Skip to content

Add optional log copying to llm-training chart #90

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
edwardsp merged 1 commit into from
Nov 15, 2025
Merged

Add optional log copying to llm-training chart #90

Show file tree
Hide file tree
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
16 changes: 16 additions & 0 deletions examples/llm-foundry/aks/helm/llm-training/templates/pytorchjob.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -43,6 +43,7 @@ spec:
ulimit -s unlimited && \
ulimit -l unlimited && \
ulimit -a && \
LOG_FILE="/tmp/$(hostname).out" && \
composer \
--world_size {{ mul .Values.training.nodes .Values.training.gpusPerNode }} \
--node_rank $RANK \
Expand All @@ -55,6 +56,13 @@ spec:
{{- range $key, $value := .Values.yamlUpdates }}
{{ $key }}={{ $value }} \
{{- end }}
2>&1 | tee "$LOG_FILE"
EXIT_CODE=$?
{{- if .Values.logs.copyLogsDir }}
mkdir -p {{ .Values.logs.copyLogsDir }}
cp "$LOG_FILE" {{ .Values.logs.copyLogsDir }}/ || echo "Failed to copy logs"
{{- end }}
exit $EXIT_CODE
env:
{{- include "llm-training.nccl-env" . | nindent 12 }}
{{- include "llm-training.sharp-env" . | nindent 12 }}
Expand Down Expand Up @@ -118,6 +126,7 @@ spec:
ulimit -s unlimited && \
ulimit -l unlimited && \
ulimit -a && \
LOG_FILE="/tmp/$(hostname).out" && \
composer \
--world_size {{ mul .Values.training.nodes .Values.training.gpusPerNode }} \
--node_rank $RANK \
Expand All @@ -130,6 +139,13 @@ spec:
{{- range $key, $value := .Values.yamlUpdates }}
{{ $key }}={{ $value }} \
{{- end }}
2>&1 | tee "$LOG_FILE"
EXIT_CODE=$?
{{- if .Values.logs.copyLogsDir }}
mkdir -p {{ .Values.logs.copyLogsDir }}
cp "$LOG_FILE" {{ .Values.logs.copyLogsDir }}/ || echo "Failed to copy logs"
{{- end }}
exit $EXIT_CODE
env:
{{- include "llm-training.nccl-env" . | nindent 12 }}
{{- include "llm-training.sharp-env" . | nindent 12 }}
Expand Down
5 changes: 5 additions & 0 deletions examples/llm-foundry/aks/helm/llm-training/values.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,6 +18,11 @@ training:
kueue:
queueName: "" # e.g., "gpu-local-queue" to use Kueue

# Logs configuration (optional)
# Set copyLogsDir to save stdout/stderr logs to a directory
logs:
copyLogsDir: "" # e.g., "/mnt/logs" - logs saved as $(hostname).out

# Additional YAML updates for the training configuration
yamlUpdates: {}
# Example:
Expand Down