open-vault/vault/cluster/cluster.go

368 lines
10 KiB
Go
Raw Normal View History

package cluster
import (
"context"
"crypto/tls"
"crypto/x509"
"errors"
"net"
"sync"
"sync/atomic"
"time"
log "github.com/hashicorp/go-hclog"
"github.com/hashicorp/vault/sdk/helper/consts"
"golang.org/x/net/http2"
)
var (
// Making this a package var allows tests to modify
HeartbeatInterval = 5 * time.Second
)
const (
ListenerAcceptDeadline = 500 * time.Millisecond
)
// Client is used to lookup a client certificate.
type Client interface {
ClientLookup(context.Context, *tls.CertificateRequestInfo) (*tls.Certificate, error)
}
// Handler exposes functions for looking up TLS configuration and handing
// off a connection for a cluster listener application.
type Handler interface {
ServerLookup(context.Context, *tls.ClientHelloInfo) (*tls.Certificate, error)
Raft Storage Backend (#6888) * Work on raft backend * Add logstore locally * Add encryptor and unsealable interfaces * Add clustering support to raft * Remove client and handler * Bootstrap raft on init * Cleanup raft logic a bit * More raft work * Work on TLS config * More work on bootstrapping * Fix build * More work on bootstrapping * More bootstrapping work * fix build * Remove consul dep * Fix build * merged oss/master into raft-storage * Work on bootstrapping * Get bootstrapping to work * Clean up FMS and node-id * Update local node ID logic * Cleanup node-id change * Work on snapshotting * Raft: Add remove peer API (#906) * Add remove peer API * Add some comments * Fix existing snapshotting (#909) * Raft get peers API (#912) * Read raft configuration * address review feedback * Use the Leadership Transfer API to step-down the active node (#918) * Raft join and unseal using Shamir keys (#917) * Raft join using shamir * Store AEAD instead of master key * Split the raft join process to answer the challenge after a successful unseal * get the follower to standby state * Make unseal work * minor changes * Some input checks * reuse the shamir seal access instead of new default seal access * refactor joinRaftSendAnswer function * Synchronously send answer in auto-unseal case * Address review feedback * Raft snapshots (#910) * Fix existing snapshotting * implement the noop snapshotting * Add comments and switch log libraries * add some snapshot tests * add snapshot test file * add TODO * More work on raft snapshotting * progress on the ConfigStore strategy * Don't use two buckets * Update the snapshot store logic to hide the file logic * Add more backend tests * Cleanup code a bit * [WIP] Raft recovery (#938) * Add recovery functionality * remove fmt.Printfs * Fix a few fsm bugs * Add max size value for raft backend (#942) * Add max size value for raft backend * Include physical.ErrValueTooLarge in the message * Raft snapshot Take/Restore API (#926) * Inital work on raft snapshot APIs * Always redirect snapshot install/download requests * More work on the snapshot APIs * Cleanup code a bit * On restore handle special cases * Use the seal to encrypt the sha sum file * Add sealer mechanism and fix some bugs * Call restore while state lock is held * Send restore cb trigger through raft log * Make error messages nicer * Add test helpers * Add snapshot test * Add shamir unseal test * Add more raft snapshot API tests * Fix locking * Change working to initalize * Add underlying raw object to test cluster core * Move leaderUUID to core * Add raft TLS rotation logic (#950) * Add TLS rotation logic * Cleanup logic a bit * Add/Remove from follower state on add/remove peer * add comments * Update more comments * Update request_forwarding_service.proto * Make sure we populate all nodes in the followerstate obj * Update times * Apply review feedback * Add more raft config setting (#947) * Add performance config setting * Add more config options and fix tests * Test Raft Recovery (#944) * Test raft recovery * Leave out a node during recovery * remove unused struct * Update physical/raft/snapshot_test.go * Update physical/raft/snapshot_test.go * fix vendoring * Switch to new raft interface * Remove unused files * Switch a gogo -> proto instance * Remove unneeded vault dep in go.sum * Update helper/testhelpers/testhelpers.go Co-Authored-By: Calvin Leung Huang <cleung2010@gmail.com> * Update vault/cluster/cluster.go * track active key within the keyring itself (#6915) * track active key within the keyring itself * lookup and store using the active key ID * update docstring * minor refactor * Small text fixes (#6912) * Update physical/raft/raft.go Co-Authored-By: Calvin Leung Huang <cleung2010@gmail.com> * review feedback * Move raft logical system into separate file * Update help text a bit * Enforce cluster addr is set and use it for raft bootstrapping * Fix tests * fix http test panic * Pull in latest raft-snapshot library * Add comment
