Files
zeekayandGitHub babbdec4b8 debrand: SigNoz -> O11y across the tree (+ otel-collector v0.144.7, schema cutover migration) (#28)
* debrand: signoz/SigNoz/SIGNOZ -> o11y/O11y/O11Y across the tree

Rename all SigNoz *branding* to O11y in file contents, file/dir names,
package names, env vars, config keys, comments, docs, SDK code, and the
ClickHouse schema identifiers the querier reads.

Collector dependency:
- Repoint github.com/hanzoai/signoz-otel-collector ->
  github.com/hanzoai/otel-collector and bump v0.144.6 -> v0.144.7
  (go.mod + go.sum reconciled via go mod tidy). Internal package paths
  also debranded upstream: signozschemamigrator -> o11yschemamigrator,
  signozlogspipelineprocessor -> o11ylogspipelineprocessor.
- Fix: v0.144.7 dropped otelconst.DistributedFieldKeysTable from the
  collector's public constants; pin FieldKeysTable = "distributed_field_keys"
  locally in pkg/telemetrymetadata/tables.go.

Package/dir renames (package decl + all importers):
  pkg/apiserver/signozapiserver -> o11yapiserver, plus signozalertmanager,
  signozauthzapi, signozglobal, signozquerier, signozruler, and pkg/signoz -> pkg/o11y.

Schema (read plane): signoz_traces/metrics/logs/metadata/analytics/meter and
signoz_index_v3/index_v2/error_index_v2/spans (+ distributed_*) -> o11y_*.
Data-preserving cutover migration added:
  deploy/clickhouse/migrations/0001_rename_signoz_to_o11y.sql
  (metadata-only RENAME DATABASE/TABLE, no drop/recreate).

Preserved (attribution / external / wire contracts):
- LICENSE + NOTICE verbatim (incl. "software from SigNoz", "Copyright ... SigNoz Inc.").
- Upstream github.com/SigNoz/* repo URLs + @SigNoz/* CODEOWNERS teams.
- Billing resource-attribute regex "signoz.workspace.*" (external wire contract).

go build ./pkg/... ./cmd/... = 0; gofmt clean; renamed+schema pkg tests pass.

* o11y: bump collector to v0.144.8 (o11y_* physical schema writer) + lockstep deploy plan

Collector v0.144.8's schema-migrator + exporters now CREATE/WRITE the same
o11y_* physical databases/tables the querier reads and the RENAME migration
(0001_rename_signoz_to_o11y.sql) targets.

Cross-checked byte-identical: writer DB set == migration target DB set (6);
writer prefixed-table set == migration target table set (8); o11y querier
reads and cloud reads are subsets of that set. Zero table-name mismatches
between writer, readers, and migration.

Collector go.mod unchanged between v0.144.7 and v0.144.8 (identical go.mod
hash) — pure source rename, no module-graph change.

Adds deploy/clickhouse/DEPLOY_signoz_to_o11y_cutover.md: the lockstep order
(collector v0.144.8 -> RENAME migration -> o11y+cloud readers) with a
fresh/scratch-ClickHouse pre-flight (create via migrator, emit trace+metric,
query back) and rollback. Ships together or telemetry blackholes.

go build ./pkg/... ./cmd/community = 0.
2026-07-09 13:08:55 -07:00

597 lines
18 KiB
Go

package o11yalertmanager
import (
"context"
"time"
amConfig "github.com/prometheus/alertmanager/config"
"github.com/prometheus/common/model"
"github.com/hanzoai/o11y/pkg/alertmanager"
"github.com/hanzoai/o11y/pkg/alertmanager/alertmanagerserver"
"github.com/hanzoai/o11y/pkg/alertmanager/alertmanagerstore/sqlalertmanagerstore"
"github.com/hanzoai/o11y/pkg/alertmanager/nfmanager"
"github.com/hanzoai/o11y/pkg/errors"
"github.com/hanzoai/o11y/pkg/factory"
"github.com/hanzoai/o11y/pkg/modules/organization"
"github.com/hanzoai/o11y/pkg/sqlstore"
"github.com/hanzoai/o11y/pkg/types"
"github.com/hanzoai/o11y/pkg/types/alertmanagertypes"
"github.com/hanzoai/o11y/pkg/types/authtypes"
"github.com/hanzoai/o11y/pkg/types/ruletypes"
"github.com/hanzoai/o11y/pkg/valuer"
)
type provider struct {
service *alertmanager.Service
config alertmanager.Config
settings factory.ScopedProviderSettings
configStore alertmanagertypes.ConfigStore
stateStore alertmanagertypes.StateStore
notificationManager nfmanager.NotificationManager
maintenanceStore alertmanagertypes.MaintenanceStore
stopC chan struct{}
}
func NewFactory(
sqlstore sqlstore.SQLStore,
orgGetter organization.Getter,
notificationManager nfmanager.NotificationManager,
maintenanceStore alertmanagertypes.MaintenanceStore,
) factory.ProviderFactory[alertmanager.Alertmanager, alertmanager.Config] {
return factory.NewProviderFactory(factory.MustNewName("o11y"), func(ctx context.Context, settings factory.ProviderSettings, config alertmanager.Config) (alertmanager.Alertmanager, error) {
return New(settings, config, sqlstore, orgGetter, notificationManager, maintenanceStore)
})
}
func New(
providerSettings factory.ProviderSettings,
config alertmanager.Config,
sqlstore sqlstore.SQLStore,
orgGetter organization.Getter,
notificationManager nfmanager.NotificationManager,
maintenanceStore alertmanagertypes.MaintenanceStore,
) (*provider, error) {
settings := factory.NewScopedProviderSettings(providerSettings, "github.com/hanzoai/o11y/pkg/alertmanager/o11yalertmanager")
configStore := sqlalertmanagerstore.NewConfigStore(sqlstore)
stateStore := sqlalertmanagerstore.NewStateStore(sqlstore)
p := &provider{
service: alertmanager.New(
settings,
config.O11y.Config,
stateStore,
configStore,
orgGetter,
notificationManager,
maintenanceStore,
),
settings: settings,
config: config,
configStore: configStore,
stateStore: stateStore,
notificationManager: notificationManager,
maintenanceStore: maintenanceStore,
stopC: make(chan struct{}),
}
return p, nil
}
func (provider *provider) Start(ctx context.Context) error {
if err := provider.service.SyncServers(ctx); err != nil {
provider.settings.Logger().ErrorContext(ctx, "failed to sync alertmanager servers", errors.Attr(err))
return err
}
ticker := time.NewTicker(provider.config.O11y.PollInterval)
defer ticker.Stop()
for {
select {
case <-provider.stopC:
return nil
case <-ticker.C:
if err := provider.service.SyncServers(ctx); err != nil {
provider.settings.Logger().ErrorContext(ctx, "failed to sync alertmanager servers", errors.Attr(err))
}
}
}
}
func (provider *provider) Stop(ctx context.Context) error {
close(provider.stopC)
return provider.service.Stop(ctx)
}
func (provider *provider) GetAlerts(ctx context.Context, orgID string, params alertmanagertypes.GettableAlertsParams) (alertmanagertypes.DeprecatedGettableAlerts, error) {
return provider.service.GetAlerts(ctx, orgID, params)
}
