Skip to content

[ci] Give the custom BYOD build a name for the agent - #65606

Merged
elliot-barn merged 1 commit into
masterfrom
elliot-barn/byod-add-host
Aug 20, 2026
Merged

elliot-barn merged 1 commit into
masterfrom
elliot-barn/byod-add-host

Conversation

@elliot-barn

Copy link
Copy Markdown
Collaborator

What

Passes --add-host rayci.localhost:host-gateway to the docker build that builds custom BYOD images, which is what wanda already passes to every image build it starts (ray-project/rayci#501).

Why

Release build 104848, on the commit that became #65578, failed hello_world_custom_byod.aws and hello_world_custom_byod.gce:

#11 6.043 error: Request failed after 3 retries in 5.7s
#11 6.043   Caused by: failed to lookup address information: No address associated with hostname
#11 ERROR: process "/bin/bash -c bash /tmp/install_python_deps.sh python_depset.lock" did not complete successfully: exit code: 2

Custom BYOD images are built here by a plain docker build, not by wanda, so they inherit none of wanda's networking flags. RAYCI_IMAGE_PIP_INDEX_URL reaches the build as http://rayci.localhost:35999/simple — a name the build has no way to resolve — and install_python_deps.sh resolves the depsets with uv, which fails hard where pip would have fallen back to PyPI. So the image build dies rather than degrading.

Everything else on that build was green (78 passed at the time, including wanda: forge, all five wheel builds, the base-extra-testdeps images, and the hello_world* release tests on AWS, GCE and released images) — this is the one path that reaches the agent proxy without going through wanda.

Unconditional, matching wanda: docker resolves host-gateway itself, and one extra hosts entry is inert for a build that never uses the name.

Testing

python -m pytest release/ray_release/tests/test_byod_build.py — 4 passed. The flag goes in after -t <image>, so the argument prefix those tests match on is unchanged.

AI assistance was used for this change.

Release 104848 failed hello_world_custom_byod.aws and .gce with

    error: Request failed after 3 retries in 5.7s
      Caused by: failed to lookup address information: No address
      associated with hostname

Custom BYOD images are built by a plain `docker build` here, not by wanda, so
they inherit none of the --add-host rayci.localhost:host-gateway that wanda
passes every image build it starts (ray-project/rayci#501). The index build
arg is a name, the build has no way to resolve it, and install_python_deps.sh
resolves these depsets with uv, which errors where pip would have fallen back
-- so the image build fails outright rather than degrading to PyPI.

Passing the same flag here makes the two build paths address the agent
identically. Unconditional, as in wanda: docker resolves host-gateway itself,
and one hosts entry is inert for a build that never uses the name.

Signed-off-by: Elliot Barnwell <elliot.barnwell@anyscale.com>
Signed-off-by: elliot-barn <elliot.barnwell@anyscale.com>
@elliot-barn
elliot-barn requested a review from a team as a code owner August 20, 2026 00:10

@gemini-code-assist gemini-code-assist Bot left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

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

Code Review

This pull request updates the custom BYOD image build command in release/ray_release/byod/build.py to include the --add-host rayci.localhost:host-gateway flag. This change ensures that the plain docker build can resolve rayci.localhost to the host gateway, preventing dependency resolution failures during the build process. There are no review comments, so I have no feedback to provide.

@elliot-barn
elliot-barn merged commit 0d8afd0 into master Aug 20, 2026
4 of 5 checks passed
@elliot-barn
elliot-barn deleted the elliot-barn/byod-add-host branch August 20, 2026 00:11
nadongjun pushed a commit to nadongjun/ray that referenced this pull request Oct 6, 2026
## What

Passes `--add-host rayci.localhost:host-gateway` to the `docker build`
that builds custom BYOD images, which is what wanda already passes to
every image build it starts (ray-project/rayci#501).

## Why

Release build 104848, on the commit that became ray-project#65578, failed
`hello_world_custom_byod.aws` and `hello_world_custom_byod.gce`:

```
ray-project#11 6.043 error: Request failed after 3 retries in 5.7s
ray-project#11 6.043   Caused by: failed to lookup address information: No address associated with hostname
ray-project#11 ERROR: process "/bin/bash -c bash /tmp/install_python_deps.sh python_depset.lock" did not complete successfully: exit code: 2
```

Custom BYOD images are built here by a plain `docker build`, not by
wanda, so they inherit none of wanda's networking flags.
`RAYCI_IMAGE_PIP_INDEX_URL` reaches the build as
`http://rayci.localhost:35999/simple` — a name the build has no way to
resolve — and `install_python_deps.sh` resolves the depsets with **uv**,
which fails hard where pip would have fallen back to PyPI. So the image
build dies rather than degrading.

Everything else on that build was green (78 passed at the time,
including `wanda: forge`, all five wheel builds, the
`base-extra-testdeps` images, and the `hello_world*` release tests on
AWS, GCE and released images) — this is the one path that reaches the
agent proxy without going through wanda.

Unconditional, matching wanda: docker resolves `host-gateway` itself,
and one extra hosts entry is inert for a build that never uses the name.

## Testing

`python -m pytest release/ray_release/tests/test_byod_build.py` — 4
passed. The flag goes in after `-t <image>`, so the argument prefix
those tests match on is unchanged.

AI assistance was used for this change.

Signed-off-by: Elliot Barnwell <elliot.barnwell@anyscale.com>
Signed-off-by: elliot-barn <elliot.barnwell@anyscale.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant