862b290545
Curated events were re-read from the newest 25 entries per channel on every collect, and the server appended whatever arrived. One real event was therefore stored again on every cadence tick for as long as it stayed in that window, so event counts measured how long a box had been up rather than what had happened to it. The agent now keeps a per-channel watermark of the last Windows event record id it reported, persisted under ProgramData so a service restart does not replay history, and sends eventrecordid with each event. The server dedupes on (assetid, eventchannel, eventrecordid) behind a unique index, so a replay from an older agent cannot double-store either. Six inventory keys the agent had been sending all along - bios, disks, domain, memory, osinfo, processor - were parsed, logged as unpersisted, and dropped. They now have columns and are written on check-in, along with user context, pending reboot, and the cpupercent/memorypercent/ idleseconds fields that were declared in the contract but always null. Session-scoped data cannot be read from the agent's LocalSystem session 0 at all: mapped drives and user-owned listeners do not exist there. A probe runs in the interactive user's session and hands results back through a spool file. It is not yet producing output on a live endpoint; the column and panel are in place and read empty until that is fixed. Also bounds the user-context script wait. It was WaitForSingleObject with INFINITE, so a hung script pinned the command worker permanently.
411 lines
18 KiB
Python
411 lines
18 KiB
Python
"""Endpoint RMM telemetry models.
|
|
|
|
Two shapes, one per concern:
|
|
|
|
EndpointState The cached CURRENT snapshot for an asset - one row per asset,
|
|
overwritten in place on every agent report. Holds the fast-
|
|
moving metrics (uptime, idle, last boot, logged-in user) and
|
|
the latest inventory snapshots (services/processes, open ports,
|
|
interfaces, scheduled tasks) as serialized JSON. This is the
|
|
cached-current-value half of the ledger pattern: it moves in
|
|
the SAME commit as any EndpointEvent rows the report carries.
|
|
|
|
EndpointEvent Append-only record of a curated Windows Event Log entry. The
|
|
agent only forwards a fixed catalog of IDs (logon/logoff,
|
|
privilege use, service install, unexpected shutdown, DCOM,
|
|
crash) so this stays a readable security/stability history
|
|
rather than a raw event firehose.
|
|
|
|
Both link to the core Asset via assetid (FK, ON DELETE CASCADE), so an endpoint
|
|
is the same physical thing as its Computer/Asset - no duplicate identity. The
|
|
agent reports a hostname; the plugin resolves it to the existing Computer.
|
|
"""
|
|
|
|
from datetime import datetime
|
|
|
|
from shopdb.api import db
|
|
|
|
|
|
def _utciso(value):
|
|
"""Naive-UTC datetime -> ISO string marked UTC, or None.
|
|
|
|
Every datetime this plugin stores is naive UTC. Serializing without the Z
|
|
leaves the receiver to guess, and JavaScript guesses browser-local.
|
|
"""
|
|
return value.isoformat() + 'Z' if value else None
|
|
|
|
|
|
class EndpointState(db.Model):
|
|
"""Latest reported state for a single endpoint (one row per asset)."""
|
|
|
|
__tablename__ = 'endpointstates'
|
|
|
|
endpointstateid = db.Column(db.Integer, primary_key=True)
|
|
|
|
# The asset this state belongs to. Unique: exactly one current snapshot per
|
|
# endpoint. Resolved from the reported hostname against the computers plugin.
|
|
assetid = db.Column(
|
|
db.Integer,
|
|
db.ForeignKey('assets.assetid', ondelete='CASCADE'),
|
|
unique=True,
|
|
nullable=False,
|
|
index=True,
|
|
)
|
|
|
|
# Read-only view of the owning asset. No cascade or backref: the asset side
|
|
# must not gain a dependency on this plugin.
|
|
asset = db.relationship('Asset', lazy='joined', viewonly=True)
|
|
|
|
hostname = db.Column(db.String(100), nullable=True)
|
|
loggedinuser = db.Column(db.String(100), nullable=True)
|
|
agentversion = db.Column(db.String(40), nullable=True)
