Skip to content

Conversation

@dayshah
Copy link
Contributor

@dayshah dayshah commented Oct 25, 2025

Description

Cherry-picking #58063 to throw an exception when trying to double send the same ref before gc because it can trigger a NIXL error. Also adding documentation for this.

Signed-off-by: Dhyey Shah <dhyey2019@gmail.com>
Signed-off-by: Qiaolin-Yu <liin1211@outlook.com>
Signed-off-by: Qiaolin Yu <liin1211@outlook.com>
Co-authored-by: Dhyey Shah <dhyey2019@gmail.com>
Co-authored-by: Stephanie Wang <smwang@cs.washington.edu>
@dayshah dayshah added the go add ONLY when ready to merge, run all tests label Oct 25, 2025
@dayshah dayshah changed the title [core][cherry-pick] RDT NIXL Limitations + Catch [core][cherry-pick] RDT NIXL Limitations + Throw Exception Oct 25, 2025
@dayshah dayshah marked this pull request as ready for review October 25, 2025 03:52
@dayshah dayshah requested review from a team as code owners October 25, 2025 03:52
if tensor in self._tensor_to_object_ids:
raise ValueError(
f"Tensor already exists in the RDT object store. Free all references to ObjectRef({obj_id}) before storing the tensor again."
)
Copy link

Choose a reason for hiding this comment

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

Bug: Misleading Error Message in Object Addition

The error message in add_object() refers to the new obj_id being added, but should reference the existing object IDs that already own the tensor. When a tensor already exists in the store, the user needs to free the OLD object ID that contains the tensor, not the NEW obj_id being added. The error message is misleading and provides incorrect guidance to the user about which ObjectRef to free.

Fix in Cursor Fix in Web

@aslonnie aslonnie merged commit b6b1fac into ray-project:releases/2.51.0 Oct 25, 2025
5 checks passed
@dayshah dayshah deleted the rdt-limit-cherrypick branch October 25, 2025 06:18
weiquanlee pushed a commit to antgroup/ant-ray that referenced this pull request Dec 11, 2025
…ct#58159)

## Description
Cherry-picking ray-project#58063 to throw an exception when trying to double send
the same ref before gc because it can trigger a NIXL error. Also adding
documentation for this.

Signed-off-by: Dhyey Shah <dhyey2019@gmail.com>
Signed-off-by: Qiaolin-Yu <liin1211@outlook.com>
Signed-off-by: Qiaolin Yu <liin1211@outlook.com>
Co-authored-by: Qiaolin Yu <liin1211@outlook.com>
Co-authored-by: Stephanie Wang <smwang@cs.washington.edu>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

go add ONLY when ready to merge, run all tests

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants