jcoffey-dev
9435115ab7
Saved, shareable multi-panel dashboards (table/line/bar/single-stat panels via gridstack + uPlot, global + per-panel time range, JSON export/import) and threshold/absence alert rules with an ok/pending/firing evaluator and webhook/Slack/PagerDuty delivery. - New /metadata component: Postgres control-plane store for dashboards, panels, notification targets, alert rules/state, and delivery log -- see docs/phase-3-dashboard-design.md for why ClickHouse's MergeTree family isn't a fit for this access pattern (needs real row-level locking and read-your-writes consistency). - api/internal/dashboards: dashboard/panel CRUD, pure -- panel query execution stays client-side, reusing the existing /query endpoint. - New /alerting service: rule/target CRUD, a ticker-driven evaluator (claim-then-evaluate concurrency control, transactional-outbox delivery, query errors and threshold zero-rows never coerced into a false transition) and webhook/Slack/PagerDuty delivery with retry/backoff. See docs/phase-3-alerting-design.md for the full state-machine design and the four correctness properties it implements. - web: /dashboards and /alerts UIs; cli: sentryctl dashboards/alerts list/get/apply, seeding a future Terraform provider's JSON contract. - hack/alert-load-test: 500 rules against real ClickHouse data, real measured results in docs/phase-3-runbook.md. Five real bugs found by actually running this against a live stack (documented in the runbook, not just fixed silently): a latent Phase 2 bug where ClickHouse rejected the timestamp format used for earliest=/latest= queries; a "now" literal token injected into query text; a GridStack/uPlot layout-timing race; JS's Date.parse being too lenient to use as a timestamp-detection heuristic; a rule's "enabled" field silently defaulting to false when omitted; and the evaluator's claim-batch-size and worker-pool-concurrency defaulting to the same value, causing 500 concurrently-due rules to take 125s to cycle through instead of the configured 60s.
75 lines
2.6 KiB
Go
75 lines
2.6 KiB
Go
package evaluator
|
|
|
|
import (
|
|
"fmt"
|
|
|
|
"github.com/sentry/sentry/alerting/internal/queryclient"
|
|
"github.com/sentry/sentry/alerting/internal/rulestore"
|
|
)
|
|
|
|
// evaluateCondition implements fixes 3 and 4 from
|
|
// /docs/phase-3-alerting-design.md: a threshold rule's query must
|
|
// resolve to exactly one row, and zero (or more than one) rows is
|
|
// returned as an error, never coerced to a value -- "nothing ran" and
|
|
// "ran and found nothing" are different failure/success modes, and
|
|
// conflating them can hide the more alarming case. This function never
|
|
// itself decides "condition false" on an error; it returns an error, and
|
|
// the caller (evaluator.go) routes that to rulestore.RecordError, never
|
|
// to ComputeTransition.
|
|
func evaluateCondition(rule rulestore.Rule, result *queryclient.Result) (conditionTrue bool, value *float64, err error) {
|
|
switch rule.ConditionType {
|
|
case rulestore.ConditionAbsence:
|
|
// The window is whatever earliest=/latest= the rule's own query
|
|
// already expresses -- no separate window field, per the design doc.
|
|
return len(result.Rows) == 0, nil, nil
|
|
|
|
case rulestore.ConditionThreshold:
|
|
if len(result.Rows) != 1 {
|
|
return false, nil, fmt.Errorf("threshold rule query returned %d rows, want exactly 1", len(result.Rows))
|
|
}
|
|
row := result.Rows[0]
|
|
if len(row) == 0 {
|
|
return false, nil, fmt.Errorf("threshold rule query returned a row with no columns")
|
|
}
|
|
v, ok := toFloat64(row[0])
|
|
if !ok {
|
|
return false, nil, fmt.Errorf("threshold rule's first column value %v is not numeric", row[0])
|
|
}
|
|
if rule.Comparator == nil || rule.ThresholdValue == nil {
|
|
return false, nil, fmt.Errorf("threshold rule is missing comparator or threshold_value")
|
|
}
|
|
return compare(v, *rule.Comparator, *rule.ThresholdValue), &v, nil
|
|
|
|
default:
|
|
return false, nil, fmt.Errorf("unknown condition_type %q", rule.ConditionType)
|
|
}
|
|
}
|
|
|
|
func compare(value float64, comparator rulestore.Comparator, threshold float64) bool {
|
|
switch comparator {
|
|
case rulestore.Gt:
|
|
return value > threshold
|
|
case rulestore.Gte:
|
|
return value >= threshold
|
|
case rulestore.Lt:
|
|
return value < threshold
|
|
case rulestore.Lte:
|
|
return value <= threshold
|
|
case rulestore.Eq:
|
|
return value == threshold
|
|
case rulestore.Ne:
|
|
return value != threshold
|
|
default:
|
|
return false
|
|
}
|
|
}
|
|
|
|
// toFloat64 handles the one shape /query responses actually come back
|
|
// as: Go's encoding/json always decodes JSON numbers into interface{}
|
|
// as float64, regardless of whether ClickHouse serialized an integer or
|
|
// a float -- there's no separate int64 case to handle.
|
|
func toFloat64(v any) (float64, bool) {
|
|
f, ok := v.(float64)
|
|
return f, ok
|
|
}
|