node_setup: retry+timeout around model downloads (xet mid-transfer stalls)
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
This commit is contained in:
@@ -8,8 +8,14 @@ cd /dev/shm && git clone https://git.draic.info/nils/jspace.git 2>/dev/null || (
|
|||||||
cd /dev/shm/jspace
|
cd /dev/shm/jspace
|
||||||
rclone copy jspace:jspace/results-loop /dev/shm/jspace/results-loop/ --include "*.json" --include "adapter_code_e399.pt"
|
rclone copy jspace:jspace/results-loop /dev/shm/jspace/results-loop/ --include "*.json" --include "adapter_code_e399.pt"
|
||||||
export HF_HUB_ENABLE_HF_TRANSFER=1
|
export HF_HUB_ENABLE_HF_TRANSFER=1
|
||||||
hf download google/gemma-4-12B-it
|
# downloads can hang mid-transfer on some vast.ai networks (xet stall):
|
||||||
hf download google/gemma-4-E2B-it
|
# retry loop, resume is free
|
||||||
|
for m in google/gemma-4-12B-it google/gemma-4-E2B-it; do
|
||||||
|
for i in 1 2 3 4 5; do
|
||||||
|
timeout 900 hf download "$m" && break
|
||||||
|
echo "RETRY $i: $m"; sleep 5
|
||||||
|
done
|
||||||
|
done
|
||||||
# probe gate: one verified generation before anything batch
|
# probe gate: one verified generation before anything batch
|
||||||
export HF_HUB_OFFLINE=1 JLENS_MODEL=google/gemma-4-12B-it
|
export HF_HUB_OFFLINE=1 JLENS_MODEL=google/gemma-4-12B-it
|
||||||
python - <<'PY'
|
python - <<'PY'
|
||||||
|
|||||||
Reference in New Issue
Block a user