|
|
# The interactive tray app versions independently of the agent service, so
|
|
# it is reported and cached as its own field (AGENT-CONTRACT.md 2). Null
|
|
# when the endpoint has no tray installed or an older agent that predates
|
|
# the field.
|
|
trayversion = db.Column(db.String(40), nullable=True)
|
|
|
|
# Fast-moving metrics
|
|
lastboottime = db.Column(db.DateTime, nullable=True)
|
|
uptimeseconds = db.Column(db.BigInteger, nullable=True)
|
|
idleseconds = db.Column(db.BigInteger, nullable=True)
|
|
cpupercent = db.Column(db.Float, nullable=True)
|
|
memorypercent = db.Column(db.Float, nullable=True)
|
|
|
|
# Latest inventory snapshots, serialized JSON (MEDIUMTEXT). Kept as the
|
|
# newest snapshot rather than one row per sample: the agent reports the full
|
|
# list each cycle and history of these lists is not what operators ask for.
|
|
servicesjson = db.Column(db.Text(length=16777215), nullable=True)
|
|
portsjson = db.Column(db.Text(length=16777215), nullable=True)
|
|
interfacesjson = db.Column(db.Text(length=16777215), nullable=True)
|
|
scheduledtasksjson = db.Column(db.Text(length=16777215), nullable=True)
|
|
|
|
# Hardware and OS inventory, same newest-snapshot rule as above. These
|
|
# arrived on the wire well before this and were logged as unknown keys, so
|
|
# the shapes are already proven against real endpoint traffic.
|
|
disksjson = db.Column(db.Text(length=16777215), nullable=True)
|
|
biosjson = db.Column(db.Text(length=16777215), nullable=True)
|
|
memoryjson = db.Column(db.Text(length=16777215), nullable=True)
|
|
processorjson = db.Column(db.Text(length=16777215), nullable=True)
|
|
osinfojson = db.Column(db.Text(length=16777215), nullable=True)
|
|
|
|
# Session-scoped data collected by a probe running as the signed-in user
|
|
# (mapped drives, user-owned listeners). Null when nobody is signed in,
|
|
# which is a normal unattended state rather than a fault.
|
|
usercontextjson = db.Column(db.Text(length=16777215), nullable=True)
|
|
|
|
# Windows domain or workgroup membership. A scalar, not JSON: it is a single
|
|
# value the fleet view filters and groups on.
|
|
domain = db.Column(db.String(255), nullable=True)
|
|
|
|
# True when the endpoint is holding a reboot. Null = never reported, which
|
|
# is deliberately distinct from a confident False.
|
|
pendingreboot = db.Column(db.Boolean, nullable=True)
|
|
|
|
# Warn-only reconciler output, refreshed on every agent check-in in the same
|
|
# commit as servicesjson. Each entry names a policy/shipped-attached monitor
|
|
# the endpoint is expected to be running but the latest snapshot does not
|
|
# report as running, so an operator sees the drift and intervenes manually.
|
|
# This is warn-only: nothing here is auto-remediated. Null = never computed.
|
|
driftjson = db.Column(db.Text(length=16777215), nullable=True)
|
|
|
|
# Last-known MTConnect agent run state, cached so the WebUI toggle can show
|
|
# a real state between command polls. Set by the agent's mtconnect_status
|
|
# result. Null = never reported (unknown).
|
|
mtconnectrunning = db.Column(db.Boolean, nullable=True)
|
|
mtconnectcheckedat = db.Column(db.DateTime, nullable=True)
|
|
|
|
# When the agent last successfully reported. There is no heartbeat separate
|
|
# from the report, so silence past the quiet-hours window is the only
|
|
# evidence an endpoint has stopped checking in.
|
|
lastreportedat = db.Column(db.DateTime, nullable=True, index=True)
|
|
|
|
createdat = db.Column(db.DateTime, nullable=False, default=datetime.utcnow)
|
|
|
|
def __repr__(self):
|
|
return f"<EndpointState {self.hostname or self.assetid}>"
|
|
|
|
def to_dict(self):
|
|
return {
|
|
'endpointstateid': self.endpointstateid,
|
|
'assetid': self.assetid,
|
|
'hostname': self.hostname,
|
|
'loggedinuser': self.loggedinuser,
|
|
'agentversion': self.agentversion,
|
|
'trayversion': self.trayversion,
|
|
'lastboottime': _utciso(self.lastboottime),
|
|
'uptimeseconds': self.uptimeseconds,
|
|
'idleseconds': self.idleseconds,
|
|
'cpupercent': self.cpupercent,
|
|
'memorypercent': self.memorypercent,
|
|
'domain': self.domain,
|
|
'pendingreboot': self.pendingreboot,
|
|
'mtconnectrunning': self.mtconnectrunning,
|
|
'mtconnectcheckedat': _utciso(self.mtconnectcheckedat),
|
|
'lastreportedat': _utciso(self.lastreportedat),
|
|
'createdat': _utciso(self.createdat),
|
|
}
|
|
|
|
|
|
class EndpointCommand(db.Model):
|
|
"""A queued operator command targeting one endpoint agent.