2019-06-20 19:14:58 +00:00
CALookup(context.Context) ([]*x509.Certificate, error)
// Handoff is used to pass the connection lifetime off to
// the handler
Handoff(context.Context, *sync.WaitGroup, chan struct{}, *tls.Conn) error
Stop() error
}
Raft Storage Backend (#6888) * Work on raft backend * Add logstore locally * Add encryptor and unsealable interfaces * Add clustering support to raft * Remove client and handler * Bootstrap raft on init * Cleanup raft logic a bit * More raft work * Work on TLS config * More work on bootstrapping * Fix build * More work on bootstrapping * More bootstrapping work * fix build * Remove consul dep * Fix build * merged oss/master into raft-storage * Work on bootstrapping * Get bootstrapping to work * Clean up FMS and node-id * Update local node ID logic * Cleanup node-id change * Work on snapshotting * Raft: Add remove peer API (#906) * Add remove peer API * Add some comments * Fix existing snapshotting (#909) * Raft get peers API (#912) * Read raft configuration * address review feedback * Use the Leadership Transfer API to step-down the active node (#918) * Raft join and unseal using Shamir keys (#917) * Raft join using shamir * Store AEAD instead of master key * Split the raft join process to answer the challenge after a successful unseal * get the follower to standby state * Make unseal work * minor changes * Some input checks * reuse the shamir seal access instead of new default seal access * refactor joinRaftSendAnswer function * Synchronously send answer in auto-unseal case * Address review feedback * Raft snapshots (#910) * Fix existing snapshotting * implement the noop snapshotting * Add comments and switch log libraries * add some snapshot tests * add snapshot test file * add TODO * More work on raft snapshotting * progress on the ConfigStore strategy * Don't use two buckets * Update the snapshot store logic to hide the file logic * Add more backend tests * Cleanup code a bit * [WIP] Raft recovery (#938) * Add recovery functionality * remove fmt.Printfs * Fix a few fsm bugs * Add max size value for raft backend (#942) * Add max size value for raft backend * Include physical.ErrValueTooLarge in the message * Raft snapshot Take/Restore API (#926) * Inital work on raft snapshot APIs * Always redirect snapshot install/download requests * More work on the snapshot APIs * Cleanup code a bit * On restore handle special cases * Use the seal to encrypt the sha sum file * Add sealer mechanism and fix some bugs * Call restore while state lock is held * Send restore cb trigger through raft log * Make error messages nicer * Add test helpers * Add snapshot test * Add shamir unseal test * Add more raft snapshot API tests * Fix locking * Change working to initalize * Add underlying raw object to test cluster core * Move leaderUUID to core * Add raft TLS rotation logic (#950) * Add TLS rotation logic * Cleanup logic a bit * Add/Remove from follower state on add/remove peer * add comments * Update more comments * Update request_forwarding_service.proto * Make sure we populate all nodes in the followerstate obj * Update times * Apply review feedback * Add more raft config setting (#947) * Add performance config setting * Add more config options and fix tests * Test Raft Recovery (#944) * Test raft recovery * Leave out a node during recovery * remove unused struct * Update physical/raft/snapshot_test.go * Update physical/raft/snapshot_test.go * fix vendoring * Switch to new raft interface * Remove unused files * Switch a gogo -> proto instance * Remove unneeded vault dep in go.sum * Update helper/testhelpers/testhelpers.go Co-Authored-By: Calvin Leung Huang <cleung2010@gmail.com> * Update vault/cluster/cluster.go * track active key within the keyring itself (#6915) * track active key within the keyring itself * lookup and store using the active key ID * update docstring * minor refactor * Small text fixes (#6912) * Update physical/raft/raft.go Co-Authored-By: Calvin Leung Huang <cleung2010@gmail.com> * review feedback * Move raft logical system into separate file * Update help text a bit * Enforce cluster addr is set and use it for raft bootstrapping * Fix tests * fix http test panic * Pull in latest raft-snapshot library * Add comment
2019-06-20 19:14:58 +00:00
type ClusterHook interface {
AddClient(alpn string, client Client)
RemoveClient(alpn string)
AddHandler(alpn string, handler Handler)
StopHandler(alpn string)
TLSConfig(ctx context.Context) (*tls.Config, error)
Addr() net.Addr
}
// Listener is the source of truth for cluster handlers and connection
// clients. It dynamically builds the cluster TLS information. It's also
// responsible for starting tcp listeners and accepting new cluster connections.
