[GAUDISW-246357] UBI images improvements by ghandoura · Pull Request #971 · vllm-project/vllm-gaudi

ghandoura · 2026-02-13T10:51:19Z

Use pip check to verify python dependencies during build
Use no-cache to reduce the docker image size
Allow using latest in sysnapse revision
TODO remove CRB
Test the changes

More changes to check and implement:

1) Around line 46, is this RUN dnf install -y python3-dnf-plugin-versionlock. It needs to start off with dnf -y update to pull in all updates before centos/epel repos get installed.
2) At line 62, it tries to install libjpeg-devel. The replacement on RHEL 9 is libjpeg-turbo-devel. It’s a drop in replacement that runs even faster. And it is in the main UBI repo and not epel.
3) Down around line 130, is a dnf -y update. Delete this block as this will pull in all kinds of Centos contamination.
4) Not strictly required, but at line 139, make it FROM gaudi-pytorch as vllm-openai to match other vllm images.
5) At line 153, there is another dnf -y update, remove it for the same reason as [FIX for upstream changes ]hpu_model_runner and add UT #3
retest

github-actions · 2026-02-13T10:51:30Z

🚧 CI Blocked

The main CI workflow was not started for the following reason:

Your branch is behind the base branch. Please merge or rebase to get the latest changes.

Copilot

Pull request overview

This PR implements improvements to the RHEL UBI (Red Hat Universal Base Image) Dockerfile for vLLM on Habana Gaudi hardware. The changes enhance the build process by adding dependency verification, reducing image size, and introducing flexible version management for Synapse packages.

Changes:

Added pip check command after vLLM installation to verify Python dependency compatibility
Added --no-cache-dir flag to all pip install commands to reduce Docker image size
Implemented support for using "latest" as a value for SYNAPSE_REVISION to automatically detect and use the newest available revision

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot

Pull request overview

Copilot reviewed 1 out of 1 changed files in this pull request and generated 1 comment.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Use --no-cache-dir for pip installs to reduce image size. Run pip check during build to validate Python dependencies. Allow SYNAPSE_REVISION as exact value (e.g. 695) or latest with revision detection. Signed-off-by: Adam Ghandoura <adam.ghandoura@intel.com>

Signed-off-by: Adam Ghandoura <adam.ghandoura@intel.com>

github-actions · 2026-02-17T13:04:13Z

🚧 CI Blocked

The main CI workflow was not started for the following reason:

Your branch is behind the base branch. Please merge or rebase to get the latest changes.

github-actions · 2026-02-17T16:27:31Z

✅ CI Passed

All checks passed successfully against the following vllm commit:
17b17c068453e6dc6af79240bb94857ae175cc51

github-actions · 2026-02-18T15:38:22Z

🚧 CI Blocked

The main CI workflow was not started for the following reason:

Your branch is behind the base branch. Please merge or rebase to get the latest changes.

Signed-off-by: Adam Ghandoura <adam.ghandoura@intel.com>

github-actions · 2026-02-18T15:56:58Z

🚧 CI Blocked

The main CI workflow was not started for the following reason:

Your branch is behind the base branch. Please merge or rebase to get the latest changes.

ghandoura · 2026-02-18T16:11:11Z

CRB is back because of new libftd-devel dependency

Signed-off-by: Adam Ghandoura <adam.ghandoura@intel.com>

github-actions · 2026-02-18T16:16:07Z

🚧 CI Blocked

The main CI workflow was not started for the following reason:

Your branch is behind the base branch. Please merge or rebase to get the latest changes.

Signed-off-by: Adam Ghandoura <adam.ghandoura@intel.com>

github-actions · 2026-02-18T16:22:33Z

🚧 CI Blocked

The main CI workflow was not started for the following reason:

Your branch is behind the base branch. Please merge or rebase to get the latest changes.

github-actions · 2026-02-18T16:22:54Z

🚧 CI Blocked

The main CI workflow was not started for the following reason:

Your branch is behind the base branch. Please merge or rebase to get the latest changes.

Signed-off-by: Adam Ghandoura <adam.ghandoura@intel.com>

github-actions · 2026-02-18T17:01:18Z

🚧 CI Blocked

The main CI workflow was not started for the following reason:

Your branch is behind the base branch. Please merge or rebase to get the latest changes.

github-actions · 2026-02-19T10:27:59Z

🚧 CI Blocked

The main CI workflow was not started for the following reason:

Your branch is behind the base branch. Please merge or rebase to get the latest changes.

