Documentation ¶
Overview ¶
Package suture provides Erlang-like supervisor trees.
This implements Erlang-esque supervisor trees, as adapted for Go. This is an industrial-strength, tested library deployed into hostile environments, not just a proof of concept or a toy.
Supervisor Tree -> SuTree -> suture -> holds your code together when it's trying to fall apart.
Why use Suture?
- You want to write bullet-resistant services that will remain available despite unforeseen failure.
- You need the code to be smart enough not to consume 100% of the CPU restarting things.
- You want to easily compose multiple such services in one program.
- You want the Erlang programmers to stop lording their supervision trees over you.
Suture has 100% test coverage, and is golint clean. This doesn't prove it free of bugs, but it shows I care.
A blog post describing the design decisions is available at http://www.jerf.org/iri/post/2930 .
Using Suture ¶
To idiomatically use Suture, create a Supervisor which is your top level "application" supervisor. This will often occur in your program's "main" function.
Create "Service"s, which implement the Service interface. .Add() them to your Supervisor. Supervisors are also services, so you can create a tree structure here, depending on the exact combination of restarts you want to create.
As a special case, when adding Supervisors to Supervisors, the "sub" supervisor will have the "super" supervisor's Log function copied. This allows you to set one log function on the "top" supervisor, and have it propagate down to all the sub-supervisors. This also allows libraries or modules to provide Supervisors without having to commit their users to a particular logging method.
Finally, as what is probably the last line of your main() function, call .Serve() on your top level supervisor. This will start all the services you've defined.
See the Example for an example, using a simple service that serves out incrementing integers.
Index ¶
- Variables
- type DefaultJitter
- type DeprecatedService
- type Event
- type EventBackoff
- type EventHook
- type EventResume
- type EventServicePanic
- type EventServiceTerminate
- type EventStopTimeout
- type EventType
- type HasSupervisor
- type Jitter
- type NoJitter
- type Service
- type ServiceToken
- type Spec
- type SprintFunc
- type Supervisor
- func (s *Supervisor) Add(service Service) ServiceToken
- func (s *Supervisor) GetSupervisor() *Supervisor
- func (s *Supervisor) Remove(id ServiceToken) error
- func (s *Supervisor) RemoveAndWait(id ServiceToken, timeout time.Duration) error
- func (s *Supervisor) Serve(ctx context.Context) error
- func (s *Supervisor) ServeBackground(ctx context.Context) <-chan error
- func (s *Supervisor) Services() []Service
- func (s *Supervisor) String() string
- func (s *Supervisor) UnstoppedServiceReport() (UnstoppedServiceReport, error)
- type UnstoppedService
- type UnstoppedServiceReport
Examples ¶
Constants ¶
This section is empty.
Variables ¶
var ErrDoNotRestart = errors.New("service should not be restarted")
ErrDoNotRestart can be returned by a service to voluntarily not be restarted. Any error that will compare with errors.Is as being this error will count as an ErrDoNotRestart.
var ErrSupervisorNotRunning = errors.New("supervisor not running")
ErrSupervisorNotRunning is returned by some methods if the supervisor is not running, either because it has not been started or because it has been terminated.
var ErrSupervisorNotStarted = errors.New("supervisor not started yet")
ErrSupervisorNotStarted is returned if you try to send control messages to a supervisor that has not started yet. See note on Supervisor struct about the legal ways to start a supervisor.
var ErrSupervisorNotTerminated = errors.New("supervisor not terminated")
ErrSupervisorNotTerminated is returned when asking for a stopped service report before the supervisor has been terminated.
var ErrTerminateSupervisorTree = errors.New("tree should be terminated")
ErrTerminateSupervisorTree can can be returned by a service to terminate the entire supervision tree above it as well. Any error that will compare with errors.Is to be ErrTerminateSupervisorTree will count as an ErrTerminateSupervisorTree.
var ErrTimeout = errors.New("waiting for service to stop has timed out")
ErrTimeout is returned when an attempt to RemoveAndWait for a service to stop has timed out.
var ErrWrongSupervisor = errors.New("wrong supervisor for this service token, no service removed")
ErrWrongSupervisor is returned by the (*Supervisor).Remove method if you pass a ServiceToken from the wrong Supervisor.
Functions ¶
This section is empty.
Types ¶
type DefaultJitter ¶
type DefaultJitter struct {
// contains filtered or unexported fields
}
DefaultJitter is the jitter function that is applied when spec.BackoffJitter is set to nil.
type DeprecatedService ¶
type DeprecatedService interface { Serve() Stop() }
type Event ¶
Event defines the interface implemented by all events Suture will generate.
Map will return a map with the details of the event serialized into a map[string]interface{}, with only the values suitable for serialization.
type EventBackoff ¶
type EventBackoff struct { Supervisor *Supervisor `json:"-"` SupervisorName string `json:"supervisor_name"` }
func (EventBackoff) Map ¶
func (e EventBackoff) Map() map[string]interface{}
func (EventBackoff) String ¶
func (e EventBackoff) String() string
func (EventBackoff) Type ¶
func (e EventBackoff) Type() EventType
type EventResume ¶
type EventResume struct { Supervisor *Supervisor `json:"-"` SupervisorName string `json:"supervisor_name"` }
func (EventResume) Map ¶
func (e EventResume) Map() map[string]interface{}
func (EventResume) String ¶
func (e EventResume) String() string
func (EventResume) Type ¶
func (e EventResume) Type() EventType
type EventServicePanic ¶
type EventServicePanic struct { Supervisor *Supervisor `json:"-"` SupervisorName string `json:"supervisor_name"` Service Service `json:"-"` ServiceName string `json:"service_name"` CurrentFailures float64 `json:"current_failures"` FailureThreshold float64 `json:"failure_threshold"` Restarting bool `json:"restarting"` PanicMsg string `json:"panic_msg"` Stacktrace string `json:"stacktrace"` }
func (EventServicePanic) Map ¶
func (e EventServicePanic) Map() map[string]interface{}
func (EventServicePanic) String ¶
func (e EventServicePanic) String() string
func (EventServicePanic) Type ¶
func (e EventServicePanic) Type() EventType
type EventServiceTerminate ¶
type EventServiceTerminate struct { Supervisor *Supervisor `json:"-"` SupervisorName string `json:"supervisor_name"` Service Service `json:"-"` ServiceName string `json:"service_name"` CurrentFailures float64 `json:"current_failures"` FailureThreshold float64 `json:"failure_threshold"` Restarting bool `json:"restarting"` Err interface{} `json:"error_msg"` }
func (EventServiceTerminate) Map ¶
func (e EventServiceTerminate) Map() map[string]interface{}
func (EventServiceTerminate) String ¶
func (e EventServiceTerminate) String() string
func (EventServiceTerminate) Type ¶
func (e EventServiceTerminate) Type() EventType
type EventStopTimeout ¶
type EventStopTimeout struct { Supervisor *Supervisor `json:"-"` SupervisorName string `json:"supervisor_name"` Service Service `json:"-"` ServiceName string `json:"service"` }
func (EventStopTimeout) Map ¶
func (e EventStopTimeout) Map() map[string]interface{}
func (EventStopTimeout) String ¶
func (e EventStopTimeout) String() string
func (EventStopTimeout) Type ¶
func (e EventStopTimeout) Type() EventType
type HasSupervisor ¶
type HasSupervisor interface {
GetSupervisor() *Supervisor
}
HasSupervisor is an interface that indicates the given struct contains a supervisor. If the struct is either already a *Supervisor, or it embeds a *Supervisor, this will already be implemented for you. Otherwise, a struct containing a supervisor will need to implement this in order to participate in the log function propagation and recursive UnstoppedService report.
It is legal for GetSupervisor to return nil, in which case the supervisor-specific behaviors will simply be ignored.
type Jitter ¶
Jitter returns the sum of the input duration and a random jitter. It is compatible with the jitter functions in github.com/lthibault/jitterbug.
type Service ¶
Service is the interface that describes a service to a Supervisor.
Serve Method ¶
The Serve method is called by a Supervisor to start the service. The service should execute within the goroutine that this is called in, that is, it should not spawn a "worker" goroutine. If this function either returns error or panics, the Supervisor will call it again.
A Serve method SHOULD do as much cleanup of the state as possible, to prevent any corruption in the previous state from crashing the service again. The beginning of a service with persistent state should generally be a few lines to initialize and clean up that state.
The error returned by the service, if any, will be part of the log message generated for it. There are two distinguished errors a Service can return:
ErrDoNotRestart indicates that the service should not be restarted and removed from the supervisor entirely.
ErrTerminateTree indicates that the Supervisor the service is running in should be terminated. If that Supervisor recursively returns that, its parent supervisor will also be terminated. (This can be controlled with configuration in the Supervisor.)
In Go 1.13 and greater, this is checked via errors.Is, so the error can be further wrapped with whatever additional info you like. Prior to Go 1.13, it will be checked via directly equality check, so the distinguished errors cannot be wrapped.