type Listener struct {
handlers map[string]Handler
clients map[string]Client
shutdown *uint32
shutdownWg *sync.WaitGroup
server *http2.Server
listenerAddrs []*net.TCPAddr
cipherSuites []uint16
logger log.Logger
l sync.RWMutex
}
func NewListener(addrs []*net.TCPAddr, cipherSuites []uint16, logger log.Logger) *Listener {
// Create the HTTP/2 server that will be shared by both RPC and regular
// duties. Doing it this way instead of listening via the server and gRPC
// allows us to re-use the same port via ALPN. We can just tell the server
// to serve a given conn and which handler to use.
h2Server := &http2.Server{
// Our forwarding connections heartbeat regularly so anything else we
// want to go away/get cleaned up pretty rapidly
IdleTimeout: 5 * HeartbeatInterval,
}
return &Listener{
handlers: make(map[string]Handler),
clients: make(map[string]Client),
shutdown: new(uint32),
shutdownWg: &sync.WaitGroup{},
server: h2Server,
listenerAddrs: addrs,
cipherSuites: cipherSuites,
logger: logger,
}
}
Raft Storage Backend (#6888) * Work on raft backend * Add logstore locally * Add encryptor and unsealable interfaces * Add clustering support to raft * Remove client and handler * Bootstrap raft on init * Cleanup raft logic a bit * More raft work * Work on TLS config * More work on bootstrapping * Fix build * More work on bootstrapping * More bootstrapping work * fix build * Remove consul dep * Fix build * merged oss/master into raft-storage * Work on bootstrapping * Get bootstrapping to work * Clean up FMS and node-id * Update local node ID logic * Cleanup node-id change * Work on snapshotting * Raft: Add remove peer API (#906) * Add remove peer API * Add some comments * Fix existing snapshotting (#909) * Raft get peers API (#912) * Read raft configuration * address review feedback * Use the Leadership Transfer API to step-down the active node (#918) * Raft join and unseal using Shamir keys (#917) * Raft join using shamir * Store AEAD instead of master key * Split the raft join process to answer the challenge after a successful unseal * get the follower to standby state * Make unseal work * minor changes * Some input checks * reuse the shamir seal access instead of new default seal access * refactor joinRaftSendAnswer function * Synchronously send answer in auto-unseal case * Address review feedback * Raft snapshots (#910) * Fix existing snapshotting * implement the noop snapshotting * Add comments and switch log libraries * add some snapshot tests * add snapshot test file * add TODO * More work on raft snapshotting * progress on the ConfigStore strategy * Don't use two buckets * Update the snapshot store logic to hide the file logic * Add more backend tests * Cleanup code a bit * [WIP] Raft recovery (#938) * Add recovery functionality * remove fmt.Printfs * Fix a few fsm bugs * Add max size value for raft backend (#942) * Add max size value for raft backend * Include physical.ErrValueTooLarge in the message * Raft snapshot Take/Restore API (#926) * Inital work on raft snapshot APIs * Always redirect snapshot install/download requests * More work on the snapshot APIs * Cleanup code a bit * On restore handle special cases * Use the seal to encrypt the sha sum file * Add sealer mechanism and fix some bugs * Call restore while state lock is held * Send restore cb trigger through raft log * Make error messages nicer * Add test helpers * Add snapshot test * Add shamir unseal test * Add more raft snapshot API tests * Fix locking * Change working to initalize * Add underlying raw object to test cluster core * Move leaderUUID to core * Add raft TLS rotation logic (#950) * Add TLS rotation logic * Cleanup logic a bit * Add/Remove from follower state on add/remove peer * add comments * Update more comments * Update request_forwarding_service.proto * Make sure we populate all nodes in the followerstate obj * Update times * Apply review feedback * Add more raft config setting (#947) * Add performance config setting * Add more config options and fix tests * Test Raft Recovery (#944) * Test raft recovery * Leave out a node during recovery * remove unused struct * Update physical/raft/snapshot_test.go * Update physical/raft/snapshot_test.go * fix vendoring * Switch to new raft interface * Remove unused files * Switch a gogo -> proto instance * Remove unneeded vault dep in go.sum * Update helper/testhelpers/testhelpers.go Co-Authored-By: Calvin Leung Huang <cleung2010@gmail.com> * Update vault/cluster/cluster.go * track active key within the keyring itself (#6915) * track active key within the keyring itself * lookup and store using the active key ID * update docstring * minor refactor * Small text fixes (#6912) * Update physical/raft/raft.go Co-Authored-By: Calvin Leung Huang <cleung2010@gmail.com> * review feedback * Move raft logical system into separate file * Update help text a bit * Enforce cluster addr is set and use it for raft bootstrapping * Fix tests * fix http test panic * Pull in latest raft-snapshot library * Add comment
2019-06-20 19:14:58 +00:00
// TODO: This probably isn't correct
func (cl *Listener) Addr() net.Addr {
return cl.listenerAddrs[0]
}
func (cl *Listener) Addrs() []*net.TCPAddr {
return cl.listenerAddrs
}
// AddClient adds a new client for an ALPN name
func (cl *Listener) AddClient(alpn string, client Client) {
cl.l.Lock()
cl.clients[alpn] = client
cl.l.Unlock()
}
// RemoveClient removes the client for the specified ALPN name
func (cl *Listener) RemoveClient(alpn string) {
cl.l.Lock()
delete(cl.clients, alpn)
cl.l.Unlock()
}
// AddHandler registers a new cluster handler for the provided ALPN name.
func (cl *Listener) AddHandler(alpn string, handler Handler) {
cl.l.Lock()
cl.handlers[alpn] = handler
cl.l.Unlock()
}
// StopHandler stops the cluster handler for the provided ALPN name, it also
// calls stop on the handler.
func (cl *Listener) StopHandler(alpn string) {
cl.l.Lock()
handler, ok := cl.handlers[alpn]
delete(cl.handlers, alpn)
cl.l.Unlock()
if ok {
handler.Stop()
}
}
2019-04-17 21:21:56 +00:00
// Handler returns the handler for the provided ALPN name
func (cl *Listener) Handler(alpn string) (Handler, bool) {
2019-07-09 07:46:08 +00:00
cl.l.RLock()
2019-04-17 21:21:56 +00:00
handler, ok := cl.handlers[alpn]
2019-07-09 07:46:08 +00:00
cl.l.RUnlock()
2019-04-17 21:21:56 +00:00
return handler, ok
}
// Server returns the http2 server that the cluster listener is using
func (cl *Listener) Server() *http2.Server {
return cl.server
}
// TLSConfig returns a tls config object that uses dynamic lookups to correctly
// authenticate registered handlers/clients
func (cl *Listener) TLSConfig(ctx context.Context) (*tls.Config, error) {
serverLookup := func(clientHello *tls.ClientHelloInfo) (*tls.Certificate, error) {
cl.logger.Debug("performing server cert lookup")
cl.l.RLock()
defer cl.l.RUnlock()
for _, v := range clientHello.SupportedProtos {
if handler, ok := cl.handlers[v]; ok {
return handler.ServerLookup(ctx, clientHello)
}
}
cl.logger.Warn("no TLS certs found for ALPN", "ALPN", clientHello.SupportedProtos)
return nil, errors.New("unsupported protocol")
}
clientLookup := func(requestInfo *tls.CertificateRequestInfo) (*tls.Certificate, error) {
cl.logger.Debug("performing client cert lookup")
cl.l.RLock()
defer cl.l.RUnlock()
for _, client := range cl.clients {
cert, err := client.ClientLookup(ctx, requestInfo)
if err == nil && cert != nil {
return cert, nil
}
}
cl.logger.Warn("no client information found")
return nil, errors.New("no client cert found")
}
serverConfigLookup := func(clientHello *tls.ClientHelloInfo) (*tls.Config, error) {
caPool := x509.NewCertPool()
ret := &tls.Config{
ClientAuth: tls.RequireAndVerifyClientCert,
GetCertificate: serverLookup,
GetClientCertificate: clientLookup,
MinVersion: tls.VersionTLS12,
RootCAs: caPool,
ClientCAs: caPool,
NextProtos: clientHello.SupportedProtos,
CipherSuites: cl.cipherSuites,
}
cl.l.RLock()
defer cl.l.RUnlock()
for _, v := range clientHello.SupportedProtos {
if handler, ok := cl.handlers[v]; ok {