PatrykWo

Build tested.

- Use pip check to verify python dependencies during build - Use no-cache to reduce the docker image size - Allow using `latest` in sysnapse revision - [x] TODO remove CRB - [x] Test the changes More changes to check and implement: - [x] 1) Around line 46, is this RUN dnf install -y python3-dnf-plugin-versionlock. It needs to start off with dnf -y update to pull in all updates before centos/epel repos get installed. - [x] 2) At line 62, it tries to install libjpeg-devel. The replacement on RHEL 9 is libjpeg-turbo-devel. It’s a drop in replacement that runs even faster. And it is in the main UBI repo and not epel. - [x] 3) Down around line 130, is a dnf -y update. Delete this block as this will pull in all kinds of Centos contamination. - [x] 4) Not strictly required, but at line 139, make it FROM gaudi-pytorch as vllm-openai to match other vllm images. - [x] 5) At line 153, there is another dnf -y update, remove it for the same reason as vllm-project#3 - [x] retest --------- Signed-off-by: Adam Ghandoura <adam.ghandoura@intel.com> Co-authored-by: Michał Kuligowski <michal.kuligowski@intel.com> Co-authored-by: Agata Dobrzyniewicz <160237065+adobrzyn@users.noreply.github.com> Co-authored-by: Patryk Wolsza <patryk.wolsza@intel.com> Signed-off-by: Rohit kumar Singh <rksingh@habana.ai>

- Use pip check to verify python dependencies during build - Use no-cache to reduce the docker image size - Allow using `latest` in sysnapse revision - [x] TODO remove CRB - [x] Test the changes More changes to check and implement: - [x] 1) Around line 46, is this RUN dnf install -y python3-dnf-plugin-versionlock. It needs to start off with dnf -y update to pull in all updates before centos/epel repos get installed. - [x] 2) At line 62, it tries to install libjpeg-devel. The replacement on RHEL 9 is libjpeg-turbo-devel. It’s a drop in replacement that runs even faster. And it is in the main UBI repo and not epel. - [x] 3) Down around line 130, is a dnf -y update. Delete this block as this will pull in all kinds of Centos contamination. - [x] 4) Not strictly required, but at line 139, make it FROM gaudi-pytorch as vllm-openai to match other vllm images. - [x] 5) At line 153, there is another dnf -y update, remove it for the same reason as vllm-project#3 - [x] retest --------- Signed-off-by: Adam Ghandoura <adam.ghandoura@intel.com> Co-authored-by: Michał Kuligowski <michal.kuligowski@intel.com> Co-authored-by: Agata Dobrzyniewicz <160237065+adobrzyn@users.noreply.github.com> Co-authored-by: Patryk Wolsza <patryk.wolsza@intel.com>

- Use pip check to verify python dependencies during build - Use no-cache to reduce the docker image size - Allow using `latest` in sysnapse revision - [x] TODO remove CRB - [x] Test the changes More changes to check and implement: - [x] 1) Around line 46, is this RUN dnf install -y python3-dnf-plugin-versionlock. It needs to start off with dnf -y update to pull in all updates before centos/epel repos get installed. - [x] 2) At line 62, it tries to install libjpeg-devel. The replacement on RHEL 9 is libjpeg-turbo-devel. It’s a drop in replacement that runs even faster. And it is in the main UBI repo and not epel. - [x] 3) Down around line 130, is a dnf -y update. Delete this block as this will pull in all kinds of Centos contamination. - [x] 4) Not strictly required, but at line 139, make it FROM gaudi-pytorch as vllm-openai to match other vllm images. - [x] 5) At line 153, there is another dnf -y update, remove it for the same reason as #3 - [x] retest --------- Signed-off-by: Adam Ghandoura <adam.ghandoura@intel.com> Co-authored-by: Michał Kuligowski <michal.kuligowski@intel.com> Co-authored-by: Agata Dobrzyniewicz <160237065+adobrzyn@users.noreply.github.com> Co-authored-by: Patryk Wolsza <patryk.wolsza@intel.com> (cherry picked from commit a3855ac)