func (provider *provider) PutAlerts(ctx context.Context, orgID string, alerts alertmanagertypes.PostableAlerts) error {
return provider.service.PutAlerts(ctx, orgID, alerts)
}
func (provider *provider) TestReceiver(ctx context.Context, orgID string, receiver *alertmanagertypes.Receiver) error {
return provider.service.TestReceiver(ctx, orgID, receiver)
}
func (provider *provider) TestAlert(ctx context.Context, orgID string, ruleID string, receiversMap map[*alertmanagertypes.PostableAlert][]string) error {
config, err := provider.notificationManager.GetNotificationConfig(orgID, ruleID)
if err != nil {
return err
}
if config.UsePolicy {
for alert := range receiversMap {
set := make(model.LabelSet)
for k, v := range alert.Labels {
set[model.LabelName(k)] = model.LabelValue(v)
}
match, err := provider.notificationManager.Match(ctx, orgID, alert.Labels[ruletypes.LabelRuleID], set)
if err != nil {
return err
}
if len(match) == 0 {
delete(receiversMap, alert)
} else {
receiversMap[alert] = match
}
}
}
return provider.service.TestAlert(ctx, orgID, receiversMap, config)
}
func (provider *provider) ListChannels(ctx context.Context, orgID string) ([]*alertmanagertypes.Channel, error) {
return provider.configStore.ListChannels(ctx, orgID)
}
func (provider *provider) ListAllChannels(ctx context.Context) ([]*alertmanagertypes.Channel, error) {
return nil, errors.Newf(errors.TypeUnsupported, errors.CodeUnsupported, "not supported by provider o11y")
}
func (provider *provider) GetChannelByID(ctx context.Context, orgID string, channelID valuer.UUID) (*alertmanagertypes.Channel, error) {
return provider.configStore.GetChannelByID(ctx, orgID, channelID)
}
func (provider *provider) UpdateChannelByReceiverAndID(ctx context.Context, orgID string, receiver *alertmanagertypes.Receiver, id valuer.UUID) error {
channel, err := provider.configStore.GetChannelByID(ctx, orgID, id)
if err != nil {
return err
}
if err := channel.Update(receiver); err != nil {
return err
}
config, err := provider.configStore.Get(ctx, orgID)
if err != nil {
return err
}
if err := config.UpdateReceiver(receiver); err != nil {
return err
}
return provider.configStore.UpdateChannel(ctx, orgID, channel, alertmanagertypes.WithCb(func(ctx context.Context) error {
return provider.configStore.Set(ctx, config)
}))
}
func (provider *provider) DeleteChannelByID(ctx context.Context, orgID string, channelID valuer.UUID) error {
channel, err := provider.configStore.GetChannelByID(ctx, orgID, channelID)
if err != nil {
return err
}
// Check if channel is referenced by any route policy (rule-based or policy-based)
policies, err := provider.notificationManager.GetRoutePoliciesByChannel(ctx, orgID, channel.Name)
if err != nil {
return err
}
if len(policies) > 0 {
names := make([]string, 0, len(policies))
for _, p := range policies {
names = append(names, p.Name)
}
return errors.NewInvalidInputf(errors.CodeInvalidInput,
"channel %q cannot be deleted because it is used by the following routing policies: %v",
channel.Name, names)
}
config, err := provider.configStore.Get(ctx, orgID)
if err != nil {
return err
}
if err := config.DeleteReceiver(channel.Name); err != nil {
return err
}
return provider.configStore.DeleteChannelByID(ctx, orgID, channelID, alertmanagertypes.WithCb(func(ctx context.Context) error {
return provider.configStore.Set(ctx, config)
}))
}
func (provider *provider) CreateChannel(ctx context.Context, orgID string, receiver *alertmanagertypes.Receiver) (*alertmanagertypes.Channel, error) {
config, err := provider.configStore.Get(ctx, orgID)
if err != nil {
