Alexander Korolev
|
7b01e19ec2
fix: HF_HOME -> /root/.cache/huggingface (fixes stale cache)
|
4 mesi fa |
Alexander Korolev
|
9241bdc682
fix: don't pass --distributed-executor-backend mp (auto-detected)
|
4 mesi fa |
Alexander Korolev
|
6f2c33a82f
fix: only add --shm-size when recipe explicitly sets it
|
4 mesi fa |
Alexander Korolev
|
bc41cdf4c9
fix(deepseek): restore --reasoning-config and reasoning_effort
|
4 mesi fa |
Alexander Korolev
|
d9c3f845c2
fix(deepseek): align parameters with working system
|
4 mesi fa |
Alexander Korolev
|
8df3f532b7
fix(deepseek): restore clean literal strings for JSON configs
|
4 mesi fa |
Alexander Korolev
|
b0c139ea80
feat: auto-select smallest cluster matching required_nodes
|
4 mesi fa |
Alexander Korolev
|
a9a35440a0
chore: cleanup recipes
|
4 mesi fa |
Alexander Korolev
|
f5f4a2ae73
docs(deepseek): use TOML literal strings for JSON configs
|
4 mesi fa |
Alexander Korolev
|
d521cfd724
docs(deepseek): use triple-quoted strings for JSON configs
|
4 mesi fa |
Alexander Korolev
|
2af1104e7f
feat: launcher auto-adds --host/--port; recipes no longer hardcode
|
4 mesi fa |
Alexander Korolev
|
1fa2811301
feat: default recipe port is 30000
|
4 mesi fa |
Alexander Korolev
|
7ba894be67
feat: per-recipe shm_size + DeepSeek V4 Flash recipe
|
4 mesi fa |
Alexander Korolev
|
7b9f963036
feat: optional cluster-level extra_entities for power commands
|
4 mesi fa |
Alexander Korolev
|
7b8651cead
feat: Home Assistant integration for power on/off
|
4 mesi fa |
Alexander Korolev
|
9c0e7680a4
feat: set safe NCCL defaults (IGNORE_CPU_AFFINITY, NET, CROSS_NIC, DEBUG)
|
4 mesi fa |
Alexander Korolev
|
77a77c0f32
docs: document every environment variable the launcher passes
|
4 mesi fa |
Alexander Korolev
|
1b02f53b07
feat: set GLOO_SOCKET_IFNAME and TP_SOCKET_IFNAME for vLLM
|
4 mesi fa |
Alexander Korolev
|
071837e830
feat: fallback to ~/.config/sm121-launcher/ for config + recipes
|
4 mesi fa |
Alexander Korolev
|
ce40c23b7a
feat: section aliases for exec subcommand
|
4 mesi fa |
Alexander Korolev
|
bfe94fa39e
feat: pre-serve hooks + exec subcommand for cluster mass-execution
|
4 mesi fa |
Alexander Korolev
|
5097d55c77
fix: redirect stderr in `cargo run logs --follow` docker pipeline
|
4 mesi fa |
Alexander Korolev
|
65cbf60fce
build: enforce quality gates via scripts/check.sh
|
4 mesi fa |
Alexander Korolev
|
df5e3a701e
feat: add 'init-host-keys' command to auto-write SSH host_keys
|
4 mesi fa |
Alexander Korolev
|
d92da29eb5
fix: shell-quote entrypoint script so 'sh -c SCRIPT' survives bash word-split
|
4 mesi fa |
Alexander Korolev
|
75860da26d
feat: test-sleep recipe streams logs; add --timestamps to docker logs
|
4 mesi fa |
Alexander Korolev
|
2b0be29648
feat: cluster defaults, configurable init_port, --headless for workers
|
4 mesi fa |
Alexander Korolev
|
afd2fc35bf
feat: switch SGLang to --dist-init-addr, add memlock/stack ulimits
|
4 mesi fa |
Alexander Korolev
|
ff9a8c6c8c
fix: test-sleep runs forever (tail -f /dev/null) for lifecycle testing
|
4 mesi fa |
Alexander Korolev
|
8194da83d4
fix: stop/logs use CLI recipe name, health uses correct label filter
|
4 mesi fa |