Repository navigation
fix(prometheus): Avoid appending duplicate pods - #123
Merged
Merged
Conversation
LeaveMyYard
approved these changes
Aug 10, 2023
Sheeproid
added a commit
that referenced
this pull request
Oct 19, 2026
#556) ## Summary - **urllib3** 2.7.0 → 2.8.0 in `pyproject.toml`, `poetry.lock`, `requirements.txt` and `enforcer/requirements.txt`. Fixes CVE-2026-97687 and CVE-2026-97689 (High). Closes dependabot alerts #122, #123, #125, #126, #128 and #129, and the matching Artifact Registry findings on `devel/krr`. - **Dockerfile**: pip is uninstalled once the requirements are installed. pip 26.2.1, the latest release, vendors its own urllib3 2.7.0, so the dependency bump alone left the CVEs in the image. krr never calls pip at runtime. The pip upgrade, requirements install and uninstall run in one `RUN`, so the upgraded pip never lands in an image layer (CodeRabbit's finding, now fixed). Linear: [ROB-1636](https://linear.app/robusta/issue/ROB-1636) ## Verification - `poetry lock --no-update` with Poetry 1.8.5: only urllib3 changes in the lock. `poetry check --lock` passes. - `pytest`: 55 passed. - Built the image for `linux/amd64` and scanned it with Trivy (Docker Scout isn't available in the sandbox). The Python packages have no HIGH or CRITICAL findings. Before this change the image showed urllib3 2.7.0 (pip's vendored copy) and msgpack 1.1.2. `docker history` shows no separate pip layer. - Built for `linux/arm64` under QEMU (from the commit before the single-`RUN` change, so the same packages): it runs on `aarch64`, Trivy reports the same results as amd64, urllib3 is 2.8.0, and pip is gone. - In both images `krr.py simple --help` works. - Not tested: a rescan in Artifact Registry. The Vanta finding clears only after CI publishes the image. Out of scope: - `gcc-14` CVE-2026-95619 (Low, no Debian fix) and 10 unfixed Debian HIGHs, unchanged from `main`. - The `enforcer` image installs pip the same way, but Vanta has not flagged it. - The enforcer README's "Python 3.9+" was already wrong before this PR. It's tracked in [ROB-1650](https://linear.app/robusta/issue/ROB-1650). --- _Generated by [Claude Code](https://claude.ai/code/session_01Fq7hqRqj2kshg8M2sFpN7h)_ --------- Signed-off-by: Claude <noreply@anthropic.com> Co-authored-by: Claude <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Hi, I noticed an incorrect pod count that causes incorrect results. I fixed the issue in this PR by simply converting the pod list to a
set.Here's my results before and after the fix:
Without filtering out deleted pods: