gsadmin/sencho
1
0
Fork 0
mirror of https://github.com/Studio-Saelix/sencho.git synced 2026-08-12 19:57:37 +00:00
Code Issues Packages Projects Releases Wiki Activity

feat(fleet): add remote node update management (#353)

Browse Source 
Add the ability to check for outdated nodes and trigger over-the-air
updates from Fleet View. Nodes self-update by pulling the latest Docker
image and recreating their container via the "last breath" pattern.

Backend:
- SelfUpdateService: self-container identification via HOSTNAME + Docker
  Compose labels, triggers pull + force-recreate
- CapabilityRegistry: runtime capability disabling via disableCapability()
- POST /api/system/update (202 + deferred self-update)
- GET /api/fleet/update-status (version comparison across fleet)
- POST /api/fleet/nodes/:nodeId/update (single node)
- POST /api/fleet/update-all (bulk remote update)
- In-memory update tracker with 5-min timeout

Frontend:
- Node Updates modal with summary stats, search filter, table layout,
  per-node Update buttons, and bulk Update All
- Version badges and update-available indicators on node cards
- ReconnectingOverlay for local node updates (polls /api/health)
- 5s fast-poll when any node is actively updating
- UpdateStatusBadge shared component for consistent badge rendering

Requires Skipper (Pro) tier. Nodes must be deployed via Docker Compose
with Docker socket access.
This commit is contained in:
Anso
2026-04-03 01:39:22 -04:00
committed by GitHub
parent d670984635
commit 87b5908288
13 changed files with 1039 additions and 21 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/features/remote-updates.mdx
+93
View File
@@ -0,0 +1,93 @@
---
title: Remote updates
description: Check for outdated nodes and trigger over-the-air updates from the Fleet View.
---
Sencho can update remote nodes directly from the dashboard. When the gateway is running a newer version than a remote node, a one-click update pulls the latest image and recreates the container automatically.
<Note>
Remote updates require **Sencho Pro** (Skipper or Admiral tier).
</Note>
## Prerequisites
Remote updates work when each node meets these conditions:
- Deployed via **Docker Compose** (the recommended method)
- The Docker socket (`/var/run/docker.sock`) is mounted into the container
- Running **Sencho v0.32.0** or later (the version that introduced the `self-update` capability)
Nodes deployed with `docker run` or orchestrators like Kubernetes do not support self-update. These nodes will show a "self-update unavailable" message.
## How it works
1. The gateway compares each node's version (from `/api/meta`) against its own version.
2. If a node is running an older version, the Fleet View shows an **Update available** badge.
3. Clicking **Update** sends a command to the remote node.
4. The remote node pulls the latest `saelix/sencho` image and recreates its own container.
5. The gateway polls the node until it comes back online with the new version.
The remote node briefly goes offline during the update (typically 10–30 seconds depending on image pull speed). The Fleet View tracks the progress and shows **Updating**, **Updated**, or **Timed out** status badges.
## Checking for updates
Open the **Fleet** tab and click **Check Updates** in the header. This opens the **Node Updates** dialog, which lists every node with its current version and update status.
<Frame>
<img src="/images/fleet/check-updates-modal.png" alt="Node Updates dialog showing update status for each node" />
</Frame>
Each node row shows:
| Element | Meaning |
|---------|---------|
| Version label (e.g. `v0.31.0`) | The node's current Sencho version |
| **Up to date** badge | Node is running the latest version |
| **Update** button | A newer version is available — click to update |
| **Updating** badge | The node is pulling and restarting |
| **Updated** badge | The node came back online with the new version |
| **Timed out** badge | The node did not come back within 5 minutes |
Nodes running a version too old to report their version will show `pre-X.Y.Z` as the current version.
## Updating a single node
Click the **Update** button next to any outdated node in the Node Updates dialog, or click the **Update to vX.Y.Z** button directly on the node card in Fleet View. For remote nodes, the update happens in the background. For the local (gateway) node, a confirmation dialog appears warning about the brief dashboard disconnection.
## Updating all nodes
Click **Update All (N)** in the Node Updates dialog to trigger updates on all outdated remote nodes simultaneously. The local node is intentionally excluded from bulk updates — update it separately to avoid losing dashboard connectivity.
## Local node updates
When updating the local (gateway) node:
1. A confirmation dialog explains that the dashboard will briefly disconnect.
2. After confirming, the server pulls the new image and restarts.
3. A reconnecting overlay appears and polls the server every 3 seconds.
4. The page automatically reloads when the server comes back.
If the server does not return within 5 minutes, a timeout message appears with a manual reload option.
## Troubleshooting
### Node shows "self-update unavailable"
The node is not running inside a Docker Compose-managed container, or the Docker socket is not mounted. Check that your `docker-compose.yml` includes the socket volume mount:
```yaml
volumes:
- /var/run/docker.sock:/var/run/docker.sock
```
### Update times out
The image pull may be slow on the remote host, or the container failed to restart. SSH into the remote host and check `docker logs sencho` for errors. You can also manually update by running:
```bash
docker compose pull && docker compose up -d
```
### Old node does not show update button
Nodes running Sencho versions before v0.32.0 do not advertise the `self-update` capability. Update them manually first, then future updates can be triggered from the dashboard.