zacheryasc/swactor
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions
swactor/apps/myelin/node-image/Dockerfile.base

35 lines
1.2 KiB
Text
Raw Normal View History

feat(myelin): replace chat app with fleet daemon Replace the single-purpose chat runtime with a persistent fleet daemon that provisions, adopts, and controls nodes through the dashboard. Add distributed job-runner actors and provider-backed deployment so jobs can materialize workspaces, execute remotely, and return outputs over iroh.
2026-08-17 22:51:08 +00:00
# Myelin node base: CUDA runtime, SSH bootstrap, and the Myelin agent only.
feat(contextual-process): replace job runner with execution contexts Add actor-owned contextual process provisioning, descriptor bootstrap, scoped namespace operations, Python bindings, Myelin orchestration, and adversarial end-to-end coverage. Remove the superseded job runner, pipeline APIs, images, entrypoints, and specifications.
2026-08-27 22:09:26 +00:00
# Frameworks belong to workload-specific images, not the control-plane agent image.
feat(mvp-chat): local e2e chat on cuda gpu Stand up an interactive end-to-end chat over a CUDA GPU, provisioning a Dockerized node that loads a GGUF model and serves prompts over TCP. - prompt_rpc: add the newline-JSON prompt protocol (`SubmitPrompt` + `PromptEvent::{TextDelta,Done,Fault}`) carried over TCP - mvp_chat: add an interactive REPL client connecting to the prompt RPC port (default 127.0.0.1:19777) - mvp_orch_one_node / mvp_one_node_chat: add the single-node orchestrator that provisions a `LocalDockerPlugin` node, loads `bartowski/Llama-3.2-1B-Instruct-GGUF` (Q4_K_M), and exposes the prompt RPC listener with boot/route/weight timeouts - mvp_node: add the GPU worker binary that spawns `tinygrad_worker.py` (default device CUDA) and ships runtime telemetry via a `ClusterFrameSink` - vastai_provisioning / bootstrap_datastream: add the vast.ai provider adapter (`VastAiProvisioningConfig`, `VastAiLeaseClient`) wrapping `swactor_vastai`, plus a bridge that folds provision stdout onto a per-node datastream - apps/mvp-node: add CUDA base/runtime Dockerfiles (nvidia/cuda 12.6.3, tinygrad 0.12.0, sshd), `mvp_entrypoint.sh` (sshd + mvp-node, held for postmortem), `local_docker_e2e.sh`, the GGUF tinygrad worker, and one-node-chat/bootstrap/vastai guarantee tests Signed-off-by: Zachery Aaron Shores-Chmielewski <zacheryasc@gmail.com>
2026-07-01 08:44:25 +00:00
# Build from workspace root:
refactor: `mvp-system` is now a standalone app, `myelin` Promote the `mvp-system` workspace library crate to a standalone application at `apps/myelin`, rebranding the MVP system along with its binaries, node image, and spec. - workspace `Cargo.toml`: swap member `crates/mvp-system` -> `apps/myelin` and drop `apps` from `exclude` so the app joins the workspace - `apps/myelin/Cargo.toml`: declare package `myelin` with `autobins = false` and explicit `[[bin]]` targets `myelin-worker`, `myelin-orchestrator`, `myelin-chat` - `apps/myelin/src`: move the whole `mvp-system` source tree and rebrand module surfaces (`chat/mod.rs`, `prompt/mod.rs`); add `bin/chat.rs` (`myelin::run_chat_from_args`) and delete the old `mvp_chat.rs` - `apps/myelin/node-image`: relocate the worker image assets from `apps/mvp-node/` (Dockerfile, Dockerfile.base, tinygrad_worker.py, entrypoint, e2e script) and rename `MVP_SYSTEM_SPEC.md` -> `MYELIN_SPEC.md` - `xtask`: rewrite build/reference paths for the rename (~1000-line churn); add `crates/dashboard/ACTOR_PANEL_SPEC.md` Signed-off-by: Zachery Aaron Shores-Chmielewski <zacheryasc@gmail.com>
