Skip to content

Share one Flux allocation per child pipeline on Tioga and Tuolumne #548

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
ZwFink merged 1 commit into from
Oct 7, 2026
Merged

Share one Flux allocation per child pipeline on Tioga and Tuolumne #548

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
25 changes: 21 additions & 4 deletions .gitlab/custom-jobs-and-variables.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -36,9 +36,9 @@ variables:
# Arguments for top level allocation
# Note: can be set to "OFF" to prevent allocation sharing.
# OPTIONAL: "-o per-resource.count=2" allows to get 2 jobs running on each node.
TIOGA_SHARED_ALLOC: "OFF"
TIOGA_SHARED_ALLOC: "-B asccasc --queue=pci --exclusive --time-limit=1h --nodes=1 -o per-resource.count=4"
# Arguments for job level allocation
TIOGA_JOB_ALLOC: "--job-name=${FLUX_JOB_NAME} -B asccasc --queue=pci --time-limit=1h --nodes=1 --begin-time=+5s"
TIOGA_JOB_ALLOC: "--job-name=${FLUX_JOB_NAME} -B asccasc --time-limit=1h --nodes=1 --begin-time=+5s"
# Arguments for performance job allocation (dedicated allocation with no overlapping).
TIOGA_PERF_ALLOC: "-B asccasc --queue=pci --exclusive --time-limit=1h --nodes=1 --begin-time=+5s"
# Add variables that should apply to all the jobs on a machine:
Expand All @@ -48,9 +48,9 @@ variables:
# Arguments for top level allocation
# Note: can be set to "OFF" to prevent allocation sharing.
# OPTIONAL: "-o per-resource.count=2" allows to get 2 jobs running on each node.
TUOLUMNE_SHARED_ALLOC: "OFF"
TUOLUMNE_SHARED_ALLOC: "-B asccasc --queue=pci --exclusive --time-limit=1h --nodes=1 -o per-resource.count=4"
# Arguments for job level allocation
TUOLUMNE_JOB_ALLOC: "--job-name=${FLUX_JOB_NAME} -B asccasc --queue=pci --time-limit=1h --nodes=1 --begin-time=+5s"
TUOLUMNE_JOB_ALLOC: "--job-name=${FLUX_JOB_NAME} -B asccasc --time-limit=1h --nodes=1 --begin-time=+5s"
# Arguments for performance job allocation (dedicated allocation with no overlapping).
TUOLUMNE_PERF_ALLOC: "-B asccasc --queue=pci --exclusive --time-limit=1h --nodes=1 --begin-time=+5s"
# Add variables that should apply to all the jobs on a machine:
Expand All @@ -73,6 +73,15 @@ variables:
after_script:
- |
if [ "${CI_JOB_STATUS}" = "canceled" ]; then
# release_resources does not run in a canceled pipeline, so free the
# shared allocation here.
ALLOC_ID="$(flux jobs --name="${ALLOC_NAME}" -n -o "{id}")"

if [ -n "${ALLOC_ID}" ]; then
echo "### Cancel Flux shared allocation ${ALLOC_NAME} (${ALLOC_ID})"
flux cancel "${ALLOC_ID}" || true
fi

URI="$(flux jobs --name="${FLUX_JOB_NAME}" -n -o "{id}")"

if [ -n "${URI}" ]; then
Expand All @@ -83,6 +92,14 @@ variables:
fi
fi

# Extends allocate_resources from the shared pipeline. It must be interruptible
# so that waiting for the allocation does not block GitLab from auto-canceling
# redundant pipelines.
allocate_resources:
interruptible: true
after_script:
- !reference [.clean-flux-allocation, after_script]

.clean-slurm-allocation:
after_script:
- |
Expand Down
Loading