Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fix pagerduty2 integration #1905

Merged
merged 1 commit into from
May 1, 2018
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions CHANGELOG.md
Original file line number Diff line number Diff line change
Expand Up @@ -25,6 +25,7 @@ alert will auto-recover.
- [#1795](https://github.com/influxdata/kapacitor/pull/1795): Support PagerDuty API v2
- [#1842](https://github.com/influxdata/kapacitor/pull/1842): Add alert inhibitors that allow an alert to supress events from other matching alerts.
- [#1776](https://github.com/influxdata/kapacitor/issues/1776): Fix bug where you could not delete a topic handler with the same name as its topic.
- [#1905](https://github.com/influxdata/kapacitor/pull/1905): Adjust PagerDuty v2 service-test names and capture detailed error messages.

## v1.4.1 [2018-03-13]

Expand Down
2 changes: 1 addition & 1 deletion alert.go
Original file line number Diff line number Diff line change
Expand Up @@ -209,7 +209,7 @@ func newAlertNode(et *ExecutingTask, n *pipeline.AlertNode, d NodeDiagnostic) (a

for _, pd := range n.PagerDuty2Handlers {
c := pagerduty2.HandlerConfig{
ServiceKey: pd.ServiceKey,
RoutingKey: pd.ServiceKey,
}
h := et.tm.PagerDuty2Service.Handler(c, ctx...)
an.handlers = append(an.handlers, h)
Expand Down
2 changes: 1 addition & 1 deletion integrations/streamer_test.go
Original file line number Diff line number Diff line change
Expand Up @@ -9295,7 +9295,7 @@ stream
c := pagerduty2.NewConfig()
c.Enabled = true
c.URL = ts.URL
c.ServiceKey = "service_key"
c.RoutingKey = "service_key"
pd := pagerduty2.NewService(c, diagService.NewPagerDuty2Handler())
pd.HTTPDService = tm.HTTPDService
tm.PagerDuty2Service = pd
Expand Down
30 changes: 15 additions & 15 deletions server/server_test.go
Original file line number Diff line number Diff line change
Expand Up @@ -7442,7 +7442,7 @@ func TestServer_UpdateConfig(t *testing.T) {
{
section: "pagerduty2",
setDefaults: func(c *server.Config) {
c.PagerDuty2.ServiceKey = "secret"
c.PagerDuty2.RoutingKey = "secret"
},
expDefaultSection: client.ConfigSection{
Link: client.Link{Relation: client.Self, Href: "/kapacitor/v1/config/pagerduty2"},
Expand All @@ -7451,11 +7451,11 @@ func TestServer_UpdateConfig(t *testing.T) {
Options: map[string]interface{}{
"enabled": false,
"global": false,
"service-key": true,
"routing-key": true,
"url": pagerduty2.DefaultPagerDuty2APIURL,
},
Redacted: []string{
"service-key",
"routing-key",
},
}},
},
Expand All @@ -7464,18 +7464,18 @@ func TestServer_UpdateConfig(t *testing.T) {
Options: map[string]interface{}{
"enabled": false,
"global": false,
"service-key": true,
"routing-key": true,
"url": pagerduty2.DefaultPagerDuty2APIURL,
},
Redacted: []string{
"service-key",
"routing-key",
},
},
updates: []updateAction{
{
updateAction: client.ConfigUpdateAction{
Set: map[string]interface{}{
"service-key": "",
"routing-key": "",
"enabled": true,
},
},
Expand All @@ -7486,11 +7486,11 @@ func TestServer_UpdateConfig(t *testing.T) {
Options: map[string]interface{}{
"enabled": true,
"global": false,
"service-key": false,
"routing-key": false,
"url": pagerduty2.DefaultPagerDuty2APIURL,
},
Redacted: []string{
"service-key",
"routing-key",
},
}},
},
Expand All @@ -7499,11 +7499,11 @@ func TestServer_UpdateConfig(t *testing.T) {
Options: map[string]interface{}{
"enabled": true,
"global": false,
"service-key": false,
"routing-key": false,
"url": pagerduty2.DefaultPagerDuty2APIURL,
},
Redacted: []string{
"service-key",
"routing-key",
},
},
},
Expand Down Expand Up @@ -8779,9 +8779,9 @@ func TestServer_ListServiceTests(t *testing.T) {
Link: client.Link{Relation: client.Self, Href: "/kapacitor/v1/service-tests/pagerduty2"},
Name: "pagerduty2",
Options: client.ServiceTestOptions{
"incident-key": "testIncidentKey",
"description": "test pagerduty2 message",
"level": "CRITICAL",
"alert_id": "testAlertID",
"description": "test pagerduty2 message",
"level": "CRITICAL",
"event_data": map[string]interface{}{
"Fields": map[string]interface{}{},
"Result": map[string]interface{}{
Expand Down Expand Up @@ -9855,7 +9855,7 @@ func TestServer_AlertHandlers(t *testing.T) {
handler: client.TopicHandler{
Kind: "pagerduty2",
Options: map[string]interface{}{
"service-key": "service_key",
"routing-key": "rkey",
},
},
setup: func(c *server.Config, ha *client.TopicHandler) (context.Context, error) {
Expand Down Expand Up @@ -9898,7 +9898,7 @@ func TestServer_AlertHandlers(t *testing.T) {
},
Timestamp: "1970-01-01T00:00:00.000000000Z",
},
RoutingKey: "service_key",
RoutingKey: "rkey",
},
}}

Expand Down
4 changes: 2 additions & 2 deletions services/pagerduty2/config.go
Original file line number Diff line number Diff line change
Expand Up @@ -15,8 +15,8 @@ type Config struct {
Enabled bool `toml:"enabled" override:"enabled"`
// The PagerDuty API URL, should not need to be changed.
URL string `toml:"url" override:"url"`
// The PagerDuty service key.
ServiceKey string `toml:"service-key" override:"service-key,redact"`
// The PagerDuty routing key, this is associated with an Event v2 API integration service.
RoutingKey string `toml:"routing-key" override:"routing-key,redact"`
// Whether every alert should automatically go to PagerDuty
Global bool `toml:"global" override:"global"`
}
Expand Down
59 changes: 25 additions & 34 deletions services/pagerduty2/service.go
Original file line number Diff line number Diff line change
Expand Up @@ -155,12 +155,11 @@ func (s *Service) Global() bool {
}

type testOptions struct {
IncidentKey string `json:"incident-key"`
Description string `json:"description"`
// Details string `json:"details"`
Level alert.Level `json:"level"`
Data alert.EventData `json:"event_data"`
Timestamp time.Time `json:"timestamp"`
AlertID string `json:"alert_id"`
Description string `json:"description"`
Level alert.Level `json:"level"`
Data alert.EventData `json:"event_data"`
Timestamp time.Time `json:"timestamp"`
}

// TestOptions returns optional values for the test harness
Expand All @@ -170,11 +169,10 @@ func (s *Service) TestOptions() interface{} {
t, _ := time.Parse(layout, str)

return &testOptions{
IncidentKey: "testIncidentKey",
AlertID: "testAlertID",
Description: "test pagerduty2 message",
Level: alert.Critical,
Timestamp: t,
// Details: html.EscapeString(`{"Test": "test_value"}`),
Data: alert.EventData{
Name: "testPagerDuty2",
Tags: make(map[string]string),
Expand All @@ -192,8 +190,8 @@ func (s *Service) Test(options interface{}) error {
}
c := s.config()
return s.Alert(
c.ServiceKey,
o.IncidentKey,
c.RoutingKey,
o.AlertID,
o.Description,
o.Level,
o.Timestamp,
Expand All @@ -205,8 +203,8 @@ func (s *Service) Test(options interface{}) error {
//
// The req headers are now required with the API v2:
// https://v2.developer.pagerduty.com/docs/migrating-to-api-v2
func (s *Service) Alert(serviceKey, incidentKey, desc string, level alert.Level, timestamp time.Time, data alert.EventData) error {
url, post, err := s.preparePost(serviceKey, incidentKey, desc, level, timestamp, data)
func (s *Service) Alert(serviceKey, alertID, desc string, level alert.Level, timestamp time.Time, data alert.EventData) error {
url, post, err := s.preparePost(serviceKey, alertID, desc, level, timestamp, data)
if err != nil {
return err
}
Expand All @@ -231,18 +229,20 @@ func (s *Service) Alert(serviceKey, incidentKey, desc string, level alert.Level,
return err
}
type response struct {
Message string `json:"message"`
Status string `json:"status"`
Message string `json:"message"`
Errors []string `json:"errors"`
}
r := &response{Message: fmt.Sprintf("failed to understand PagerDuty2 response. code: %d content: %s", resp.StatusCode, string(body))}
b := bytes.NewReader(body)
dec := json.NewDecoder(b)
dec.Decode(r)
return errors.New(r.Message)
return fmt.Errorf("Status: %s, Message: %s Errors: %v", r.Status, r.Message, r.Errors)
}
return nil
}

