gitadmin/foxhunt
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

fix(ci): correct KUBERNETES_NODE_SELECTOR format to key=value

Browse Source 
GitLab Runner expects KUBERNETES_NODE_SELECTOR_<NAME>: "key=value" format,
not KUBERNETES_NODE_SELECTOR_<key>: "value". Fix both compile (ci-compile-cpu)
and supervised training (ci-training) node selector overrides.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
This commit is contained in:
jgrusewski
2026-03-01 12:27:55 +01:00
parent 02850c210d
commit f6cc67378c
1 changed files with 2 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
.gitlab-ci.yml
View File

@@ -292,7 +292,7 @@ build-foxhunt-training-runtime:
variables:
# No CUDA_COMPUTE_CAP — CPU-only builds don't compile CUDA kernels
CARGO_BUILD_JOBS: "14"
KUBERNETES_NODE_SELECTOR_k8s.scaleway.com/pool-name: "ci-compile-cpu"
KUBERNETES_NODE_SELECTOR_POOL: "k8s.scaleway.com/pool-name=ci-compile-cpu"
KUBERNETES_CPU_REQUEST: "14000m"
KUBERNETES_CPU_LIMIT: "15500m"
KUBERNETES_MEMORY_REQUEST: "24Gi"
@@ -578,7 +578,7 @@ write-manifest:
# nvidia runtime provided by runner default (runtime_class_name = "nvidia")
# Route to L40S pool (48GB VRAM for large supervised models)
# Overrides runner default to L40S pool
KUBERNETES_NODE_SELECTOR_k8s.scaleway.com/pool-name: "ci-training"
KUBERNETES_NODE_SELECTOR_POOL: "k8s.scaleway.com/pool-name=ci-training"
# Supervised is GPU-bound — minimal CPU. Low request allows 3-4 concurrent jobs.
KUBERNETES_CPU_REQUEST: "1000m"
KUBERNETES_CPU_LIMIT: "2000m"