feat(docker): schema for lifecycle events + swarm topology
CI / lint (push) Successful in 3s
CI / unit (push) Successful in 47s
CI / integration (push) Successful in 2m20s
CI / publish (push) Successful in 58s

Adds the milestone-77 storage that doesn't fit on the per-container row:

  * docker_events — lifecycle (start/stop/die/oom/health_change), to be
    derived by diffing consecutive host snapshots; host-scoped, indexed for
    timeline lookups (host_id, container_name, at) and retention pruning (at).
  * docker_swarm_services / docker_swarm_nodes — manager-reported Swarm
    topology (desired-vs-running replicas, node role/availability/status).

Migration docker_004 extends the docker branch (down_revision docker_003);
purely additive, no DROP+recreate. event/mode/role are plain strings (no
CHECK whitelist), matching how docker_containers models status. Integration
guard asserts the three new host-scoped tables exist.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_016Jg27rgypiW2efULXJDtMC
This commit is contained in:
2026-06-18 20:47:46 -04:00
parent 82c3d2cf36
commit fee654b53a
3 changed files with 185 additions and 0 deletions
@@ -0,0 +1,77 @@
"""Docker lifecycle events + swarm topology tables
Adds the storage for milestone-77 monitoring depth that doesn't fit on the
per-container row:
* docker_events — lifecycle (start/stop/die/oom/health_change) derived by
diffing consecutive host snapshots; host-scoped, indexed for both timeline
lookups (host_id, container_name, at) and retention pruning (at).
* docker_swarm_services / docker_swarm_nodes — manager-reported Swarm
topology (desired-vs-running replicas, node role/availability/status).
All new tables — purely additive, so no DROP+recreate. `event`/`mode`/`role`
etc. are plain strings (no CHECK whitelist), matching how docker_containers
models `status`; keeps the value sets open without a migration per addition
(so rule 36 doesn't apply here).
Revision ID: docker_004_events_swarm
Revises: docker_003_container_enrichment
Create Date: 2026-06-19
"""
from typing import Sequence, Union
from alembic import op
import sqlalchemy as sa
revision: str = "docker_004_events_swarm"
down_revision: Union[str, None] = "docker_003_container_enrichment"
branch_labels: Union[str, Sequence[str], None] = None
depends_on: Union[str, Sequence[str], None] = None
def upgrade() -> None:
op.create_table(
"docker_events",
sa.Column("id", sa.String(36), primary_key=True),
sa.Column("host_id", sa.String(36),
sa.ForeignKey("hosts.id", ondelete="CASCADE"), nullable=False),
sa.Column("container_name", sa.String(255), nullable=False),
sa.Column("event", sa.String(16), nullable=False),
sa.Column("detail", sa.String(255), nullable=True),
sa.Column("at", sa.DateTime(timezone=True), nullable=False),
)
op.create_index("ix_docker_events_host_container_time", "docker_events",
["host_id", "container_name", "at"])
op.create_index("ix_docker_events_at", "docker_events", ["at"])
op.create_table(
"docker_swarm_services",
sa.Column("host_id", sa.String(36),
sa.ForeignKey("hosts.id", ondelete="CASCADE"), primary_key=True),
sa.Column("service_name", sa.String(255), primary_key=True),
sa.Column("mode", sa.String(16), nullable=False, server_default="replicated"),
sa.Column("desired", sa.Integer, nullable=False, server_default="0"),
sa.Column("running", sa.Integer, nullable=False, server_default="0"),
sa.Column("image", sa.String(512), nullable=True),
sa.Column("updated_at", sa.DateTime(timezone=True), nullable=False),
)
op.create_table(
"docker_swarm_nodes",
sa.Column("host_id", sa.String(36),
sa.ForeignKey("hosts.id", ondelete="CASCADE"), primary_key=True),
sa.Column("node_id", sa.String(64), primary_key=True),
sa.Column("hostname", sa.String(255), nullable=False, server_default=""),
sa.Column("role", sa.String(16), nullable=False, server_default="worker"),
sa.Column("availability", sa.String(16), nullable=False, server_default="active"),
sa.Column("status", sa.String(16), nullable=False, server_default="unknown"),
sa.Column("leader", sa.Boolean, nullable=False, server_default=sa.false()),