return nil, err
}
if err := config.CreateReceiver(receiver); err != nil {
return nil, err
}
channel, err := alertmanagertypes.NewChannelFromReceiver(receiver, orgID)
if err != nil {
return nil, err
}
err = provider.configStore.CreateChannel(ctx, channel, alertmanagertypes.WithCb(func(ctx context.Context) error {
return provider.configStore.Set(ctx, config)
}))
if err != nil {
return nil, err
}
return channel, nil
}
func (provider *provider) Config() alertmanagerserver.Config {
return provider.config.O11y.Config
}
func (provider *provider) SetConfig(ctx context.Context, config *alertmanagertypes.Config) error {
return provider.configStore.Set(ctx, config)
}
func (provider *provider) GetConfig(ctx context.Context, orgID string) (*alertmanagertypes.Config, error) {
return provider.configStore.Get(ctx, orgID)
}
func (provider *provider) SetDefaultConfig(ctx context.Context, orgID string) error {
config, err := alertmanagertypes.NewDefaultConfig(provider.config.O11y.Global, provider.config.O11y.Route, orgID)
if err != nil {
return err
}
return provider.configStore.Set(ctx, config)
}
func (provider *provider) Collect(ctx context.Context, orgID valuer.UUID) (map[string]any, error) {
channels, err := provider.configStore.ListChannels(ctx, orgID.String())
if err != nil {
return nil, err
}
return alertmanagertypes.NewStatsFromChannels(channels), nil
}
func (provider *provider) SetNotificationConfig(ctx context.Context, orgID valuer.UUID, ruleId string, config *alertmanagertypes.NotificationConfig) error {
err := provider.notificationManager.SetNotificationConfig(orgID.StringValue(), ruleId, config)
if err != nil {
return err
}
return nil
}
func (provider *provider) DeleteNotificationConfig(ctx context.Context, orgID valuer.UUID, ruleId string) error {
err := provider.notificationManager.DeleteNotificationConfig(orgID.StringValue(), ruleId)
if err != nil {
return err
}
return nil
}
func (provider *provider) CreateRoutePolicy(ctx context.Context, routeRequest *alertmanagertypes.PostableRoutePolicy) (*alertmanagertypes.GettableRoutePolicy, error) {
claims, err := authtypes.ClaimsFromContext(ctx)
if err != nil {
return nil, err
}
orgID, err := valuer.NewUUID(claims.OrgID)
if err != nil {
return nil, err
}
if err := routeRequest.Validate(); err != nil {
return nil, err
}
route := alertmanagertypes.RoutePolicy{
Expression: routeRequest.Expression,
ExpressionKind: routeRequest.ExpressionKind,
Name: routeRequest.Name,
Description: routeRequest.Description,
Enabled: true,
Tags: routeRequest.Tags,
Channels: routeRequest.Channels,
OrgID: claims.OrgID,
Identifiable: types.Identifiable{
ID: valuer.GenerateUUID(),
},
UserAuditable: types.UserAuditable{
CreatedBy: claims.Email,
UpdatedBy: claims.Email,
},
TimeAuditable: types.TimeAuditable{
CreatedAt: time.Now(),
UpdatedAt: time.Now(),
},
}
err = provider.notificationManager.CreateRoutePolicy(ctx, orgID.String(), &route)
if err != nil {
return nil, err
}
return &alertmanagertypes.GettableRoutePolicy{
PostableRoutePolicy: *routeRequest,
ID: route.ID.StringValue(),
CreatedAt: route.CreatedAt,
UpdatedAt: route.UpdatedAt,
CreatedBy: &route.CreatedBy,
UpdatedBy: &route.UpdatedBy,
}, nil
}
func (provider *provider) CreateRoutePolicies(ctx context.Context, routeRequests []*alertmanagertypes.PostableRoutePolicy) ([]*alertmanagertypes.GettableRoutePolicy, error) {
claims, err := authtypes.ClaimsFromContext(ctx)
if err != nil {
return nil, err
}
orgID, err := valuer.NewUUID(claims.OrgID)
if err != nil {
return nil, err
}
if len(routeRequests) == 0 {