|
|
|
|
The WebUI enqueues a named action (server allowlist only, never an arbitrary
|
|
command string) for an asset. The agent claims the next pending command on
|
|
its long-poll, executes it locally, and reports the outcome back, which
|
|
moves the row to a terminal status in the same request. This is a small
|
|
at-most-one-in-flight queue, not a general job system.
|
|
"""
|
|
|
|
__tablename__ = 'endpointcommands'
|
|
|
|
# Lifecycle: pending -> running -> succeeded|failed. A command is claimed
|
|
# (pending -> running) by exactly one agent poll, then finalized by its
|
|
# result post. A command that is claimed but whose agent never reports back
|
|
# (agent killed/uninstalled/offline mid-flight) would otherwise pin the
|
|
# at-most-one guard forever; the enqueue path reaps such rows to STATUS_EXPIRED
|
|
# so a dead in-flight command cannot wedge the endpoint's command queue.
|
|
STATUS_PENDING = 'pending'
|
|
STATUS_RUNNING = 'running'
|
|
STATUS_SUCCEEDED = 'succeeded'
|
|
STATUS_FAILED = 'failed'
|
|
STATUS_EXPIRED = 'expired'
|
|
|
|
# In-flight staleness ceilings. A RUNNING command is expected to finish in
|
|
# seconds to a couple of minutes; the generous ceiling tolerates a slow
|
|
# app install while still freeing the queue when the agent is gone. A
|
|
# PENDING command is claimed within one poll (~25s) by a live agent, so a
|
|
# long-unclaimed row means the agent is offline and the queue should not
|
|
# stay wedged behind it.
|
|
RUNNING_STALE_SECONDS = 15 * 60
|
|
PENDING_STALE_SECONDS = 60 * 60
|
|
|
|
endpointcommandid = db.Column(db.Integer, primary_key=True)
|
|
|
|
assetid = db.Column(
|
|
db.Integer,
|
|
db.ForeignKey('assets.assetid', ondelete='CASCADE'),
|
|
nullable=False,
|
|
index=True,
|
|
)
|
|
|
|
asset = db.relationship('Asset', lazy='joined', viewonly=True)
|
|
|
|
# The allowlisted action name (for example 'mtconnect_start'). Validated
|
|
# against the server allowlist at enqueue time, so only known names land.
|
|
action = db.Column(db.String(64), nullable=False)
|
|
|
|
# Generic actions (service_* / process_*) carry their subject over the wire:
|
|
# target is a Windows service name or a process image (for example
|
|
# 'DncMain.exe'); targetkind is 'service' or 'process'. Fixed-target actions
|
|
# (mtconnect_*) leave both null. This is what lets a site build the agent
|
|
# once and drive every future policy/app purely from server data.
|
|
target = db.Column(db.String(255), nullable=True)
|
|
targetkind = db.Column(db.String(16), nullable=True)
|
|
|
|
status = db.Column(db.String(20), nullable=False,
|
|
default=STATUS_PENDING, index=True)
|
|
|
|
# Who queued it (userid) for audit; the agent never sees this.
|
|
requestedby = db.Column(db.Integer, nullable=True)
|
|
|
|
# The content digest as it stood WHEN AN OPERATOR APPROVED THIS PUSH.
|
|
#
|
|
# Delivery descriptors are re-resolved at claim time, minutes or hours
|
|
# later, so without this the bytes an operator approved and the bytes the
|
|
# agent runs are two different lookups against mutable content. Whoever can
|
|
# replace a script body or an installer package could swap it in the gap
|
|
# and the digest would faithfully describe the new bytes.
|
|
#
|
|
# Set at enqueue for run_script and update_app; null for actions that
|
|
# deliver no content. The claim path refuses a command whose descriptor no
|
|
# longer matches, so a swap fails the command instead of shipping.
|
|
pinnedsha256 = db.Column(db.String(64), nullable=True)
|
|
|
|
# Agent-reported outcome. resultjson holds structured detail (for example
|
|
# the observed MTConnect running state); resultmessage is a short summary.
|
|
resultmessage = db.Column(db.Text, nullable=True)
|
|
resultjson = db.Column(db.Text(length=16777215), nullable=True)
|
|
|
|
createdat = db.Column(db.DateTime, nullable=False, default=datetime.utcnow)
|
|
claimedat = db.Column(db.DateTime, nullable=True)
|
|
completedat = db.Column(db.DateTime, nullable=True)
|
|
|
|
__table_args__ = (
|
|
db.Index('idx_endpointcommands_assetstatus', 'assetid', 'status'),
|
|
)
|
|
|
|
def __repr__(self):
|
|
return f"<EndpointCommand {self.action} asset={self.assetid} {self.status}>"
|
|
|
|
def to_dict(self):
|
|
return {
|
|
'endpointcommandid': self.endpointcommandid,
|
|
'assetid': self.assetid,
|
|
'action': self.action,
|
|
'target': self.target,
|
|
'targetkind': self.targetkind,
|
|
'status': self.status,
|
|
'requestedby': self.requestedby,
|
|
'resultmessage': self.resultmessage,
|
|
'createdat': _utciso(self.createdat),
|
|
'claimedat': _utciso(self.claimedat),
|
|
'completedat': _utciso(self.completedat),
|
|
}
|
|
|
|
|
|
class EndpointEvent(db.Model):
|
|
"""A single curated Windows Event Log record forwarded by the agent."""
|
|
|
|
__tablename__ = 'endpointevents'
|
|
|
|
endpointeventid = db.Column(db.Integer, primary_key=True)
|
|
|
|
assetid = db.Column(
|
|
db.Integer,
|
|
db.ForeignKey('assets.assetid', ondelete='CASCADE'),
|
|
nullable=False,
|
|
index=True,
|
|
)
|
|
|
|
asset = db.relationship('Asset', lazy='joined', viewonly=True)
|
|
|
|
# The Windows Event ID (for example 4625 failed logon, 41 unexpected
|
|
# shutdown, 1001 application crash). Indexed because alerting queries filter
|
|
# on it.
|
|
eventid = db.Column(db.Integer, nullable=False, index=True)
|
|
|
|
# Event Log channel the record came from: Security, System, Application.
|
|
eventchannel = db.Column(db.String(50), nullable=True)
|
|
eventsource = db.Column(db.String(255), nullable=True)
|
|
|
|
# The Windows event record id: monotonic and unique within its channel on a
|
|
# given machine. This is the dedup key. The agent also uses it as its
|
|
# harvest watermark, so in normal operation a record is sent once; the
|
|
# unique index below is what guarantees it even when an agent is reinstalled
|
|
# (losing its watermark) or two reports overlap. Nullable because an agent
|
|
# older than this field still reports, and MySQL permits repeated NULLs in a
|
|
# unique index, so those degrade to the previous append-only behavior rather
|
|
# than failing.
|
|
eventrecordid = db.Column(db.BigInteger, nullable=True)
|
|
|
|
# When the event happened on the endpoint (from the record), not when it
|
|
# reached ShopDB.
|
|
eventtime = db.Column(db.DateTime, nullable=True, index=True)
|
|
|
|
eventmessage = db.Column(db.Text, nullable=True)
|
|
|
|
# Structured fields the agent extracted for this ID (for example the target
|
|
# account on a 4625). Advisory index only, not part of dedup.
|
|
payloadjson = db.Column(db.Text(length=16777215), nullable=True)
|
|
|
|
recordedat = db.Column(db.DateTime, nullable=False, default=datetime.utcnow)
|
|
|
|
__table_args__ = (
|
|
db.Index('idx_endpointevents_assetevent', 'assetid', 'eventtime'),
|
|
db.Index('idx_endpointevents_assetrecord',
|
|
'assetid', 'eventchannel', 'eventrecordid', unique=True),
|
|
)
|
|
|
|
def __repr__(self):
|
|
return f"<EndpointEvent {self.eventid} asset={self.assetid}>"
|
|
|
|
def to_dict(self):
|
|
return {
|
|
'endpointeventid': self.endpointeventid,
|
|
'assetid': self.assetid,
|
|
'eventid': self.eventid,
|
|
'eventrecordid': self.eventrecordid,
|
|
'eventchannel': self.eventchannel,
|
|
'eventsource': self.eventsource,
|
|
'eventtime': _utciso(self.eventtime),
|
|
'eventmessage': self.eventmessage,
|
|
'recordedat': _utciso(self.recordedat),
|
|
}
|
|
|
|
|
|
class EndpointApplication(db.Model):
|
|
"""One installed-software (SBOM) entry reported for an endpoint.
