098cbc9c67
all of the packages and tests work with both grpc and drpc. we'll probably need to do some jenkins pipelines to run the tests with drpc as well. most of the changes are really due to a bit of cleanup of the pkg/transport.Client api into an rpc.Dialer in the spirit of a net.Dialer. now that we don't need observers, we can pass around stateless configuration to everything rather than stateful things that issue observations. it also adds a DialAddressID for the case where we don't have a pb.Node, but we do have an address and want to assert some ID. this happened pretty frequently, and now there's no more weird contortions creating custom tls options, etc. a lot of the other changes are being consistent/using the abstractions in the rpc package to do rpc style things like finding peer information, or checking status codes. Change-Id: Ief62875e21d80a21b3c56a5a37f45887679f9412
117 lines
2.7 KiB
Go
117 lines
2.7 KiB
Go
// Copyright (C) 2019 Storj Labs, Inc.
|
|
// See LICENSE for copying information.
|
|
|
|
package contact
|
|
|
|
import (
|
|
"context"
|
|
"math/rand"
|
|
"time"
|
|
|
|
"github.com/zeebo/errs"
|
|
"go.uber.org/zap"
|
|
"golang.org/x/sync/errgroup"
|
|
|
|
"storj.io/storj/internal/sync2"
|
|
"storj.io/storj/pkg/pb"
|
|
"storj.io/storj/pkg/rpc"
|
|
"storj.io/storj/storagenode/trust"
|
|
)
|
|
|
|
// Chore is the contact chore for nodes announcing themselves to their trusted satellites
|
|
//
|
|
// architecture: Chore
|
|
type Chore struct {
|
|
log *zap.Logger
|
|
service *Service
|
|
dialer rpc.Dialer
|
|
|
|
trust *trust.Pool
|
|
|
|
maxSleep time.Duration
|
|
Loop *sync2.Cycle
|
|
}
|
|
|
|
// NewChore creates a new contact chore
|
|
func NewChore(log *zap.Logger, interval time.Duration, maxSleep time.Duration, trust *trust.Pool, dialer rpc.Dialer, service *Service) *Chore {
|
|
return &Chore{
|
|
log: log,
|
|
service: service,
|
|
dialer: dialer,
|
|
|
|
trust: trust,
|
|
|
|
maxSleep: maxSleep,
|
|
Loop: sync2.NewCycle(interval),
|
|
}
|
|
}
|
|
|
|
// Run the contact chore on a regular interval with jitter
|
|
func (chore *Chore) Run(ctx context.Context) (err error) {
|
|
defer mon.Task()(&ctx)(&err)
|
|
chore.log.Info("Storagenode contact chore starting up")
|
|
|
|
return chore.Loop.Run(ctx, func(ctx context.Context) error {
|
|
if err := chore.randomDurationSleep(ctx); err != nil {
|
|
return err
|
|
}
|
|
if err := chore.pingSatellites(ctx); err != nil {
|
|
chore.log.Error("pingSatellites failed", zap.Error(err))
|
|
}
|
|
return nil
|
|
})
|
|
}
|
|
|
|
func (chore *Chore) pingSatellites(ctx context.Context) (err error) {
|
|
defer mon.Task()(&ctx)(&err)
|
|
var group errgroup.Group
|
|
self := chore.service.Local()
|
|
satellites := chore.trust.GetSatellites(ctx)
|
|
for _, satellite := range satellites {
|
|
satellite := satellite
|
|
addr, err := chore.trust.GetAddress(ctx, satellite)
|
|
if err != nil {
|
|
chore.log.Error("getting satellite address", zap.Error(err))
|
|
continue
|
|
}
|
|
group.Go(func() error {
|
|
conn, err := chore.dialer.DialAddressID(ctx, addr, satellite)
|
|
if err != nil {
|
|
return err
|
|
}
|
|
defer func() { err = errs.Combine(err, conn.Close()) }()
|
|
|
|
_, err = conn.NodeClient().CheckIn(ctx, &pb.CheckInRequest{
|
|
Address: self.Address.GetAddress(),
|
|
Version: &self.Version,
|
|
Capacity: &self.Capacity,
|
|
Operator: &self.Operator,
|
|
})
|
|
|
|
return err
|
|
})
|
|
}
|
|
|
|
return group.Wait()
|
|
}
|
|
|
|
// randomDurationSleep sleeps for random interval in [0;maxSleep)
|
|
// returns error if context was cancelled
|
|
func (chore *Chore) randomDurationSleep(ctx context.Context) error {
|
|
if chore.maxSleep <= 0 {
|
|
return nil
|
|
}
|
|
jitter := time.Duration(rand.Int63n(int64(chore.maxSleep)))
|
|
if !sync2.Sleep(ctx, jitter) {
|
|
return ctx.Err()
|
|
}
|
|
|
|
return nil
|
|
}
|
|
|
|
// Close stops the contact chore
|
|
func (chore *Chore) Close() error {
|
|
chore.Loop.Close()
|
|
return nil
|
|
}
|