0
self.is_size_sub_failure_cb_called: bool = False

def table_schema(self, table_key: TableKeyImpl,
schema_cb: Callable[[pa.Schema, Optional[pa.Schema]], None],
Expand Down Expand Up @@ -164,10 +165,12 @@ def _th_partition_size_changes(self, table_key: TableKeyImpl, table_location_key
rbs.append(pa_table.to_batches()[0])
new_pa_table = pa.Table.from_batches(rbs)
self.partitions[table_location_key] = new_pa_table
size_cb(new_pa_table.num_rows)
if self.sub_partition_size_fail_test:
failure_cb(Exception("table location size subscription failure"))
self.is_size_sub_failure_cb_called = True
return
else:
size_cb(new_pa_table.num_rows)
time.sleep(0.1)

def subscribe_to_table_location_size(self, table_key: TableKeyImpl,
Expand Down Expand Up @@ -357,9 +360,17 @@ def test_partition_size_sub_failure(self):
data_service = TableDataService(backend)
backend.sub_partition_size_fail_test = True
table = data_service.make_table(TableKeyImpl("test"), refreshing=True)
table.coalesce()
# wait for the size subscription failure callback to be called for 5 seconds
for _ in range(50):
time.sleep(0.1)
if backend.is_size_sub_failure_cb_called:
break
else:
self.fail("size subscription failure callback was not called in 5s")
Comment on lines +363 to +370
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This is probably better than what you had, although I really don't like seeing this kind of sleep loop; does Python not have condition variables?


with self.assertRaises(Exception) as cm:
# failure_cb will be called in the background thread after 2 PUG cycles, 3 seconds timeout should be enough
self.wait_ticking_table_update(table, 600, 3)
self.wait_ticking_table_update(table, 600, 1)
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I suspect you need 2 seconds (2 cycles) to be safe if you're using a real PeriodicUpdateGraph with a 1 second interval. The failure is buffered after the roots are processed on one cycle, it won't be delivered until the next.


self.assertTrue(table.j_table.isFailed())

Expand Down