Once the service has been instructed to stop, the Service SHOULD NOT be reused in any other supervisor! Because of the impossibility of guaranteeing that the service has fully stopped in Go, you can't prove that you won't be starting two goroutines using the exact same memory to store state, causing completely unpredictable behavior.
Serve should not return until the service has actually stopped. "Stopped" here is defined as "the service will stop servicing any further requests in the future". Any mandatory cleanup related to the Service should also have been performed.
If a service does not stop within the supervisor's timeout duration, the supervisor will log an entry to that effect. This does not guarantee that the service is hung; it may still get around to being properly stopped in the future. Until the service is fully stopped, both the service and the spawned goroutine trying to stop it will be "leaked".
Stringer Interface ¶
When a Service is added to a Supervisor, the Supervisor will create a string representation of that service used for logging.
If you implement the fmt.Stringer interface, that will be used.
If you do not implement the fmt.Stringer interface, a default fmt.Sprintf("%#v") will be used.
func AsService ¶
func AsService(service DeprecatedService) Service
AsService converts old-style suture service to a new style suture service.
type ServiceToken ¶
type ServiceToken struct {
// contains filtered or unexported fields
}
ServiceToken is an opaque identifier that can be used to terminate a service that has been Add()ed to a Supervisor.
type Spec ¶
type Spec struct { EventHook EventHook Sprint SprintFunc FailureDecay float64 FailureThreshold float64 FailureBackoff time.Duration BackoffJitter Jitter Timeout time.Duration PassThroughPanics bool DontPropagateTermination bool }
Spec is used to pass arguments to the New function to create a supervisor. See the New function for full documentation.
type SprintFunc ¶
type SprintFunc func(interface{}) string
SprintFunc formats an arbitrary Go value into a string. It is used by the supervisor to format the value of a call to recover().
type Supervisor ¶
type Supervisor struct { Name string // contains filtered or unexported fields }
Supervisor is the core type of the module that represents a Supervisor.
Supervisors should be constructed either by New or NewSimple.
Once constructed, a Supervisor should be started in one of three ways:
- Calling .Serve(ctx).
- Calling .ServeBackground(ctx).
- Adding it to an existing Supervisor.
Calling Serve will cause the supervisor to run until the passed-in context is cancelled. Often one of the last lines of the "main" func for a program will be to call one of the Serve methods.
Calling ServeBackground will CORRECTLY start the supervisor running in a new goroutine. It is risky to directly run
go supervisor.Serve()
because that will briefly create a race condition as it starts up, if you try to .Add() services immediately afterward.
func New ¶
func New(name string, spec Spec) *Supervisor
New is the full constructor function for a supervisor.
The name is a friendly human name for the supervisor, used in logging. Suture does not care if this is unique, but it is good for your sanity if it is.
If not set, the following values are used:
- EventHook: A function is created that uses log.Print.
- FailureDecay: 30 seconds
- FailureThreshold: 5 failures
- FailureBackoff: 15 seconds
- Timeout: 10 seconds
- BackoffJitter: DefaultJitter
The EventHook function will be called when errors occur. Suture will log the following:
- When a service has failed, with a descriptive message about the current backoff status, and whether it was immediately restarted
- When the supervisor has gone into its backoff mode, and when it exits it
- When a service fails to stop
The failureRate, failureThreshold, and failureBackoff controls how failures are handled, in order to avoid the supervisor failure case where the program does nothing but restarting failed services. If you do not care how failures behave, the default values should be fine for the vast majority of services, but if you want the details:
The supervisor tracks the number of failures that have occurred, with an exponential decay on the count. Every FailureDecay seconds, the number of failures that have occurred is cut in half. (This is done smoothly with an exponential function.) When a failure occurs, the number of failures is incremented by one. When the number of failures passes the FailureThreshold, the entire service waits for FailureBackoff seconds before attempting any further restarts, at which point it resets its failure count to zero.
Timeout is how long Suture will wait for a service to properly terminate.
The PassThroughPanics options can be set to let panics in services propagate and crash the program, should this be desirable.
DontPropagateTermination indicates whether this supervisor tree will propagate a ErrTerminateTree if a child process returns it. If false, this supervisor will itself return an error that will terminate its parent. If true, it will merely return ErrDoNotRestart. false by default.
Example (Simple) ¶
package main import ( "context" "fmt" ) type Incrementor struct { current int next chan int stop chan bool } func (i *Incrementor) Stop() { fmt.Println("Stopping the service") i.stop <- true } func (i *Incrementor) Serve(ctx context.Context) error { for { select { case i.next <- i.current: i.current++ case <-ctx.Done(): // This message on i.stop is just to synchronize // this test with the example code so the output is // consistent for the test code; most services // would just "return nil" here. fmt.Println("Stopping the service") i.stop <- true return nil } } } func main() { supervisor := NewSimple("Supervisor") service := &Incrementor{0, make(chan int), make(chan bool)} supervisor.Add(service) ctx, cancel := context.WithCancel(context.Background()) supervisor.ServeBackground(ctx) fmt.Println("Got:", <-service.next) fmt.Println("Got:", <-service.next) cancel() // We sync here just to guarantee the output of "Stopping the service" <-service.stop }
Output: Got: 0 Got: 1 Stopping the service
func NewSimple ¶
func NewSimple(name string) *Supervisor
NewSimple is a convenience function to create a service with just a name and the sensible defaults.
func (*Supervisor) Add ¶
func (s *Supervisor) Add(service Service) ServiceToken
Add adds a service to this supervisor.
If the supervisor is currently running, the service will be started immediately. If the supervisor has not been started yet, the service will be started when the supervisor is. If the supervisor was already stopped, this is a no-op returning an empty service-token.
The returned ServiceID may be passed to the Remove method of the Supervisor to terminate the service.
As a special behavior, if the service added is itself a supervisor, the supervisor being added will copy the EventHook function from the Supervisor it is being added to. This allows factoring out providing a Supervisor from its logging. This unconditionally overwrites the child Supervisor's logging functions.
func (*Supervisor) GetSupervisor ¶
func (s *Supervisor) GetSupervisor() *Supervisor
func (*Supervisor) Remove ¶
func (s *Supervisor) Remove(id ServiceToken) error
Remove will remove the given service from the Supervisor, and attempt to Stop() it. The ServiceID token comes from the Add() call. This returns without waiting for the service to stop.
func (*Supervisor) RemoveAndWait ¶
func (s *Supervisor) RemoveAndWait(id ServiceToken, timeout time.Duration) error
RemoveAndWait will remove the given service from the Supervisor and attempt to Stop() it. It will wait up to the given timeout value for the service to terminate. A timeout value of 0 means to wait forever.
If a nil error is returned from this function, then the service was terminated normally. If either the supervisor terminates or the timeout passes, ErrTimeout is returned. (If this isn't even the right supervisor ErrWrongSupervisor is returned.)
func (*Supervisor) Serve ¶
func (s *Supervisor) Serve(ctx context.Context) error
Serve starts the supervisor. You should call this on the top-level supervisor, but nothing else.
func (*Supervisor) ServeBackground ¶
func (s *Supervisor) ServeBackground(ctx context.Context) <-chan error
ServeBackground starts running a supervisor in its own goroutine. When this method returns, the supervisor is guaranteed to be in a running state. The returned one-buffered channel receives the error returned by .Serve.
func (*Supervisor) Services ¶
func (s *Supervisor) Services() []Service
Services returns a []Service containing a snapshot of the services this Supervisor is managing.
func (*Supervisor) String ¶
func (s *Supervisor) String() string
String implements the fmt.Stringer interface.
func (*Supervisor) UnstoppedServiceReport ¶
func (s *Supervisor) UnstoppedServiceReport() (UnstoppedServiceReport, error)
UnstoppedServiceReport will return a report of what services failed to stop when the supervisor was stopped. This call will return when the supervisor is done shutting down. It will hang on a supervisor that has not been stopped, because it will not be "done shutting down".
Calling this on a supervisor will return a report for the whole supervisor tree under it.
WARNING: Technically, any use of the returned data structure is a TOCTOU violation: https://en.wikipedia.org/wiki/Time-of-check_to_time-of-use Since the data structure was generated and returned to you, any of these services may have stopped since then.
However, this can still be useful information at program teardown time. For instance, logging that a service failed to stop as expected is still useful, as even if it shuts down later, it was still later than you expected.
But if you cast the Service objects back to their underlying objects and start trying to manipulate them ("shut down harder!"), be sure to account for the possibility they are in fact shut down before you get them.
If there are no services to report, the UnstoppedServiceReport will be nil. A zero-length constructed slice is never returned.
type UnstoppedService ¶
type UnstoppedService struct { SupervisorPath []*Supervisor Service Service Name string ServiceToken ServiceToken }
An UnstoppedService is the component member of an UnstoppedServiceReport.
The SupervisorPath is the path down the supervisor tree to the given service.
type UnstoppedServiceReport ¶
type UnstoppedServiceReport []UnstoppedService
An UnstoppedServiceReport will be returned by StopWithReport, reporting which services failed to stop.