|
|
|
|
The agent enumerates both Windows Uninstall registry hives and emits a
|
|
per-item {displayname, displayversion, publisher}, de-duped on
|
|
name+version. The server stores that as a per-asset SNAPSHOT: on each
|
|
report the existing rows for the asset are replaced with the reported set,
|
|
mirroring how the services/ports inventories are kept as the newest list
|
|
rather than an append-only history. History of installs is not what
|
|
operators ask for; the current inventory is.
|
|
|
|
appid is groundwork for a future link to the core applications catalog
|
|
(the "Managed Applications" concept): a nullable, unenforced-by-match FK
|
|
that a later matcher can populate to associate a discovered install with a
|
|
managed Application row. No matching logic exists yet - the column and the
|
|
relationship are laid down here so the association can be added without a
|
|
second migration.
|
|
"""
|
|
|
|
__tablename__ = 'endpointapplications'
|
|
|
|
# BigInteger on MySQL, Integer on SQLite. This table is rewritten wholesale
|
|
# on every check-in, so the key never stops climbing and a 32-bit column
|
|
# eventually stops accepting inserts - see the migration for the arithmetic.
|
|
#
|
|
# The variant is not decoration: SQLite only autoincrements a column typed
|
|
# exactly INTEGER PRIMARY KEY, so a plain BigInteger makes every insert fail
|
|
# with a NOT NULL violation on the test database. Nothing caught that,
|
|
# because no test inserts into this table.
|
|
endpointapplicationid = db.Column(
|
|
db.BigInteger().with_variant(db.Integer, 'sqlite'), primary_key=True)
|
|
|
|
# The asset this install belongs to. Not unique: many applications per
|
|
# endpoint. CASCADE so removing the asset clears its SBOM snapshot.
|
|
assetid = db.Column(
|
|
db.Integer,
|
|
db.ForeignKey('assets.assetid', ondelete='CASCADE'),
|
|
nullable=False,
|
|
index=True,
|
|
)
|
|
|
|
asset = db.relationship('Asset', lazy='joined', viewonly=True)
|
|
|
|
# Wire fields, stored under the same lowercase/no-underscore names the agent
|
|
# emits. displayname is always present (the agent skips entries without it);
|
|
# displayversion and publisher default to '' on the agent side.
|
|
displayname = db.Column(db.String(255), nullable=False, index=True)
|
|
displayversion = db.Column(db.String(128), nullable=True)
|
|
publisher = db.Column(db.String(255), nullable=True)
|
|
|
|
# Groundwork link to the core applications catalog ("Managed Applications").
|
|
# Nullable and unpopulated for now: no discovery-to-catalog matcher is wired
|
|
# yet. SET NULL so retiring a managed Application does not delete discovered
|
|
# SBOM rows - the install is still on the machine, it is just no longer
|
|
# managed.
|
|
appid = db.Column(
|
|
db.Integer,
|
|
db.ForeignKey('applications.appid', ondelete='SET NULL'),
|
|
nullable=True,
|
|
index=True,
|
|
)
|
|
|
|
application = db.relationship('Application', lazy='select', viewonly=True)
|
|
|
|
# When this snapshot row was written (the report that produced it).
|
|
recordedat = db.Column(db.DateTime, nullable=False, default=datetime.utcnow)
|
|
|
|
__table_args__ = (
|
|
db.Index('idx_endpointapplications_assetname', 'assetid', 'displayname'),
|
|
)
|
|
|
|
def __repr__(self):
|
|
return f"<EndpointApplication {self.displayname!r} asset={self.assetid}>"
|
|
|
|
def to_dict(self):
|
|
return {
|
|
'endpointapplicationid': self.endpointapplicationid,
|
|
'assetid': self.assetid,
|
|
'displayname': self.displayname,
|
|
'displayversion': self.displayversion,
|
|
'publisher': self.publisher,
|
|
'appid': self.appid,
|
|
'recordedat': _utciso(self.recordedat),
|
|
}
|