Raft Storage Backend (#6888) * Work on raft backend * Add logstore locally * Add encryptor and unsealable interfaces * Add clustering support to raft * Remove client and handler * Bootstrap raft on init * Cleanup raft logic a bit * More raft work * Work on TLS config * More work on bootstrapping * Fix build * More work on bootstrapping * More bootstrapping work * fix build * Remove consul dep * Fix build * merged oss/master into raft-storage * Work on bootstrapping * Get bootstrapping to work * Clean up FMS and node-id * Update local node ID logic * Cleanup node-id change * Work on snapshotting * Raft: Add remove peer API (#906) * Add remove peer API * Add some comments * Fix existing snapshotting (#909) * Raft get peers API (#912) * Read raft configuration * address review feedback * Use the Leadership Transfer API to step-down the active node (#918) * Raft join and unseal using Shamir keys (#917) * Raft join using shamir * Store AEAD instead of master key * Split the raft join process to answer the challenge after a successful unseal * get the follower to standby state * Make unseal work * minor changes * Some input checks * reuse the shamir seal access instead of new default seal access * refactor joinRaftSendAnswer function * Synchronously send answer in auto-unseal case * Address review feedback * Raft snapshots (#910) * Fix existing snapshotting * implement the noop snapshotting * Add comments and switch log libraries * add some snapshot tests * add snapshot test file * add TODO * More work on raft snapshotting * progress on the ConfigStore strategy * Don't use two buckets * Update the snapshot store logic to hide the file logic * Add more backend tests * Cleanup code a bit * [WIP] Raft recovery (#938) * Add recovery functionality * remove fmt.Printfs * Fix a few fsm bugs * Add max size value for raft backend (#942) * Add max size value for raft backend * Include physical.ErrValueTooLarge in the message * Raft snapshot Take/Restore API (#926) * Inital work on raft snapshot APIs * Always redirect snapshot install/download requests * More work on the snapshot APIs * Cleanup code a bit * On restore handle special cases * Use the seal to encrypt the sha sum file * Add sealer mechanism and fix some bugs * Call restore while state lock is held * Send restore cb trigger through raft log * Make error messages nicer * Add test helpers * Add snapshot test * Add shamir unseal test * Add more raft snapshot API tests * Fix locking * Change working to initalize * Add underlying raw object to test cluster core * Move leaderUUID to core * Add raft TLS rotation logic (#950) * Add TLS rotation logic * Cleanup logic a bit * Add/Remove from follower state on add/remove peer * add comments * Update more comments * Update request_forwarding_service.proto * Make sure we populate all nodes in the followerstate obj * Update times * Apply review feedback * Add more raft config setting (#947) * Add performance config setting * Add more config options and fix tests * Test Raft Recovery (#944) * Test raft recovery * Leave out a node during recovery * remove unused struct * Update physical/raft/snapshot_test.go * Update physical/raft/snapshot_test.go * fix vendoring * Switch to new raft interface * Remove unused files * Switch a gogo -> proto instance * Remove unneeded vault dep in go.sum * Update helper/testhelpers/testhelpers.go Co-Authored-By: Calvin Leung Huang <cleung2010@gmail.com> * Update vault/cluster/cluster.go * track active key within the keyring itself (#6915) * track active key within the keyring itself * lookup and store using the active key ID * update docstring * minor refactor * Small text fixes (#6912) * Update physical/raft/raft.go Co-Authored-By: Calvin Leung Huang <cleung2010@gmail.com> * review feedback * Move raft logical system into separate file * Update help text a bit * Enforce cluster addr is set and use it for raft bootstrapping * Fix tests * fix http test panic * Pull in latest raft-snapshot library * Add comment
2019-06-20 19:14:58 +00:00
caList, err := handler.CALookup(ctx)
if err != nil {
return nil, err
}
Raft Storage Backend (#6888) * Work on raft backend * Add logstore locally * Add encryptor and unsealable interfaces * Add clustering support to raft * Remove client and handler * Bootstrap raft on init * Cleanup raft logic a bit * More raft work * Work on TLS config * More work on bootstrapping * Fix build * More work on bootstrapping * More bootstrapping work * fix build * Remove consul dep * Fix build * merged oss/master into raft-storage * Work on bootstrapping * Get bootstrapping to work * Clean up FMS and node-id * Update local node ID logic * Cleanup node-id change * Work on snapshotting * Raft: Add remove peer API (#906) * Add remove peer API * Add some comments * Fix existing snapshotting (#909) * Raft get peers API (#912) * Read raft configuration * address review feedback * Use the Leadership Transfer API to step-down the active node (#918) * Raft join and unseal using Shamir keys (#917) * Raft join using shamir * Store AEAD instead of master key * Split the raft join process to answer the challenge after a successful unseal * get the follower to standby state * Make unseal work * minor changes * Some input checks * reuse the shamir seal access instead of new default seal access * refactor joinRaftSendAnswer function * Synchronously send answer in auto-unseal case * Address review feedback * Raft snapshots (#910) * Fix existing snapshotting * implement the noop snapshotting * Add comments and switch log libraries * add some snapshot tests * add snapshot test file * add TODO * More work on raft snapshotting * progress on the ConfigStore strategy * Don't use two buckets * Update the snapshot store logic to hide the file logic * Add more backend tests * Cleanup code a bit * [WIP] Raft recovery (#938) * Add recovery functionality * remove fmt.Printfs * Fix a few fsm bugs * Add max size value for raft backend (#942) * Add max size value for raft backend * Include physical.ErrValueTooLarge in the message * Raft snapshot Take/Restore API (#926) * Inital work on raft snapshot APIs * Always redirect snapshot install/download requests * More work on the snapshot APIs * Cleanup code a bit * On restore handle special cases * Use the seal to encrypt the sha sum file * Add sealer mechanism and fix some bugs * Call restore while state lock is held * Send restore cb trigger through raft log * Make error messages nicer * Add test helpers * Add snapshot test * Add shamir unseal test * Add more raft snapshot API tests * Fix locking * Change working to initalize * Add underlying raw object to test cluster core * Move leaderUUID to core * Add raft TLS rotation logic (#950) * Add TLS rotation logic * Cleanup logic a bit * Add/Remove from follower state on add/remove peer * add comments * Update more comments * Update request_forwarding_service.proto * Make sure we populate all nodes in the followerstate obj * Update times * Apply review feedback * Add more raft config setting (#947) * Add performance config setting * Add more config options and fix tests * Test Raft Recovery (#944) * Test raft recovery * Leave out a node during recovery * remove unused struct * Update physical/raft/snapshot_test.go * Update physical/raft/snapshot_test.go * fix vendoring * Switch to new raft interface * Remove unused files * Switch a gogo -> proto instance * Remove unneeded vault dep in go.sum * Update helper/testhelpers/testhelpers.go Co-Authored-By: Calvin Leung Huang <cleung2010@gmail.com> * Update vault/cluster/cluster.go * track active key within the keyring itself (#6915) * track active key within the keyring itself * lookup and store using the active key ID * update docstring * minor refactor * Small text fixes (#6912) * Update physical/raft/raft.go Co-Authored-By: Calvin Leung Huang <cleung2010@gmail.com> * review feedback * Move raft logical system into separate file * Update help text a bit * Enforce cluster addr is set and use it for raft bootstrapping * Fix tests * fix http test panic * Pull in latest raft-snapshot library * Add comment
2019-06-20 19:14:58 +00:00
for _, ca := range caList {
caPool.AddCert(ca)
}
return ret, nil
}
}
cl.logger.Warn("no TLS config found for ALPN", "ALPN", clientHello.SupportedProtos)
return nil, errors.New("unsupported protocol")
}
return &tls.Config{
ClientAuth: tls.RequireAndVerifyClientCert,
GetCertificate: serverLookup,
GetClientCertificate: clientLookup,
GetConfigForClient: serverConfigLookup,
MinVersion: tls.VersionTLS12,
CipherSuites: cl.cipherSuites,
}, nil
}