func (s *Service) sendResolve(c Config, serviceKey, incidentKey string) (string, io.Reader, error) {
func (s *Service) sendResolve(c Config, serviceKey, alertID string) (string, io.Reader, error) {
// create a new AlertPayload for us to fire off
type Resolve struct {
RoutingKey string `json:"routing_key"`
Expand All @@ -253,12 +253,12 @@ func (s *Service) sendResolve(c Config, serviceKey, incidentKey string) (string,
ap := Resolve{}

if serviceKey == "" {
ap.RoutingKey = c.ServiceKey
ap.RoutingKey = c.RoutingKey
} else {
ap.RoutingKey = serviceKey
}

ap.DedupKey = incidentKey
ap.DedupKey = alertID
ap.EventAction = "resolve"

var post bytes.Buffer
Expand All @@ -272,7 +272,7 @@ func (s *Service) sendResolve(c Config, serviceKey, incidentKey string) (string,
}

// preparePost is a helper method that sets up the payload for transmission to PagerDuty
func (s *Service) preparePost(serviceKey, incidentKey, desc string, level alert.Level, timestamp time.Time, data alert.EventData) (string, io.Reader, error) {
func (s *Service) preparePost(serviceKey, alertID, desc string, level alert.Level, timestamp time.Time, data alert.EventData) (string, io.Reader, error) {
c := s.config()
if !c.Enabled {
return "", nil, errors.New("service is not enabled")
Expand All @@ -290,7 +290,7 @@ func (s *Service) preparePost(serviceKey, incidentKey, desc string, level alert.
severity = "info"
default:
// default is a 'resolve' function
return s.sendResolve(c, serviceKey, incidentKey)
return s.sendResolve(c, serviceKey, alertID)
}

// create a new AlertPayload for us to fire off
Expand All @@ -299,33 +299,24 @@ func (s *Service) preparePost(serviceKey, incidentKey, desc string, level alert.
}

if serviceKey == "" {
ap.RoutingKey = c.ServiceKey
ap.RoutingKey = c.RoutingKey
} else {
ap.RoutingKey = serviceKey
}

ap.Client = "kapacitor"
ap.ClientURL = s.HTTPDService.URL()
ap.DedupKey = incidentKey
ap.DedupKey = alertID
ap.EventAction = eventType

ap.Payload.CustomDetails = make(map[string]interface{})
ap.Payload.CustomDetails["result"] = data.Result

// The API doesn't explicitly mention a requirement for nanosecond resolution but payloads seem to
// fail if we don't include it (even zeroes). This hack is not graceful, but adds a negligible
// nanosecond resolution to our timestamp
m := timestamp.Format("2006-01-02T15:04:05.000000000Z07:00")
// m := timestamp.Format(time.RFC3339Nano)
// if timestamp.Nanosecond() == 0 {
// m = time.Unix(timestamp.Unix(), 1).In(timestamp.Location()).Format(time.RFC3339Nano)
// }

ap.Payload.Class = data.TaskName
ap.Payload.Severity = severity
ap.Payload.Source = "unknown"
ap.Payload.Summary = desc
ap.Payload.Timestamp = m
ap.Payload.Timestamp = timestamp.Format("2006-01-02T15:04:05.000000000Z07:00")

if _, ok := data.Tags["host"]; ok {
ap.Payload.Source = data.Tags["host"]
Expand All @@ -344,9 +335,9 @@ func (s *Service) preparePost(serviceKey, incidentKey, desc string, level alert.

// HandlerConfig defines the high-level struct required to connect to PagerDuty
type HandlerConfig struct {
// The service key to use for the alert.
// The routing key to use for the alert.
// Defaults to the value in the configuration if empty.
ServiceKey string `mapstructure:"service-key"`
RoutingKey string `mapstructure:"routing-key"`
}

type handler struct {
Expand All @@ -367,7 +358,7 @@ func (s *Service) Handler(c HandlerConfig, ctx ...keyvalue.T) alert.Handler {
// Handle is a bound method to the handler that processes a given alert
func (h *handler) Handle(event alert.Event) {
if err := h.s.Alert(
h.c.ServiceKey,
h.c.RoutingKey,
event.State.ID,
event.State.Message,
event.State.Level,
Expand Down