Files
FabledSteward/steward/models/monitors.py
T
bvandeusenandClaude Opus 5.5 e595bd922a
CI / lint (push) Successful in 3s
CI / unit (push) Failing after 1m23s
CI / integration (push) Successful in 1m12s
CI / publish (push) Skipped
feat(monitors): problems first, a page per monitor, adding on its own page (#5121)
The Monitors list now opens with a count strip (down, up with how many
degraded, pending, paused) and sorts down, degraded, pending, up, paused,
by name within each. Degraded is up but worth a look: uptime at or below
its warning floor, or the latest response past its critical cutoff, both
from Settings → Thresholds; a slow-but-tolerable reply is coloured, not
sorted.

A monitor had no page of its own: every link went to the list. Now
/monitors/<id> shows its state and since when, latency, uptime for 24h/7d/
30d, TLS days, the last checks, recent failures with their errors, and its
settings (HTTP header names, never their values). Status links to it, and
adding or editing returns to it.

Adding is its own page (/monitors/new, form.html shared with edit) behind
an "Add monitor" button; a refused add re-renders the form with the error
and what was typed instead of a bare 400 text. Row actions are labelled
icon buttons, type names read Ping/TCP/DNS/HTTP (|monitor_type), and the
heartbeat uses pill-up/pill-down, so HEX_CEILING drops to 56.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-06 17:54:30 -04:00

120 lines
5.0 KiB
Python

from __future__ import annotations
import enum
import json
import uuid
from datetime import datetime, timezone
from sqlalchemy import (
Boolean, CheckConstraint, DateTime, Float, ForeignKey, Index, Integer, String, Text,
)
from sqlalchemy.orm import Mapped, mapped_column
from .base import Base
class MonitorType(str, enum.Enum):
"""The kinds of synthetic/uptime check a Monitor can run."""
icmp = "icmp" # ICMP echo via the system ping binary
tcp = "tcp" # TCP connect to target:port
dns = "dns" # DNS A/AAAA resolution, optional expected-IP assertion
http = "http" # HTTP(S) request — status / content / TLS expiry
# Whitelist for the CHECK constraint. `type` is a plain String + CHECK rather
# than a native Postgres enum so a new monitor type ships as a same-change
# DROP/ADD CONSTRAINT (family rule 36) instead of the heavier ALTER TYPE ADD
# VALUE a native enum would force.
MONITOR_TYPE_VALUES = tuple(t.value for t in MonitorType)
# How a page names each type.
MONITOR_TYPE_LABELS = {"icmp": "Ping", "tcp": "TCP", "dns": "DNS", "http": "HTTP"}
_TYPE_CHECK = "type IN ('icmp', 'tcp', 'dns', 'http')"
class Monitor(Base):
"""A single configured uptime/synthetic check of any type.
Unifies what used to be host-coupled ping/DNS facets (columns on `hosts`)
and the standalone HTTP plugin monitor. `host_id` is OPTIONAL: set when the
check belongs to a registered Host (then it surfaces on the host hub), or
NULL for a free-standing custom destination. Type-specific settings live in
`config_json` so adding a new type never widens this table.
"""
__tablename__ = "monitors"
id: Mapped[str] = mapped_column(
String(36), primary_key=True, default=lambda: str(uuid.uuid4())
)
name: Mapped[str] = mapped_column(String(128), nullable=False)
type: Mapped[str] = mapped_column(String(16), nullable=False)
# Address (icmp/tcp/dns) or URL (http) — free-form, not tied to a Host row.
target: Mapped[str] = mapped_column(String(2048), nullable=False)
host_id: Mapped[str | None] = mapped_column(
String(36), ForeignKey("hosts.id", ondelete="CASCADE"), nullable=True, index=True
)
# Type-specific config as a JSON object. Keys by type:
# icmp -> {}
# tcp -> {"port": int}
# dns -> {"expected_ip": str | None}
# http -> {"method", "expected_status", "content_match", "headers",
# "timeout_seconds", "follow_redirects", "verify_ssl"}
config_json: Mapped[str] = mapped_column(Text, nullable=False, default="{}")
enabled: Mapped[bool] = mapped_column(Boolean, nullable=False, default=True)
# 0 = use the global monitors.poll_interval_seconds setting.
check_interval_seconds: Mapped[int] = mapped_column(Integer, nullable=False, default=0)
# Updated after each check so the scheduler can compute next-run efficiently.
last_checked_at: Mapped[datetime | None] = mapped_column(
DateTime(timezone=True), nullable=True
)
created_at: Mapped[datetime] = mapped_column(
DateTime(timezone=True), nullable=False,
default=lambda: datetime.now(timezone.utc),
)
__table_args__ = (
CheckConstraint(_TYPE_CHECK, name="ck_monitors_type"),
)
@property
def config(self) -> dict:
try:
return json.loads(self.config_json or "{}")
except (ValueError, TypeError):
return {}
def set_config(self, cfg: dict) -> None:
self.config_json = json.dumps(cfg or {})
class MonitorResult(Base):
"""One check result for a Monitor, across all types.
Superset of the old per-type result columns: response_ms is universal,
while status_code/content_matched/tls_expires_at (http) and resolved_ip
(dns) are populated only for the types that produce them.
"""
__tablename__ = "monitor_results"
id: Mapped[str] = mapped_column(
String(36), primary_key=True, default=lambda: str(uuid.uuid4())
)
monitor_id: Mapped[str] = mapped_column(
String(36), ForeignKey("monitors.id", ondelete="CASCADE"),
nullable=False, index=True,
)
checked_at: Mapped[datetime] = mapped_column(
DateTime(timezone=True), nullable=False,
default=lambda: datetime.now(timezone.utc), index=True,
)
is_up: Mapped[bool] = mapped_column(Boolean, nullable=False, default=False)
response_ms: Mapped[float | None] = mapped_column(Float, nullable=True)
status_code: Mapped[int | None] = mapped_column(Integer, nullable=True) # http
resolved_ip: Mapped[str | None] = mapped_column(String(255), nullable=True) # dns
content_matched: Mapped[bool | None] = mapped_column(Boolean, nullable=True) # http
tls_expires_at: Mapped[datetime | None] = mapped_column(
DateTime(timezone=True), nullable=True # http
)
error_msg: Mapped[str | None] = mapped_column(String(512), nullable=True)
__table_args__ = (
# The status/heartbeat queries page the latest N per monitor by time.
Index("ix_monitor_results_monitor_checked", "monitor_id", "checked_at"),
)