- Use pip check to verify python dependencies during build - Use no-cache to reduce the docker image size - Allow using `latest` in sysnapse revision - [x] TODO remove CRB - [x] Test the changes More changes to check and implement: - [x] 1) Around line 46, is this RUN dnf install -y python3-dnf-plugin-versionlock. It needs to start off with dnf -y update to pull in all updates before centos/epel repos get installed. - [x] 2) At line 62, it tries to install libjpeg-devel. The replacement on RHEL 9 is libjpeg-turbo-devel. It’s a drop in replacement that runs even faster. And it is in the main UBI repo and not epel. - [x] 3) Down around line 130, is a dnf -y update. Delete this block as this will pull in all kinds of Centos contamination. - [x] 4) Not strictly required, but at line 139, make it FROM gaudi-pytorch as vllm-openai to match other vllm images. - [x] 5) At line 153, there is another dnf -y update, remove it for the same reason as #3 - [x] retest --------- Signed-off-by: Adam Ghandoura <adam.ghandoura@intel.com> Co-authored-by: Michał Kuligowski <michal.kuligowski@intel.com> Co-authored-by: Agata Dobrzyniewicz <160237065+adobrzyn@users.noreply.github.com> Co-authored-by: Patryk Wolsza <patryk.wolsza@intel.com> (cherry picked from commit a3855ac) Signed-off-by: PatrykWo <patryk.wolsza@intel.com>

…0.15.1 (#1049) ## Summary This PR cherry-picks all RHEL/UBI Dockerfile changes merged to `main` after `releases/v0.15.1` into the v0.15.1 release branch. ## Cherry-picked PRs | PR | Commit | Description | |----|--------|-------------| | [#923](#923) | `6d15fdc` | [GAUDISW-244821] Modify UBI docker to support both internal and external builds | | [#811](#811) | `a0a0d36` | Fix reported version of vllm | | [#713](#713) | `40a425f` | Create UBI based vLLM docker build instructions | | [#974](#974) | `6db03ad` | [GADC-941] Add libfdt-devel (new habanalabs-thunk dependency) to UBI Dockerfile | | [#971](#971) | `a3855ac` | [GAUDISW-246357] UBI images improvements | | [#1008](#1008) | `b3b2fb3` | Fix Dockerfile for RHEL 9.6 build by updating package installation order | ## Key changes in `.cd/Dockerfile.rhel.ubi.vllm` - Added new build args: `OS_VERSION`, `OS_STRING`, `PT_MODULES_REPO_NAME`, `PT_PACKAGE_NAME_NON_DEFAULT_PYTHON_SUBSTRING`, `PYPI_INDEX_URL`, `HABANA_RPM_REPO_PATH` - Support `SYNAPSE_REVISION=latest` (auto-detects newest available revision) - Detected Synapse revision stored in `/etc/habanalabs/synapse_revision` for use across stages - `dnf install` uses `--allowerasing` throughout; `openssl-fips-provider-so` removal has `|| true` to support RHEL 9.4 - Added packages: `libomp`, `libjpeg-turbo-devel` (replaces `libjpeg-devel`), `libfdt-devel` - `pip` calls use `--no-cache-dir` - vLLM install: replaced `use_existing_torch.py` with `pip install -r <(sed '/^torch/d' requirements/build.txt)` - `pip check` added after installation - Final stage renamed to `AS vllm-openai` - `OS_STRING` is now parametric (supports both RHEL 9.4 and 9.6) --------- Signed-off-by: Michal Muszynski <mmuszynski@habana.ai> Signed-off-by: PatrykWo <patryk.wolsza@intel.com> Signed-off-by: Adam Ghandoura <adam.ghandoura@intel.com> Signed-off-by: mhelf-intel <monika.helfer@intel.com> Signed-off-by: Michal Muszynski <michal.muszynski@intel.com> Co-authored-by: Michal Muszynski <141021743+mmuszynskihabana@users.noreply.github.com> Co-authored-by: Adam Ghandoura <adam.ghandoura@intel.com> Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com> Co-authored-by: Iryna Boiko <iboiko@habana.ai> Co-authored-by: aghandoura <adam.ghandoura@gmail.com> Co-authored-by: mhelf-intel <monika.helfer@intel.com> Co-authored-by: Michał Kuligowski <michal.kuligowski@intel.com> Co-authored-by: Agata Dobrzyniewicz <160237065+adobrzyn@users.noreply.github.com>