2026-08-01 09:46:55 +00:00
# docker build -f apps/myelin/node-image/Dockerfile.base -t myelin-node-base:cuda12.6 .
feat(mvp-chat): local e2e chat on cuda gpu Stand up an interactive end-to-end chat over a CUDA GPU, provisioning a Dockerized node that loads a GGUF model and serves prompts over TCP. - prompt_rpc: add the newline-JSON prompt protocol (`SubmitPrompt` + `PromptEvent::{TextDelta,Done,Fault}`) carried over TCP - mvp_chat: add an interactive REPL client connecting to the prompt RPC port (default 127.0.0.1:19777) - mvp_orch_one_node / mvp_one_node_chat: add the single-node orchestrator that provisions a `LocalDockerPlugin` node, loads `bartowski/Llama-3.2-1B-Instruct-GGUF` (Q4_K_M), and exposes the prompt RPC listener with boot/route/weight timeouts - mvp_node: add the GPU worker binary that spawns `tinygrad_worker.py` (default device CUDA) and ships runtime telemetry via a `ClusterFrameSink` - vastai_provisioning / bootstrap_datastream: add the vast.ai provider adapter (`VastAiProvisioningConfig`, `VastAiLeaseClient`) wrapping `swactor_vastai`, plus a bridge that folds provision stdout onto a per-node datastream - apps/mvp-node: add CUDA base/runtime Dockerfiles (nvidia/cuda 12.6.3, tinygrad 0.12.0, sshd), `mvp_entrypoint.sh` (sshd + mvp-node, held for postmortem), `local_docker_e2e.sh`, the GGUF tinygrad worker, and one-node-chat/bootstrap/vastai guarantee tests Signed-off-by: Zachery Aaron Shores-Chmielewski <zacheryasc@gmail.com>
2026-07-01 08:44:25 +00:00
feat(myelin): replace chat app with fleet daemon Replace the single-purpose chat runtime with a persistent fleet daemon that provisions, adopts, and controls nodes through the dashboard. Add distributed job-runner actors and provider-backed deployment so jobs can materialize workspaces, execute remotely, and return outputs over iroh.
2026-08-17 22:51:08 +00:00
FROM nvidia/cuda:12.6.3-runtime-ubuntu24.04
feat(mvp-chat): local e2e chat on cuda gpu Stand up an interactive end-to-end chat over a CUDA GPU, provisioning a Dockerized node that loads a GGUF model and serves prompts over TCP. - prompt_rpc: add the newline-JSON prompt protocol (`SubmitPrompt` + `PromptEvent::{TextDelta,Done,Fault}`) carried over TCP - mvp_chat: add an interactive REPL client connecting to the prompt RPC port (default 127.0.0.1:19777) - mvp_orch_one_node / mvp_one_node_chat: add the single-node orchestrator that provisions a `LocalDockerPlugin` node, loads `bartowski/Llama-3.2-1B-Instruct-GGUF` (Q4_K_M), and exposes the prompt RPC listener with boot/route/weight timeouts - mvp_node: add the GPU worker binary that spawns `tinygrad_worker.py` (default device CUDA) and ships runtime telemetry via a `ClusterFrameSink` - vastai_provisioning / bootstrap_datastream: add the vast.ai provider adapter (`VastAiProvisioningConfig`, `VastAiLeaseClient`) wrapping `swactor_vastai`, plus a bridge that folds provision stdout onto a per-node datastream - apps/mvp-node: add CUDA base/runtime Dockerfiles (nvidia/cuda 12.6.3, tinygrad 0.12.0, sshd), `mvp_entrypoint.sh` (sshd + mvp-node, held for postmortem), `local_docker_e2e.sh`, the GGUF tinygrad worker, and one-node-chat/bootstrap/vastai guarantee tests Signed-off-by: Zachery Aaron Shores-Chmielewski <zacheryasc@gmail.com>
2026-07-01 08:44:25 +00:00
RUN apt-get update && \
apt-get install -y --no-install-recommends \
ca-certificates \
procps \
runtime: checkpoint distributed execution and retained-node deployment Integrate namespace and source-route lifecycle changes, contextual process cleanup, Python binding updates, and Myelin worker/orchestrator recovery. Keep the shared control contracts, deployment identity fencing, SSH bootstrap adapters, paid admission accounting, and VastAI cleanup implementation together with their consumers. Migrate Iroh dependencies and telemetry transport/collection with dashboard and demo callsites, workspace build configuration, and actor-control-flow policy updates. This is an intermediate development checkpoint, not paid-provider qualification. Review verification: contextual_process_guarantees (4 tests), telemetry_transport (4 tests), and shared control contracts (5 tests) passed. Historical five-node redeployment and campaign execution passed individually; complete ordered qualification remains pending.
