bvandeusenandClaude Opus 5.5 423493e2bf fix(docker): a restarted agent's re-sent log lines are stored once (#5809) ...

An agent keeps its log cursors in memory, so after a restart it seeds each
container from its last 50 lines, which Steward already holds; docker_logs
has no unique key, so every update or reboot doubled them. The ingest now
reads what it already has for each container over the batch's own span,
in one query, and skips those lines. A push of new lines overlaps nothing.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

2026-10-15 10:08:38 -04:00

..

migrations

refactor(docker): container CPU and memory live in the one metric store (#5752)

2026-10-14 14:39:56 -04:00

templates/docker

feat(norms): hosts, containers and monitor response times are graded against their own normal (#5742)

2026-10-14 16:17:04 -04:00

__init__.py

feat(swarm): services short, nodes down and managers lost are alertable (#5788)

2026-10-14 22:23:52 -04:00

dedup.py

fix(docker): move dedup helper to a model-free module (CI green)

2026-06-20 23:07:33 -04:00

disk_view.py

feat(docker): Docker disk is recorded over time per host, the Disk page shows each host's growth and folds images under their repository, and the overview has one disk line per host (#5304–#5306)

2026-10-07 20:16:13 -04:00

ingest.py

fix(docker): a restarted agent's re-sent log lines are stored once (#5809)

2026-10-15 10:08:38 -04:00

mcp_tools.py

fix(docker): a swarm service back to all its replicas leaves the overview and dashboard (#5722)

2026-10-14 16:50:31 -04:00

models.py

refactor(docker): container CPU and memory live in the one metric store (#5752)

2026-10-14 14:39:56 -04:00

overview.py

feat(docker): the Docker page leads with a verdict, lists services and containers as filterable tables, and each service has its own page with logs across its tasks (#5298–#5302)

2026-10-07 20:08:38 -04:00

plugin.yaml

feat(docker): per-host collection via the host agent; drop central scrape

2026-06-18 17:54:36 -04:00

retention.py

refactor(docker): container CPU and memory live in the one metric store (#5752)

2026-10-14 14:39:56 -04:00

routes.py

fix(docker): a swarm service back to all its replicas leaves the overview and dashboard (#5722)

2026-10-14 16:50:31 -04:00

service.py

feat(norms): hosts, containers and monitor response times are graded against their own normal (#5742)

2026-10-14 16:17:04 -04:00

swarm_health.py

feat(swarm): services short, nodes down and managers lost are alertable (#5788)

2026-10-14 22:23:52 -04:00

swarm_nodes.py

feat(docker): the Swarm page is laid out by node: role, what its containers use, and the services placed on each (#5303)

2026-10-07 20:11:46 -04:00

swarm_view.py

feat(docker): the Docker page leads with a verdict, lists services and containers as filterable tables, and each service has its own page with logs across its tasks (#5298–#5302)

2026-10-07 20:08:38 -04:00