From 5ef18c992c6e3efaaa1ce9374d1233a9fce5f3ea Mon Sep 17 00:00:00 2001 From: Bryan Van Deusen Date: Fri, 9 Oct 2026 08:33:32 -0400 Subject: [PATCH] feat(alerts): Send test on a rule checks that its alerts reach someone (#5466) #5466 "Send test: check a rule's delivery from the rule". A rule's page gets a Send test button (operator role), and MCP gets send_test_alert. Each sends a TEST alert for that rule now, by email to everyone who gets alerts and by webhook. The alert carries the rule's name and condition, at the latest reading of what it watches; a rule covering hosts uses its first host. The result for each channel shows under the button, or "Nothing is set up to send to". Nothing fires, no event is written, and the test is audited as alert_rule.tested. An alert with no reading now says "has no reading yet" instead of "is None". Docs: alerting.md also gains the burst email from #5465 "One email per burst of alerts". Co-Authored-By: Claude Opus 5.5 --- docs/core/alerting.md | 21 +++- docs/core/mcp.md | 2 +- steward/alerts/routes.py | 15 +++ steward/alerts/service.py | 53 ++++++++++ steward/audit/service.py | 1 + steward/core/mail.py | 2 + steward/mcp/tools/alerts.py | 23 +++- steward/templates/alerts/_test_result.html | 4 + steward/templates/alerts/rules_form.html | 9 ++ tests/core/test_alert_test_alert.py | 52 +++++++++ tests/core/test_mail.py | 6 +- tests/integration/test_alert_send_test.py | 116 +++++++++++++++++++++ tests/mcp/test_monitoring_writes.py | 2 +- 13 files changed, 299 insertions(+), 7 deletions(-) create mode 100644 steward/templates/alerts/_test_result.html create mode 100644 tests/core/test_alert_test_alert.py create mode 100644 tests/integration/test_alert_send_test.py diff --git a/docs/core/alerting.md b/docs/core/alerting.md index 3b7f504..cf461f3 100644 --- a/docs/core/alerting.md +++ b/docs/core/alerting.md @@ -134,6 +134,14 @@ Configured at `/settings/` under SMTP. Settings: Email is skipped if `smtp.host` is empty. +Alerts that arrive together go out as one email (`core/alert_burst.py`). The +first alert opens a 30-second window. When it closes, each recipient gets one +message listing every alert in the window, oldest first, in their own zone, +with a subject like "[Steward] 3 alerts: 2 firing, 1 resolved". One alert on +its own keeps its usual email. Every alert's event records whether the email +went. A burst still waiting when Steward stops is not sent; its alerts stay on +the Alerts page and in the history. + ### Webhook Configured at `/settings/` under Webhook. The body is a Jinja2 template rendered to JSON. Content-Type is always `application/json`. If the rendered template is not valid JSON, the delivery is logged as failed and no request is sent. @@ -154,7 +162,18 @@ Default template (Discord-compatible): {"content": "**{{ alert.state }}** — {{ alert.resource }} — {{ alert.rule_name }} ({{ alert.metric }} = {{ alert.value }})"} ``` -Webhook is skipped if `webhook.url` is empty. +Webhook is skipped if `webhook.url` is empty. Webhooks are not gathered into +bursts: each alert posts on its own, at once. + +### Send test + +**Send test** on a rule's page (operator role), or `send_test_alert` over MCP, +sends a `TEST` alert for that rule now, by email and webhook, the same way a +real alert would go. It uses the rule's name and condition and the latest +reading of what the rule watches. A rule that covers hosts uses its first +host. Each channel's result shows under the button. Nothing fires, the rule's +history doesn't change, and the test is recorded in the audit log as +`alert_rule.tested`. --- diff --git a/docs/core/mcp.md b/docs/core/mcp.md index 5148d1f..271576e 100644 --- a/docs/core/mcp.md +++ b/docs/core/mcp.md @@ -86,7 +86,7 @@ Each role includes the tools of the roles above it in this list. |---|---| | Automation | `run_playbook`, `cancel_ansible_run`, `create_schedule`, `update_schedule`, `delete_schedule` | | Monitors | `create_monitor`, `update_monitor`, `set_monitor_enabled` | -| Alerts | `create_alert_rule`, `update_alert_rule`, `acknowledge_alert`, `create_maintenance_window`, `end_maintenance_window` | +| Alerts | `create_alert_rule`, `update_alert_rule`, `acknowledge_alert`, `send_test_alert`, `create_maintenance_window`, `end_maintenance_window` | ### Admin: destructive and remediation diff --git a/steward/alerts/routes.py b/steward/alerts/routes.py index c49e49c..8084621 100644 --- a/steward/alerts/routes.py +++ b/steward/alerts/routes.py @@ -420,6 +420,21 @@ async def toggle_rule(rule_id: str): return redirect(url_for("alerts.list_alerts")) +@alerts_bp.post("/rules//test") +@require_role(UserRole.operator) +async def test_rule(rule_id: str): + """Send test (#5466): a TEST notification for this rule, now, by every + channel its alerts use; the result shows under the button.""" + try: + sent = await alert_svc.send_test(current_app, rule_id) + except LookupError: + return "Not found", 404 + await log_audit(current_app, session.get("user_id"), session.get("username", ""), + "alert_rule.tested", entity_type="alert_rule", entity_id=sent["rule"], + detail={"rule_id": rule_id, "ok": sent["ok"], "lines": sent["lines"]}) + return await render_template("alerts/_test_result.html", ok=sent["ok"], lines=sent["lines"]) + + @alerts_bp.post("/rules//delete") @require_role(UserRole.admin) async def delete_rule(rule_id: str): diff --git a/steward/alerts/service.py b/steward/alerts/service.py index f591c70..77124d8 100644 --- a/steward/alerts/service.py +++ b/steward/alerts/service.py @@ -602,6 +602,59 @@ def part_of(rule: AlertRule, rules) -> AlertRule | None: return next((r for r in rules if r.id == rule.parent_id), None) if rule.parent_id else None +async def test_alert(db, rule: AlertRule) -> dict: + """What a Send test carries for `rule` (#5466): a TEST alert with the + rule's name and condition, on its resource at the latest reading. A rule + covering hosts tests on its first host.""" + target = rule + if rule.covers != "resource": + target = (await db.execute( + select(AlertRule).where(AlertRule.parent_id == rule.id) + .order_by(AlertRule.resource_name).limit(1))).scalar_one_or_none() or rule + return {"rule_name": rule.name, "state": "TEST", "metric": rule.metric_name, + "value": await latest_value(db, target) if target.resource_name else None, + "operator": rule.operator.value, "threshold": rule.threshold, + "resource": target.resource_name or covers_text(rule), + "source_module": rule.source_module, + "note": "A test of this rule's notifications. Nothing has fired.", + "timestamp": datetime.now(timezone.utc).isoformat()} + + +async def send_test(app, rule_id: str) -> dict: + """Send a TEST notification for a rule now, by every channel a real alert + of it would use: email to whoever gets alerts, each in their own zone, and + the webhook. Not held for a burst, and not recorded in the rule's history. + + Returns {"rule", "ok", "lines"}: ok when something was sent and nothing + failed; lines say what happened to each channel. LookupError for no rule. + """ + from urllib.parse import urlsplit + from steward.core.mail import recipients + from steward.core.notifications import dispatch_notifications + smtp, webhook = app.config.get("SMTP", {}), app.config.get("WEBHOOK", {}) + async with app.db_sessionmaker() as db: + rule = await db.get(AlertRule, rule_id) + if rule is None: + raise LookupError(rule_id) + alert = await test_alert(db, rule) + to = await recipients(db, "alerts", smtp) if smtp.get("host") else [] + results = await dispatch_notifications(smtp, webhook, alert, to) + lines = [] + if email := results.get("email"): + who = ", ".join(r.address for r in to) + lines.append(f"Email sent to {who}." if email["sent"] + else f"Email to {who} failed: {email['error']}") + if hook := results.get("webhook"): + host = urlsplit(webhook["url"]).hostname or webhook["url"] + lines.append(f"Webhook to {host} delivered." if hook["sent"] + else f"Webhook to {host} failed: {hook['error']}") + if not lines: + lines.append("Nothing is set up to send to: no email server with someone to send to, " + "and no webhook.") + return {"rule": rule.name, "ok": bool(results) and all(r["sent"] for r in results.values()), + "lines": lines, "email": results.get("email"), "webhook": results.get("webhook")} + + class PartOfCoveringRule(ValueError): """A child of a covering rule is changed through its parent, never alone.""" diff --git a/steward/audit/service.py b/steward/audit/service.py index 67c4f72..7366fc2 100644 --- a/steward/audit/service.py +++ b/steward/audit/service.py @@ -149,6 +149,7 @@ ACTIONS = { "alert_rule.disabled": "turned off alert rule", "alert_rule.deleted": "deleted alert rule", "alert_rule.acknowledged": "acknowledged the alert from", + "alert_rule.tested": "sent a test alert for rule", "alert_rule.action_set": "set the playbook run by alert rule", "alert_rule.action_cleared": "removed the playbook from alert rule", "maintenance_window.created": "scheduled maintenance", diff --git a/steward/core/mail.py b/steward/core/mail.py index 0a022e3..87ae983 100644 --- a/steward/core/mail.py +++ b/steward/core/mail.py @@ -222,6 +222,8 @@ def alert_facts(alert: dict, tz: tzinfo, show_utc: bool = True) -> dict: elif state == "resolved": headline = f"{metric_label(metric)} on {resource} is back within its limit" + ( f", at {reading}." if reading else ".") + elif reading is None: + headline = f"{metric_label(metric)} on {resource} has no reading yet." else: headline = f"{metric_label(metric)} on {resource} is {reading}." at = datetime.fromisoformat(alert["timestamp"]) diff --git a/steward/mcp/tools/alerts.py b/steward/mcp/tools/alerts.py index 4718ef8..e386e8d 100644 --- a/steward/mcp/tools/alerts.py +++ b/steward/mcp/tools/alerts.py @@ -6,7 +6,7 @@ from datetime import datetime, timedelta, timezone from sqlalchemy import select from steward.alerts import service as alert_svc -from steward.mcp._context import current_caller +from steward.mcp._context import current_caller, mcp_app from steward.models.alerts import AlertOperator, AlertRule, AlertState, MaintenanceWindow from steward.models.users import UserRole @@ -391,6 +391,25 @@ async def acknowledge_alert(rule: str) -> dict: "acknowledged_at": iso(state.acknowledged_at)} +async def send_test_alert(rule: str) -> dict: + """Send a TEST notification for a rule now, to check it reaches someone: + email to everyone who gets alerts (each in their own zone) and the + webhook, the same channels a real alert of this rule would use. Nothing + fires and the rule's history is unchanged. + + rule: the rule's name or id (list_alert_rules). Returns `ok` (something + was sent and nothing failed) and one line per channel saying what + happened. Recorded in the audit log. + """ + async with db_session() as db: + r = await _rule(db, rule) + sent = await alert_svc.send_test(mcp_app(), r.id) + await audit_mcp("alert_rule.tested", "alert_rule", r.name, + {"rule_id": r.id, "ok": sent["ok"], "lines": sent["lines"]}) + return {"rule_id": r.id, "rule_name": r.name, "ok": sent["ok"], "result": sent["lines"], + "email": sent["email"], "webhook": sent["webhook"]} + + def _when(value: str, what: str) -> datetime: try: dt = datetime.fromisoformat(value.strip().replace("Z", "+00:00")) @@ -467,7 +486,7 @@ def register(register_tool) -> None: for fn in (list_alerts, get_alert_history, list_alert_rules, list_maintenance_windows, get_alert_rule_options): register_tool(fn, role=UserRole.viewer, read_only=True) - for fn in (create_alert_rule, update_alert_rule, acknowledge_alert, + for fn in (create_alert_rule, update_alert_rule, acknowledge_alert, send_test_alert, create_maintenance_window, end_maintenance_window): register_tool(fn, role=UserRole.operator, read_only=False) register_tool(delete_alert_rule, role=UserRole.admin, read_only=False) diff --git a/steward/templates/alerts/_test_result.html b/steward/templates/alerts/_test_result.html new file mode 100644 index 0000000..303e129 --- /dev/null +++ b/steward/templates/alerts/_test_result.html @@ -0,0 +1,4 @@ +{# alerts/_test_result.html — what a rule's Send test did (#5466), one line per channel #} +
+ {% for line in lines %}
{{ line }}
{% endfor %} +
diff --git a/steward/templates/alerts/rules_form.html b/steward/templates/alerts/rules_form.html index 584f655..82bf25d 100644 --- a/steward/templates/alerts/rules_form.html +++ b/steward/templates/alerts/rules_form.html @@ -141,6 +141,15 @@ {%- else %}It notifies nobody yet: no email server with recipients and no webhook is set {%- if session.user_role == 'admin' %} (Settings → Notifications){% endif %}.{% endif -%}

+ {% if rule %} +
+ +

Sends a TEST alert for this rule as saved, now, to everywhere above. + Nothing fires, and the rule's history doesn't change.

+
+
+ {% endif %}