return []*alertmanagertypes.GettableRoutePolicy{}, nil
}
routes := make([]*alertmanagertypes.RoutePolicy, 0, len(routeRequests))
results := make([]*alertmanagertypes.GettableRoutePolicy, 0, len(routeRequests))
for _, routeRequest := range routeRequests {
if err := routeRequest.Validate(); err != nil {
return nil, err
}
route := &alertmanagertypes.RoutePolicy{
Expression: routeRequest.Expression,
ExpressionKind: routeRequest.ExpressionKind,
Name: routeRequest.Name,
Description: routeRequest.Description,
Enabled: true,
Tags: routeRequest.Tags,
Channels: routeRequest.Channels,
OrgID: claims.OrgID,
Identifiable: types.Identifiable{
ID: valuer.GenerateUUID(),
},
UserAuditable: types.UserAuditable{
CreatedBy: claims.Email,
UpdatedBy: claims.Email,
},
TimeAuditable: types.TimeAuditable{
CreatedAt: time.Now(),
UpdatedAt: time.Now(),
},
}
routes = append(routes, route)
results = append(results, &alertmanagertypes.GettableRoutePolicy{
PostableRoutePolicy: *routeRequest,
ID: route.ID.StringValue(),
CreatedAt: route.CreatedAt,
UpdatedAt: route.UpdatedAt,
CreatedBy: &route.CreatedBy,
UpdatedBy: &route.UpdatedBy,
})
}
err = provider.notificationManager.CreateRoutePolicies(ctx, orgID.String(), routes)
if err != nil {
return nil, err
}
return results, nil
}
func (provider *provider) GetRoutePolicyByID(ctx context.Context, routeID string) (*alertmanagertypes.GettableRoutePolicy, error) {
claims, err := authtypes.ClaimsFromContext(ctx)
if err != nil {
return nil, err
}
orgID, err := valuer.NewUUID(claims.OrgID)
if err != nil {
return nil, err
}
route, err := provider.notificationManager.GetRoutePolicyByID(ctx, orgID.String(), routeID)
if err != nil {
return nil, err
}
return &alertmanagertypes.GettableRoutePolicy{
PostableRoutePolicy: alertmanagertypes.PostableRoutePolicy{
Expression: route.Expression,
ExpressionKind: route.ExpressionKind,
Channels: route.Channels,
Name: route.Name,
Description: route.Description,
Tags: route.Tags,
},
ID: route.ID.StringValue(),
CreatedAt: route.CreatedAt,
UpdatedAt: route.UpdatedAt,
CreatedBy: &route.CreatedBy,
UpdatedBy: &route.UpdatedBy,
}, nil
}
func (provider *provider) GetAllRoutePolicies(ctx context.Context) ([]*alertmanagertypes.GettableRoutePolicy, error) {
claims, err := authtypes.ClaimsFromContext(ctx)
if err != nil {
return nil, err
}
orgID, err := valuer.NewUUID(claims.OrgID)
if err != nil {
return nil, err
}
routes, err := provider.notificationManager.GetAllRoutePolicies(ctx, orgID.String())
if err != nil {
return nil, err
}
results := make([]*alertmanagertypes.GettableRoutePolicy, 0, len(routes))
for _, route := range routes {
results = append(results, &alertmanagertypes.GettableRoutePolicy{
PostableRoutePolicy: alertmanagertypes.PostableRoutePolicy{
Expression: route.Expression,
ExpressionKind: route.ExpressionKind,
Channels: route.Channels,
Name: route.Name,
Description: route.Description,
Tags: route.Tags,
},
ID: route.ID.StringValue(),
CreatedAt: route.CreatedAt,
UpdatedAt: route.UpdatedAt,
CreatedBy: &route.CreatedBy,
UpdatedBy: &route.UpdatedBy,
})
}
return results, nil
}
func (provider *provider) UpdateRoutePolicyByID(ctx context.Context, routeID string, route *alertmanagertypes.PostableRoutePolicy) (*alertmanagertypes.GettableRoutePolicy, error) {
claims, err := authtypes.ClaimsFromContext(ctx)
if err != nil {
return nil, errors.NewInvalidInputf(errors.CodeUnauthenticated, "invalid claims: %v", err)
}
orgID, err := valuer.NewUUID(claims.OrgID)
if err != nil {
return nil, err
}
if routeID == "" {