…0.15.1 (#1049) This PR cherry-picks all RHEL/UBI Dockerfile changes merged to `main` after `releases/v0.15.1` into the v0.15.1 release branch. | PR | Commit | Description | |----|--------|-------------| | [#923](#923) | `6d15fdc` | [GAUDISW-244821] Modify UBI docker to support both internal and external builds | | [#811](#811) | `a0a0d36` | Fix reported version of vllm | | [#713](#713) | `40a425f` | Create UBI based vLLM docker build instructions | | [#974](#974) | `6db03ad` | [GADC-941] Add libfdt-devel (new habanalabs-thunk dependency) to UBI Dockerfile | | [#971](#971) | `a3855ac` | [GAUDISW-246357] UBI images improvements | | [#1008](#1008) | `b3b2fb3` | Fix Dockerfile for RHEL 9.6 build by updating package installation order | - Added new build args: `OS_VERSION`, `OS_STRING`, `PT_MODULES_REPO_NAME`, `PT_PACKAGE_NAME_NON_DEFAULT_PYTHON_SUBSTRING`, `PYPI_INDEX_URL`, `HABANA_RPM_REPO_PATH` - Support `SYNAPSE_REVISION=latest` (auto-detects newest available revision) - Detected Synapse revision stored in `/etc/habanalabs/synapse_revision` for use across stages - `dnf install` uses `--allowerasing` throughout; `openssl-fips-provider-so` removal has `|| true` to support RHEL 9.4 - Added packages: `libomp`, `libjpeg-turbo-devel` (replaces `libjpeg-devel`), `libfdt-devel` - `pip` calls use `--no-cache-dir` - vLLM install: replaced `use_existing_torch.py` with `pip install -r <(sed '/^torch/d' requirements/build.txt)` - `pip check` added after installation - Final stage renamed to `AS vllm-openai` - `OS_STRING` is now parametric (supports both RHEL 9.4 and 9.6) --------- Signed-off-by: Michal Muszynski <mmuszynski@habana.ai> Signed-off-by: PatrykWo <patryk.wolsza@intel.com> Signed-off-by: Adam Ghandoura <adam.ghandoura@intel.com> Signed-off-by: mhelf-intel <monika.helfer@intel.com> Signed-off-by: Michal Muszynski <michal.muszynski@intel.com> Co-authored-by: Michal Muszynski <141021743+mmuszynskihabana@users.noreply.github.com> Co-authored-by: Adam Ghandoura <adam.ghandoura@intel.com> Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com> Co-authored-by: Iryna Boiko <iboiko@habana.ai> Co-authored-by: aghandoura <adam.ghandoura@gmail.com> Co-authored-by: mhelf-intel <monika.helfer@intel.com> Co-authored-by: Michał Kuligowski <michal.kuligowski@intel.com> Co-authored-by: Agata Dobrzyniewicz <160237065+adobrzyn@users.noreply.github.com>

- Use pip check to verify python dependencies during build - Use no-cache to reduce the docker image size - Allow using `latest` in sysnapse revision - [x] TODO remove CRB - [x] Test the changes More changes to check and implement: - [x] 1) Around line 46, is this RUN dnf install -y python3-dnf-plugin-versionlock. It needs to start off with dnf -y update to pull in all updates before centos/epel repos get installed. - [x] 2) At line 62, it tries to install libjpeg-devel. The replacement on RHEL 9 is libjpeg-turbo-devel. It’s a drop in replacement that runs even faster. And it is in the main UBI repo and not epel. - [x] 3) Down around line 130, is a dnf -y update. Delete this block as this will pull in all kinds of Centos contamination. - [x] 4) Not strictly required, but at line 139, make it FROM gaudi-pytorch as vllm-openai to match other vllm images. - [x] 5) At line 153, there is another dnf -y update, remove it for the same reason as #3 - [x] retest --------- Signed-off-by: Adam Ghandoura <adam.ghandoura@intel.com> Co-authored-by: Michał Kuligowski <michal.kuligowski@intel.com> Co-authored-by: Agata Dobrzyniewicz <160237065+adobrzyn@users.noreply.github.com> Co-authored-by: Patryk Wolsza <patryk.wolsza@intel.com>

Copilot AI review requested due to automatic review settings February 13, 2026 10:51

ghandoura requested review from adobrzyn, afierka-intel, iboiko-habana, kamil-kaczor, ksmusz, mgawarkiewicz-intel, michalkuligowski and xuechendi as code owners February 13, 2026 10:51