// Run starts the tcp listeners and will accept connections until stop is
Raft Storage Backend (#6888) * Work on raft backend * Add logstore locally * Add encryptor and unsealable interfaces * Add clustering support to raft * Remove client and handler * Bootstrap raft on init * Cleanup raft logic a bit * More raft work * Work on TLS config * More work on bootstrapping * Fix build * More work on bootstrapping * More bootstrapping work * fix build * Remove consul dep * Fix build * merged oss/master into raft-storage * Work on bootstrapping * Get bootstrapping to work * Clean up FMS and node-id * Update local node ID logic * Cleanup node-id change * Work on snapshotting * Raft: Add remove peer API (#906) * Add remove peer API * Add some comments * Fix existing snapshotting (#909) * Raft get peers API (#912) * Read raft configuration * address review feedback * Use the Leadership Transfer API to step-down the active node (#918) * Raft join and unseal using Shamir keys (#917) * Raft join using shamir * Store AEAD instead of master key * Split the raft join process to answer the challenge after a successful unseal * get the follower to standby state * Make unseal work * minor changes * Some input checks * reuse the shamir seal access instead of new default seal access * refactor joinRaftSendAnswer function * Synchronously send answer in auto-unseal case * Address review feedback * Raft snapshots (#910) * Fix existing snapshotting * implement the noop snapshotting * Add comments and switch log libraries * add some snapshot tests * add snapshot test file * add TODO * More work on raft snapshotting * progress on the ConfigStore strategy * Don't use two buckets * Update the snapshot store logic to hide the file logic * Add more backend tests * Cleanup code a bit * [WIP] Raft recovery (#938) * Add recovery functionality * remove fmt.Printfs * Fix a few fsm bugs * Add max size value for raft backend (#942) * Add max size value for raft backend * Include physical.ErrValueTooLarge in the message * Raft snapshot Take/Restore API (#926) * Inital work on raft snapshot APIs * Always redirect snapshot install/download requests * More work on the snapshot APIs * Cleanup code a bit * On restore handle special cases * Use the seal to encrypt the sha sum file * Add sealer mechanism and fix some bugs * Call restore while state lock is held * Send restore cb trigger through raft log * Make error messages nicer * Add test helpers * Add snapshot test * Add shamir unseal test * Add more raft snapshot API tests * Fix locking * Change working to initalize * Add underlying raw object to test cluster core * Move leaderUUID to core * Add raft TLS rotation logic (#950) * Add TLS rotation logic * Cleanup logic a bit * Add/Remove from follower state on add/remove peer * add comments * Update more comments * Update request_forwarding_service.proto * Make sure we populate all nodes in the followerstate obj * Update times * Apply review feedback * Add more raft config setting (#947) * Add performance config setting * Add more config options and fix tests * Test Raft Recovery (#944) * Test raft recovery * Leave out a node during recovery * remove unused struct * Update physical/raft/snapshot_test.go * Update physical/raft/snapshot_test.go * fix vendoring * Switch to new raft interface * Remove unused files * Switch a gogo -> proto instance * Remove unneeded vault dep in go.sum * Update helper/testhelpers/testhelpers.go Co-Authored-By: Calvin Leung Huang <cleung2010@gmail.com> * Update vault/cluster/cluster.go * track active key within the keyring itself (#6915) * track active key within the keyring itself * lookup and store using the active key ID * update docstring * minor refactor * Small text fixes (#6912) * Update physical/raft/raft.go Co-Authored-By: Calvin Leung Huang <cleung2010@gmail.com> * review feedback * Move raft logical system into separate file * Update help text a bit * Enforce cluster addr is set and use it for raft bootstrapping * Fix tests * fix http test panic * Pull in latest raft-snapshot library * Add comment
2019-06-20 19:14:58 +00:00
// called. This function blocks so should be called in a goroutine.
