feat: s6-overlay service supervision + wdocker CLI#136
Merged
evangineer merged 10 commits intomainfrom Mar 13, 2026
Merged
Conversation
…sender - Merge Dockerfile.hc-0.6.1-rc.1 and Dockerfile.unyt into docker/Dockerfile - Bump Holochain to 0.6.1-rc.3 - Include log-sender v0.1.4 in base image (no longer a separate unyt variant) - Simplify docker-compose.yml to single edgenode + log-collector service - Delete legacy Dockerfiles (hc-0.5.6, hc-0.6.0-go-pion, hc-0.6.0-custom-go-pion, unyt) - Delete legacy compose files and MULTI_IMAGE_STRATEGY.md - Simplify run_tests_multi.sh and build-images.sh for single image - Simplify release.yml: one build job, tags as both version and latest - Update all BATS tests: remove is_unyt()/is_hc_0_6_0() version gates Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
v0.1.5 ships separate log-sender-x86_64 and log-sender-aarch64 assets, enabling proper arm64 support. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Update kando fixtures to v0.17.1 (compatible with HC 0.6.1-rc.3 bundle manifest v0) - New SHA256: a4cdee64fe32720077e0aade94630f24d0da5e91da33ccbe5bfd894d9d359f28 - Remove initZomeCalls (profiles zome API changed) - Skip webhapp tests pending a HC 0.6.1-rc.3 compatible rhymez release (rhymez 0.1.5 uses bundle manifest v1, HC 0.6.1-rc.3 expects v0) - Fix integration_data_pipeline.bats: use real conductor config instead of empty dummy file (log-sender panicked reading data_root_path from empty YAML) - Fix log_sender_debug.bats: allow exit 124 (timeout) as success for 'timeout 5 log-sender service' step - Add mkdir -p /usr/local/bin to Dockerfile log-sender layer (build fix) Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
…al behaviour Log-sender v0.1.5 only: 1. Sends dbSize proofs (one per DHT database per interval) 2. Forwards JSONL entries with k="fetchedOps" from --report-path files Previous tests wrote k="metric" JSONL (ignored by log-sender) and queried `WHERE source LIKE 'integration_test_%'` — but the metrics table has no source column. Both issues meant wait_for_database_data always returned 0. Fixes: - Use k="fetchedOps" in all test JSONL (the type log-sender actually forwards) - Replace source-filtered DB queries with METRICS_BASELINE delta approach: @setup captures total metric count; tests assert count increased by >= N - Redesign "data integrity" test to verify proof timestamp validity and drone registration rather than non-existent metric_value/source fields - Redesign "cleanup" test to verify fresh re-registration still works (source-based DELETE was a no-op with no source column) Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Add log_tool to Container Commands section (log-sender now in every image) - Update kando sandbox example from v0.13.0 to v0.17.1 - Fix Conductor Configuration: replace incorrect 'lair_root must be empty' with actual keystore path (/var/local/lib/holochain/ks) Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
kando v0.17.1 ships a .webhapp and is compatible with HC 0.6.1-rc.3. - Add kando-webhapp.json (inner .happ SHA256: a4cdee64...) - Add kando-webhapp-badsha.json (wrong SHA256 to test checksum failure) - Add kando-webhapp-webhappsha.json (whole .webhapp SHA256 to test that providing the bundle hash rather than the inner .happ hash fails) - Remove skips from webhapp.bats and update to use new fixtures Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Replace tini+entrypoint.sh with s6-overlay v3.2.0.2 as PID 1 init
- Services: setup (oneshot), conductor, log-sender, logrotate-cron (longruns)
- conductor and log-sender auto-restart on crash via s6 supervision
- log-sender waits for config.json before starting (no polling loop)
- Removes monitor_log_sender() bash loop from entrypoint
- Add nodejs + npm + @theweave/wdocker@0.15.3 CLI tool
- Workaround for broken file: workspace deps in published package
via jq patch of package.json before npm install
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
tini+entrypoint.shwith s6-overlay v3.2.0.2 as PID 1, providing proper service supervision with automatic restart on crash@theweave/wdockerv0.15.3 CLI tool (wdocker+wdaemonbinaries) for managing always-on Moss group nodes, as documented in the EdgeNode Moss GuideServices (s6-rc.d)
setupconductorlog-senderconfig.json, then runs log-sender — auto-restarts on crashlogrotate-cronNotable details
conductorandlog-sendernow automatically restart if they crash (previously relied on Docker restart policy / bash polling loop)log-senderwaits for/etc/log-sender/config.jsonbefore starting — no more bash polling loop inmonitor_log_sender()wdockerpackage has a known upstream packaging bug wherefile:workspace paths are used for@theweave/utils,@theweave/group-client, and@theweave/api— worked around with ajqpatch ofpackage.jsonbeforenpm installentrypoint.shis no longer used as the container entrypoint; init logic lives ins6-overlay/s6-rc.d/setup/init.shTest plan
multi_installneeds initZomeCalls-compatible hApp)wdocker --versionreturns0.15.3wdaemonbinary presentwdocker run <name>creates a conductor and joins a Moss group🤖 Generated with Claude Code