Skip to content
Merged
Changes from all commits
Commits
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
5 changes: 4 additions & 1 deletion checkpoint_engine/ps.py
Original file line number Diff line number Diff line change
Expand Up @@ -1287,6 +1287,7 @@ class UpdateRequest(BaseModel):
update_url: str | None = None
inference_group_ranks: list[int] = []
timeout: float = 300.0
uds: str | None = None

def wrap_exception(func: Callable[[], None]) -> Response:
try:
Expand Down Expand Up @@ -1319,7 +1320,9 @@ def update_func(socket_paths: list[tuple[str, str]]):
return
if req.inference_group_ranks:
socket_paths = [socket_paths[i] for i in req.inference_group_ranks]
request_inference_to_update(req.update_url, dict(socket_paths), timeout=req.timeout)
request_inference_to_update(
req.update_url, dict(socket_paths), timeout=req.timeout, uds=req.uds
)

return wrap_exception(lambda: ps.update(checkpoint_name, update_func, ranks=req.ranks))

Expand Down