2026-09-14 09:20:56 +00:00
python3 \
python3-pip \
feat(mvp-chat): local e2e chat on cuda gpu Stand up an interactive end-to-end chat over a CUDA GPU, provisioning a Dockerized node that loads a GGUF model and serves prompts over TCP. - prompt_rpc: add the newline-JSON prompt protocol (`SubmitPrompt` + `PromptEvent::{TextDelta,Done,Fault}`) carried over TCP - mvp_chat: add an interactive REPL client connecting to the prompt RPC port (default 127.0.0.1:19777) - mvp_orch_one_node / mvp_one_node_chat: add the single-node orchestrator that provisions a `LocalDockerPlugin` node, loads `bartowski/Llama-3.2-1B-Instruct-GGUF` (Q4_K_M), and exposes the prompt RPC listener with boot/route/weight timeouts - mvp_node: add the GPU worker binary that spawns `tinygrad_worker.py` (default device CUDA) and ships runtime telemetry via a `ClusterFrameSink` - vastai_provisioning / bootstrap_datastream: add the vast.ai provider adapter (`VastAiProvisioningConfig`, `VastAiLeaseClient`) wrapping `swactor_vastai`, plus a bridge that folds provision stdout onto a per-node datastream - apps/mvp-node: add CUDA base/runtime Dockerfiles (nvidia/cuda 12.6.3, tinygrad 0.12.0, sshd), `mvp_entrypoint.sh` (sshd + mvp-node, held for postmortem), `local_docker_e2e.sh`, the GGUF tinygrad worker, and one-node-chat/bootstrap/vastai guarantee tests Signed-off-by: Zachery Aaron Shores-Chmielewski <zacheryasc@gmail.com>
2026-07-01 08:44:25 +00:00
openssh-server && \
apt-get clean && \
rm -rf /var/lib/apt/lists/* \
/var/cache/apt/archives/* \
/var/cache/apt/*.bin \
/var/log/apt/* \
/var/log/dpkg.log \
/tmp/* \
/var/tmp/* \
/usr/share/doc/* \
/usr/share/man/* \
/usr/share/info/* && \
feat(contextual-process): replace job runner with execution contexts Add actor-owned contextual process provisioning, descriptor bootstrap, scoped namespace operations, Python bindings, Myelin orchestration, and adversarial end-to-end coverage. Remove the superseded job runner, pipeline APIs, images, entrypoints, and specifications.
2026-08-27 22:09:26 +00:00
mkdir -p /usr/local/share/myelin /var/cache/myelin-contexts /var/log
feat(mvp-chat): local e2e chat on cuda gpu Stand up an interactive end-to-end chat over a CUDA GPU, provisioning a Dockerized node that loads a GGUF model and serves prompts over TCP. - prompt_rpc: add the newline-JSON prompt protocol (`SubmitPrompt` + `PromptEvent::{TextDelta,Done,Fault}`) carried over TCP - mvp_chat: add an interactive REPL client connecting to the prompt RPC port (default 127.0.0.1:19777) - mvp_orch_one_node / mvp_one_node_chat: add the single-node orchestrator that provisions a `LocalDockerPlugin` node, loads `bartowski/Llama-3.2-1B-Instruct-GGUF` (Q4_K_M), and exposes the prompt RPC listener with boot/route/weight timeouts - mvp_node: add the GPU worker binary that spawns `tinygrad_worker.py` (default device CUDA) and ships runtime telemetry via a `ClusterFrameSink` - vastai_provisioning / bootstrap_datastream: add the vast.ai provider adapter (`VastAiProvisioningConfig`, `VastAiLeaseClient`) wrapping `swactor_vastai`, plus a bridge that folds provision stdout onto a per-node datastream - apps/mvp-node: add CUDA base/runtime Dockerfiles (nvidia/cuda 12.6.3, tinygrad 0.12.0, sshd), `mvp_entrypoint.sh` (sshd + mvp-node, held for postmortem), `local_docker_e2e.sh`, the GGUF tinygrad worker, and one-node-chat/bootstrap/vastai guarantee tests Signed-off-by: Zachery Aaron Shores-Chmielewski <zacheryasc@gmail.com>