func (cl *Listener) Run(ctx context.Context) error {
// Get our TLS config
tlsConfig, err := cl.TLSConfig(ctx)
if err != nil {
cl.logger.Error("failed to get tls configuration when starting cluster listener", "error", err)
return err
}
// The server supports all of the possible protos
tlsConfig.NextProtos = []string{"h2", consts.RequestForwardingALPN, consts.PerfStandbyALPN, consts.PerformanceReplicationALPN, consts.DRReplicationALPN}
for i, laddr := range cl.listenerAddrs {
// closeCh is used to shutdown the spawned goroutines once this
// function returns
closeCh := make(chan struct{})
if cl.logger.IsInfo() {
cl.logger.Info("starting listener", "listener_address", laddr)
}
// Create a TCP listener. We do this separately and specifically
// with TCP so that we can set deadlines.
tcpLn, err := net.ListenTCP("tcp", laddr)
if err != nil {
cl.logger.Error("error starting listener", "error", err)
continue
}
if laddr.String() != tcpLn.Addr().String() {
// If we listened on port 0, record the port the OS gave us.
cl.listenerAddrs[i] = tcpLn.Addr().(*net.TCPAddr)
}
// Wrap the listener with TLS
tlsLn := tls.NewListener(tcpLn, tlsConfig)
if cl.logger.IsInfo() {
cl.logger.Info("serving cluster requests", "cluster_listen_address", tlsLn.Addr())
}
cl.shutdownWg.Add(1)
// Start our listening loop
go func(closeCh chan struct{}, tlsLn net.Listener) {
defer func() {
cl.shutdownWg.Done()
tlsLn.Close()
close(closeCh)
}()
for {
if atomic.LoadUint32(cl.shutdown) > 0 {
return
}
// Set the deadline for the accept call. If it passes we'll get
// an error, causing us to check the condition at the top
// again.
tcpLn.SetDeadline(time.Now().Add(ListenerAcceptDeadline))
// Accept the connection
conn, err := tlsLn.Accept()
if err != nil {
if err, ok := err.(net.Error); ok && !err.Timeout() {
cl.logger.Debug("non-timeout error accepting on cluster port", "error", err)
}
if conn != nil {
conn.Close()
}
continue
}
if conn == nil {
continue
}
// Type assert to TLS connection and handshake to populate the
// connection state
tlsConn := conn.(*tls.Conn)
// Set a deadline for the handshake. This will cause clients
// that don't successfully auth to be kicked out quickly.
// Cluster connections should be reliable so being marginally
// aggressive here is fine.
err = tlsConn.SetDeadline(time.Now().Add(30 * time.Second))
if err != nil {
if cl.logger.IsDebug() {
cl.logger.Debug("error setting deadline for cluster connection", "error", err)
}
tlsConn.Close()
continue
}
err = tlsConn.Handshake()
if err != nil {
if cl.logger.IsDebug() {
cl.logger.Debug("error handshaking cluster connection", "error", err)
}
tlsConn.Close()
continue
}
// Now, set it back to unlimited
err = tlsConn.SetDeadline(time.Time{})
if err != nil {
if cl.logger.IsDebug() {
cl.logger.Debug("error setting deadline for cluster connection", "error", err)
}
tlsConn.Close()
continue
}
cl.l.RLock()
handler, ok := cl.handlers[tlsConn.ConnectionState().NegotiatedProtocol]
cl.l.RUnlock()
if !ok {
cl.logger.Debug("unknown negotiated protocol on cluster port")
tlsConn.Close()
continue
}
if err := handler.Handoff(ctx, cl.shutdownWg, closeCh, tlsConn); err != nil {
cl.logger.Error("error handling cluster connection", "error", err)
continue
}
}
}(closeCh, tlsLn)
}
return nil
}
// Stop stops the cluster listner
func (cl *Listener) Stop() {
// Set the shutdown flag. This will cause the listeners to shut down
// within the deadline in clusterListenerAcceptDeadline
atomic.StoreUint32(cl.shutdown, 1)
cl.logger.Info("forwarding rpc listeners stopped")
// Wait for them all to shut down
cl.shutdownWg.Wait()
cl.logger.Info("rpc listeners successfully shut down")
}