return nil, errors.NewInvalidInputf(errors.CodeInvalidInput, "routeID cannot be empty")
}
if route == nil {
return nil, errors.NewInvalidInputf(errors.CodeInvalidInput, "route cannot be nil")
}
if err := route.Validate(); err != nil {
return nil, errors.NewInvalidInputf(errors.CodeInvalidInput, "invalid route: %v", err)
}
existingRoute, err := provider.notificationManager.GetRoutePolicyByID(ctx, claims.OrgID, routeID)
if err != nil {
return nil, errors.NewInvalidInputf(errors.CodeNotFound, "route not found: %v", err)
}
updatedRoute := &alertmanagertypes.RoutePolicy{
Expression: route.Expression,
ExpressionKind: route.ExpressionKind,
Name: route.Name,
Description: route.Description,
Tags: route.Tags,
Channels: route.Channels,
OrgID: claims.OrgID,
Identifiable: existingRoute.Identifiable,
UserAuditable: types.UserAuditable{
CreatedBy: existingRoute.CreatedBy,
UpdatedBy: claims.Email,
},
TimeAuditable: types.TimeAuditable{
CreatedAt: existingRoute.CreatedAt,
UpdatedAt: time.Now(),
},
}
err = provider.notificationManager.DeleteRoutePolicy(ctx, orgID.String(), routeID)
if err != nil {
return nil, errors.NewInvalidInputf(errors.CodeInternal, "error deleting existing route: %v", err)
}
err = provider.notificationManager.CreateRoutePolicy(ctx, orgID.String(), updatedRoute)
if err != nil {
return nil, err
}
return &alertmanagertypes.GettableRoutePolicy{
PostableRoutePolicy: *route,
ID: updatedRoute.ID.StringValue(),
CreatedAt: updatedRoute.CreatedAt,
UpdatedAt: updatedRoute.UpdatedAt,
CreatedBy: &updatedRoute.CreatedBy,
UpdatedBy: &updatedRoute.UpdatedBy,
}, nil
}
func (provider *provider) DeleteRoutePolicyByID(ctx context.Context, routeID string) error {
claims, err := authtypes.ClaimsFromContext(ctx)
if err != nil {
return errors.NewInvalidInputf(errors.CodeUnauthenticated, "invalid claims: %v", err)
}
orgID, err := valuer.NewUUID(claims.OrgID)
if err != nil {
return err
}
if routeID == "" {
return errors.NewInvalidInputf(errors.CodeInvalidInput, "routeID cannot be empty")
}
return provider.notificationManager.DeleteRoutePolicy(ctx, orgID.String(), routeID)
}
func (provider *provider) CreateInhibitRules(ctx context.Context, orgID valuer.UUID, rules []amConfig.InhibitRule) error {
config, err := provider.configStore.Get(ctx, orgID.String())
if err != nil {
return err
}
if err := config.AddInhibitRules(rules); err != nil {
return err
}
return provider.configStore.Set(ctx, config)
}
func (provider *provider) DeleteAllRoutePoliciesByRuleId(ctx context.Context, names string) error {
claims, err := authtypes.ClaimsFromContext(ctx)
if err != nil {
return errors.NewInvalidInputf(errors.CodeUnauthenticated, "invalid claims: %v", err)
}
orgID, err := valuer.NewUUID(claims.OrgID)
if err != nil {
return err
}
return provider.notificationManager.DeleteAllRoutePoliciesByName(ctx, orgID.String(), names)
}
func (provider *provider) UpdateAllRoutePoliciesByRuleId(ctx context.Context, names string, routes []*alertmanagertypes.PostableRoutePolicy) error {
err := provider.DeleteAllRoutePoliciesByRuleId(ctx, names)
if err != nil {
return errors.NewInvalidInputf(errors.CodeInternal, "error deleting the routes: %v", err)
}
_, err = provider.CreateRoutePolicies(ctx, routes)
return err
}
func (provider *provider) DeleteAllInhibitRulesByRuleId(ctx context.Context, orgID valuer.UUID, ruleId string) error {
config, err := provider.configStore.Get(ctx, orgID.String())
if err != nil {
return err
}
if err := config.DeleteRuleIDInhibitor(ruleId); err != nil {
return err
}
return provider.configStore.Set(ctx, config)
}