sa.Column("updated_at", sa.DateTime(timezone=True), nullable=False),
)
def downgrade() -> None:
op.drop_table("docker_swarm_nodes")
op.drop_table("docker_swarm_services")
op.drop_index("ix_docker_events_at", table_name="docker_events")
op.drop_index("ix_docker_events_host_container_time", table_name="docker_events")
op.drop_table("docker_events")
+86
View File
@@ -85,3 +85,89 @@ class DockerMetric(Base):
Index("ix_docker_metrics_host_container_time",
"host_id", "container_name", "scraped_at"),
)
class DockerEvent(Base):
"""Lifecycle events derived by diffing consecutive host snapshots.
The agent only reports current state, so Steward synthesises lifecycle by
comparing each new snapshot against the stored DockerContainer state for the
host: a container that appears → `start`; one that vanishes → `stop`; a
transition to a stopped status with a non-zero exit → `die`; an OOM-kill
flip → `oom`; a health status change → `health_change`. Scoped to the
reporting host (names are only unique within a host). `event` is a plain
string (no CHECK whitelist), matching how `status` is modelled on
DockerContainer — keeps the value set open without a migration per addition.
"""
__tablename__ = "docker_events"
id: Mapped[str] = mapped_column(
String(36), primary_key=True, default=lambda: str(uuid.uuid4())
)
host_id: Mapped[str] = mapped_column(
String(36), ForeignKey("hosts.id", ondelete="CASCADE"), nullable=False
)
container_name: Mapped[str] = mapped_column(String(255), nullable=False)
event: Mapped[str] = mapped_column(String(16), nullable=False)
# start | stop | die | oom | health_change
detail: Mapped[str | None] = mapped_column(String(255), nullable=True)
# e.g. "exit 137", "healthy→unhealthy", image on start
at: Mapped[datetime] = mapped_column(
DateTime(timezone=True), nullable=False,
default=lambda: datetime.now(timezone.utc),
)
# Timeline lookups filter on (host_id, container_name) and sort by time;
# retention prunes by `at` alone — index both access paths.
__table_args__ = (
Index("ix_docker_events_host_container_time",
"host_id", "container_name", "at"),
Index("ix_docker_events_at", "at"),
)
class DockerSwarmService(Base):
"""A Swarm service as seen by a manager host: desired vs running replicas.
Manager-scoped — only a host that is a Swarm manager reports these, so a
single-manager cluster yields one row set keyed by that host. Service names
are unique within a swarm; we still key by (host_id, service_name) so two
managers reporting independently don't collide.
"""
__tablename__ = "docker_swarm_services"
host_id: Mapped[str] = mapped_column(
String(36), ForeignKey("hosts.id", ondelete="CASCADE"), primary_key=True
)
service_name: Mapped[str] = mapped_column(String(255), primary_key=True)
mode: Mapped[str] = mapped_column(String(16), nullable=False, default="replicated")
# replicated | global
desired: Mapped[int] = mapped_column(Integer, nullable=False, default=0)
running: Mapped[int] = mapped_column(Integer, nullable=False, default=0)
image: Mapped[str | None] = mapped_column(String(512), nullable=True)
updated_at: Mapped[datetime] = mapped_column(
DateTime(timezone=True), nullable=False,
default=lambda: datetime.now(timezone.utc),
)
class DockerSwarmNode(Base):
"""A Swarm node as seen by a manager host: role / availability / status."""
__tablename__ = "docker_swarm_nodes"
host_id: Mapped[str] = mapped_column(
String(36), ForeignKey("hosts.id", ondelete="CASCADE"), primary_key=True
)
node_id: Mapped[str] = mapped_column(String(64), primary_key=True)
hostname: Mapped[str] = mapped_column(String(255), nullable=False, default="")
role: Mapped[str] = mapped_column(String(16), nullable=False, default="worker")
# manager | worker
availability: Mapped[str] = mapped_column(String(16), nullable=False, default="active")
# active | pause | drain
status: Mapped[str] = mapped_column(String(16), nullable=False, default="unknown")
# ready | down | unknown
leader: Mapped[bool] = mapped_column(Boolean, nullable=False, default=False)
updated_at: Mapped[datetime] = mapped_column(
DateTime(timezone=True), nullable=False,
default=lambda: datetime.now(timezone.utc),
)