2026-07-01 08:44:25 +00:00
refactor: `mvp-system` is now a standalone app, `myelin` Promote the `mvp-system` workspace library crate to a standalone application at `apps/myelin`, rebranding the MVP system along with its binaries, node image, and spec. - workspace `Cargo.toml`: swap member `crates/mvp-system` -> `apps/myelin` and drop `apps` from `exclude` so the app joins the workspace - `apps/myelin/Cargo.toml`: declare package `myelin` with `autobins = false` and explicit `[[bin]]` targets `myelin-worker`, `myelin-orchestrator`, `myelin-chat` - `apps/myelin/src`: move the whole `mvp-system` source tree and rebrand module surfaces (`chat/mod.rs`, `prompt/mod.rs`); add `bin/chat.rs` (`myelin::run_chat_from_args`) and delete the old `mvp_chat.rs` - `apps/myelin/node-image`: relocate the worker image assets from `apps/mvp-node/` (Dockerfile, Dockerfile.base, tinygrad_worker.py, entrypoint, e2e script) and rename `MVP_SYSTEM_SPEC.md` -> `MYELIN_SPEC.md` - `xtask`: rewrite build/reference paths for the rename (~1000-line churn); add `crates/dashboard/ACTOR_PANEL_SPEC.md` Signed-off-by: Zachery Aaron Shores-Chmielewski <zacheryasc@gmail.com>
2026-08-01 09:46:55 +00:00
COPY apps/myelin/node-image/myelin_entrypoint.sh /usr/local/bin/myelin_entrypoint.sh
RUN chmod +x /usr/local/bin/myelin_entrypoint.sh
feat(mvp-chat): local e2e chat on cuda gpu Stand up an interactive end-to-end chat over a CUDA GPU, provisioning a Dockerized node that loads a GGUF model and serves prompts over TCP. - prompt_rpc: add the newline-JSON prompt protocol (`SubmitPrompt` + `PromptEvent::{TextDelta,Done,Fault}`) carried over TCP - mvp_chat: add an interactive REPL client connecting to the prompt RPC port (default 127.0.0.1:19777) - mvp_orch_one_node / mvp_one_node_chat: add the single-node orchestrator that provisions a `LocalDockerPlugin` node, loads `bartowski/Llama-3.2-1B-Instruct-GGUF` (Q4_K_M), and exposes the prompt RPC listener with boot/route/weight timeouts - mvp_node: add the GPU worker binary that spawns `tinygrad_worker.py` (default device CUDA) and ships runtime telemetry via a `ClusterFrameSink` - vastai_provisioning / bootstrap_datastream: add the vast.ai provider adapter (`VastAiProvisioningConfig`, `VastAiLeaseClient`) wrapping `swactor_vastai`, plus a bridge that folds provision stdout onto a per-node datastream - apps/mvp-node: add CUDA base/runtime Dockerfiles (nvidia/cuda 12.6.3, tinygrad 0.12.0, sshd), `mvp_entrypoint.sh` (sshd + mvp-node, held for postmortem), `local_docker_e2e.sh`, the GGUF tinygrad worker, and one-node-chat/bootstrap/vastai guarantee tests Signed-off-by: Zachery Aaron Shores-Chmielewski <zacheryasc@gmail.com>
2026-07-01 08:44:25 +00:00
ARG DEPLOY_PUBKEY=""
refactor: `mvp-system` is now a standalone app, `myelin` Promote the `mvp-system` workspace library crate to a standalone application at `apps/myelin`, rebranding the MVP system along with its binaries, node image, and spec. - workspace `Cargo.toml`: swap member `crates/mvp-system` -> `apps/myelin` and drop `apps` from `exclude` so the app joins the workspace - `apps/myelin/Cargo.toml`: declare package `myelin` with `autobins = false` and explicit `[[bin]]` targets `myelin-worker`, `myelin-orchestrator`, `myelin-chat` - `apps/myelin/src`: move the whole `mvp-system` source tree and rebrand module surfaces (`chat/mod.rs`, `prompt/mod.rs`); add `bin/chat.rs` (`myelin::run_chat_from_args`) and delete the old `mvp_chat.rs` - `apps/myelin/node-image`: relocate the worker image assets from `apps/mvp-node/` (Dockerfile, Dockerfile.base, tinygrad_worker.py, entrypoint, e2e script) and rename `MVP_SYSTEM_SPEC.md` -> `MYELIN_SPEC.md` - `xtask`: rewrite build/reference paths for the rename (~1000-line churn); add `crates/dashboard/ACTOR_PANEL_SPEC.md` Signed-off-by: Zachery Aaron Shores-Chmielewski <zacheryasc@gmail.com>
2026-08-01 09:46:55 +00:00
RUN if [ -n "$DEPLOY_PUBKEY" ]; then printf '%s\n' "$DEPLOY_PUBKEY" > /etc/myelin_deploy_key.pub; fi
feat(mvp-chat): local e2e chat on cuda gpu Stand up an interactive end-to-end chat over a CUDA GPU, provisioning a Dockerized node that loads a GGUF model and serves prompts over TCP. - prompt_rpc: add the newline-JSON prompt protocol (`SubmitPrompt` + `PromptEvent::{TextDelta,Done,Fault}`) carried over TCP - mvp_chat: add an interactive REPL client connecting to the prompt RPC port (default 127.0.0.1:19777) - mvp_orch_one_node / mvp_one_node_chat: add the single-node orchestrator that provisions a `LocalDockerPlugin` node, loads `bartowski/Llama-3.2-1B-Instruct-GGUF` (Q4_K_M), and exposes the prompt RPC listener with boot/route/weight timeouts - mvp_node: add the GPU worker binary that spawns `tinygrad_worker.py` (default device CUDA) and ships runtime telemetry via a `ClusterFrameSink` - vastai_provisioning / bootstrap_datastream: add the vast.ai provider adapter (`VastAiProvisioningConfig`, `VastAiLeaseClient`) wrapping `swactor_vastai`, plus a bridge that folds provision stdout onto a per-node datastream - apps/mvp-node: add CUDA base/runtime Dockerfiles (nvidia/cuda 12.6.3, tinygrad 0.12.0, sshd), `mvp_entrypoint.sh` (sshd + mvp-node, held for postmortem), `local_docker_e2e.sh`, the GGUF tinygrad worker, and one-node-chat/bootstrap/vastai guarantee tests Signed-off-by: Zachery Aaron Shores-Chmielewski <zacheryasc@gmail.com>
2026-07-01 08:44:25 +00:00
refactor: `mvp-system` is now a standalone app, `myelin` Promote the `mvp-system` workspace library crate to a standalone application at `apps/myelin`, rebranding the MVP system along with its binaries, node image, and spec. - workspace `Cargo.toml`: swap member `crates/mvp-system` -> `apps/myelin` and drop `apps` from `exclude` so the app joins the workspace - `apps/myelin/Cargo.toml`: declare package `myelin` with `autobins = false` and explicit `[[bin]]` targets `myelin-worker`, `myelin-orchestrator`, `myelin-chat` - `apps/myelin/src`: move the whole `mvp-system` source tree and rebrand module surfaces (`chat/mod.rs`, `prompt/mod.rs`); add `bin/chat.rs` (`myelin::run_chat_from_args`) and delete the old `mvp_chat.rs` - `apps/myelin/node-image`: relocate the worker image assets from `apps/mvp-node/` (Dockerfile, Dockerfile.base, tinygrad_worker.py, entrypoint, e2e script) and rename `MVP_SYSTEM_SPEC.md` -> `MYELIN_SPEC.md` - `xtask`: rewrite build/reference paths for the rename (~1000-line churn); add `crates/dashboard/ACTOR_PANEL_SPEC.md` Signed-off-by: Zachery Aaron Shores-Chmielewski <zacheryasc@gmail.com>
2026-08-01 09:46:55 +00:00
ENTRYPOINT ["/usr/local/bin/myelin_entrypoint.sh"]
Reference in a new issue Copy permalink