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 |
Alexander Korolev
|
6c6c70e0b9
feat: add test-sleep recipe for cluster testing
|
4 mesi fa |
Alexander Korolev
|
4d3c297c50
feat: add --rm to containers by default, --keep flag to persist
|
4 mesi fa |
Alexander Korolev
|
91e273a649
feat: health command shows all containers, marks ours (sm121)
|
4 mesi fa |
Alexander Korolev
|
a726b83bcf
feat: mount HF cache directory in Docker containers
|
4 mesi fa |
Alexander Korolev
|
38290ec6f5
refactor: split docker.rs and tmux/mod.rs under 500 lines each
|
4 mesi fa |
Alexander Korolev
|
303e0ce9be
feat: SGLang engine support + minimax-m3 recipe + README
|
4 mesi fa |
Alexander Korolev
|
0a54cfc240
refactor: split codebase into <500-line modules, fix review issues
|
4 mesi fa |
Alexander Korolev
|
750a89980c
refactor: integrate health.rs, deduplicate select_cluster, fix Ray backend, add 21 tests
|
4 mesi fa |
Alexander Korolev
|
ff2838f901
feat: tmux launch variant with sequential Ray startup, deep health checks, multi-window layout
|
4 mesi fa |
Alexander Korolev
|
abcb899ef7
feat: doctor command with repair, distributed_backend ray/mp switch
|
6 mesi fa |
Alexander Korolev
|
8db6d414a3
refactor: comprehensive rewrite — concurrent SSH, labels, config-driven clusters, host key verification
|
6 mesi fa |
Alexander Korolev
|
491d7a66f6
Initial commit: sm121-launcher — DGX Spark cluster launcher for vLLM/SGLang
|
6 mesi fa |