[maximusunc]

Summary

Two monitor dashboard improvements.

1. Image version on worker cards

• The release workflow passes the release tag into every image as a SHEPHERD_VERSION build arg. All worker Dockerfiles and shepherd_server/Dockerfile declare it, and local builds default to dev. A deployment can also override it through the environment.
• The worker heartbeat now includes version. The monitor carries it per replica and lists the distinct versions for each worker type.
• Card: shows the version under the worker name. When replicas run different versions (e.g. mid-rollout), the line turns amber and reads Mixed: v2.7.0 ×1, v2.8.1 ×2.
• Replica detail popup: gains a Version column.
• Images built before this change report no version and show as unknown until the next release.

2. Backlog no longer counts running tasks

Workers XDEL a stream entry only after finishing it, so XLEN includes tasks already being worked on. The card's Backlog was raw XLEN, so a single running query showed as Running 1 / Backlog 1. Backlog is now XLEN − pending, clamped at 0.

The utilization bar and the recorded backlog:* history series still use XLEN (total load). This keeps the History page's "Queue backlog (XLEN)" chart and its utilization heatmap unchanged. The popup's "Queue length" label is renamed "Backlog" to match the card.

Testing

• New unit tests cover the heartbeat version field, the version rollup, the backlog calculation (including the clamp) and utilization. The monitor tests and tests/unit/test_worker_lifecycle.py pass (104 tests).
• The dashboard was rendered with mock snapshot data in headless Chromium to check how the cards and popup look.

---

Generated by Claude Code

[codecov]

Codecov Report

❌ Patch coverage is 88.88889% with 1 line in your changes missing coverage. Please review.
✅ Project coverage is 54.14%. Comparing base (0e19968) to head (d1c6221).
⚠️ Report is 1 commits behind head on devel.

Files with missing lines	Patch %	Lines
workers/monitor/poller.py	80.00%	1 Missing ⚠️

Files with missing lines	Coverage Δ
shepherd_utils/heartbeat.py	55.23% <100.00%> (+0.86%)	⬆️
workers/monitor/poller.py	53.67% <80.00%> (+2.99%)	⬆️

... and 2 files with indirect coverage changes

---

Continue to review full report in Codecov by Harness.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 5786ce5...d1c6221. Read the comment docs.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.