Copilot started reviewing on behalf of ghandoura February 13, 2026 10:51 View session

Copilot AI reviewed Feb 13, 2026

View reviewed changes

Comment thread .cd/Dockerfile.rhel.ubi.vllm Outdated

Comment thread .cd/Dockerfile.rhel.ubi.vllm Outdated

ghandoura force-pushed the ubi_improvements branch from 5284c46 to 18ca0d8 Compare February 13, 2026 11:54

ghandoura requested a review from Copilot February 13, 2026 11:59

Copilot started reviewing on behalf of ghandoura February 13, 2026 11:59 View session

Copilot AI reviewed Feb 13, 2026

View reviewed changes

Comment thread .cd/Dockerfile.rhel.ubi.vllm

github-actions Bot mentioned this pull request Feb 13, 2026

🚦 Team Review Dashboard #701

Open

ghandoura force-pushed the ubi_improvements branch from f19d1f0 to 03bb55d Compare February 13, 2026 15:39

ghandoura added 3 commits February 13, 2026 16:41

ubi image: add comment about image lable

453108f

Signed-off-by: Adam Ghandoura <adam.ghandoura@intel.com>

remove crb repo and add allow-erasing

41a2cb3

Signed-off-by: Adam Ghandoura <adam.ghandoura@intel.com>

ghandoura force-pushed the ubi_improvements branch from 03bb55d to 41a2cb3 Compare February 13, 2026 15:41

Merge branch 'main' into ubi_improvements

9c472de

Merge branch 'main' into ubi_improvements

0d2080b

more fixes

310ece6

Signed-off-by: Adam Ghandoura <adam.ghandoura@intel.com>

ghandoura force-pushed the ubi_improvements branch from 2eb6e82 to 310ece6 Compare February 18, 2026 15:54

add crb back because of libftd-devel

bf22daf

Signed-off-by: Adam Ghandoura <adam.ghandoura@intel.com>

rename last layer

f3ebd9b

Signed-off-by: Adam Ghandoura <adam.ghandoura@intel.com>

ghandoura force-pushed the ubi_improvements branch from f66f1a3 to f3ebd9b Compare February 18, 2026 16:22

fix typo

8b7fbb6

Signed-off-by: Adam Ghandoura <adam.ghandoura@intel.com>

PatrykWo self-assigned this Feb 19, 2026

PatrykWo added the skip-gaudi-tests label Feb 19, 2026

PatrykWo approved these changes Feb 19, 2026

View reviewed changes

Merge branch 'main' into ubi_improvements

2c82e75

mgawarkiewicz-intel approved these changes Feb 19, 2026

View reviewed changes

afierka-intel merged commit a3855ac into vllm-project:main Feb 19, 2026
16 checks passed

PatrykWo mentioned this pull request Feb 26, 2026

[Cherry-pick] Backport RHEL/UBI Dockerfile improvements to releases/v0.15.1 #1049

Merged

PatrykWo mentioned this pull request Feb 26, 2026

Fix RHEL Dockerfile build order and remove obsolete TencentOS Dockerfile #1056

Merged

Conversation

ghandoura commented Feb 13, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

github-actions Bot commented Feb 13, 2026

🚧 CI Blocked

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Uh oh!

github-actions Bot commented Feb 17, 2026

🚧 CI Blocked

Uh oh!

github-actions Bot commented Feb 17, 2026

✅ CI Passed

Uh oh!

github-actions Bot commented Feb 18, 2026

🚧 CI Blocked

Uh oh!

github-actions Bot commented Feb 18, 2026

🚧 CI Blocked

Uh oh!

ghandoura commented Feb 18, 2026

Uh oh!

github-actions Bot commented Feb 18, 2026

🚧 CI Blocked

Uh oh!

github-actions Bot commented Feb 18, 2026

🚧 CI Blocked

Uh oh!

github-actions Bot commented Feb 18, 2026

🚧 CI Blocked

Uh oh!

github-actions Bot commented Feb 18, 2026

🚧 CI Blocked

Uh oh!

github-actions Bot commented Feb 19, 2026

🚧 CI Blocked

Uh oh!

PatrykWo left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

7 participants

ghandoura commented Feb 13, 2026 •

edited

Loading