minio/cmd/lock-rest-server.go
Klaus Post 51aa59a737
perf: websocket grid connectivity for all internode communication (#18461)
This PR adds a WebSocket grid feature that allows servers to communicate via 
a single two-way connection.

There are two request types:

* Single requests, which are `[]byte => ([]byte, error)`. This is for efficient small
  roundtrips with small payloads.

* Streaming requests which are `[]byte, chan []byte => chan []byte (and error)`,
  which allows for different combinations of full two-way streams with an initial payload.

Only a single stream is created between two machines - and there is, as such, no
server/client relation since both sides can initiate and handle requests. Which server
initiates the request is decided deterministically on the server names.

Requests are made through a mux client and server, which handles message
passing, congestion, cancelation, timeouts, etc.

If a connection is lost, all requests are canceled, and the calling server will try
to reconnect. Registered handlers can operate directly on byte 
slices or use a higher-level generics abstraction.

There is no versioning of handlers/clients, and incompatible changes should
be handled by adding new handlers.

The request path can be changed to a new one for any protocol changes.

First, all servers create a "Manager." The manager must know its address 
as well as all remote addresses. This will manage all connections.
To get a connection to any remote, ask the manager to provide it given
the remote address using.

```
func (m *Manager) Connection(host string) *Connection
```

All serverside handlers must also be registered on the manager. This will
make sure that all incoming requests are served. The number of in-flight 
requests and responses must also be given for streaming requests.

The "Connection" returned manages the mux-clients. Requests issued
to the connection will be sent to the remote.

* `func (c *Connection) Request(ctx context.Context, h HandlerID, req []byte) ([]byte, error)`
   performs a single request and returns the result. Any deadline provided on the request is
   forwarded to the server, and canceling the context will make the function return at once.

* `func (c *Connection) NewStream(ctx context.Context, h HandlerID, payload []byte) (st *Stream, err error)`
   will initiate a remote call and send the initial payload.

```Go
// A Stream is a two-way stream.
// All responses *must* be read by the caller.
// If the call is canceled through the context,
//The appropriate error will be returned.
type Stream struct {
	// Responses from the remote server.
	// Channel will be closed after an error or when the remote closes.
	// All responses *must* be read by the caller until either an error is returned or the channel is closed.
	// Canceling the context will cause the context cancellation error to be returned.
	Responses <-chan Response

	// Requests sent to the server.
	// If the handler is defined with 0 incoming capacity this will be nil.
	// Channel *must* be closed to signal the end of the stream.
	// If the request context is canceled, the stream will no longer process requests.
	Requests chan<- []byte
}

type Response struct {
	Msg []byte
	Err error
}
```

There are generic versions of the server/client handlers that allow the use of type
safe implementations for data types that support msgpack marshal/unmarshal.
2023-11-20 17:09:35 -08:00

181 lines
5.9 KiB
Go

// Copyright (c) 2015-2021 MinIO, Inc.
//
// This file is part of MinIO Object Storage stack
//
// This program is free software: you can redistribute it and/or modify
// it under the terms of the GNU Affero General Public License as published by
// the Free Software Foundation, either version 3 of the License, or
// (at your option) any later version.
//
// This program is distributed in the hope that it will be useful
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU Affero General Public License for more details.
//
// You should have received a copy of the GNU Affero General Public License
// along with this program. If not, see <http://www.gnu.org/licenses/>.
package cmd
import (
"context"
"time"
"github.com/minio/minio/internal/dsync"
"github.com/minio/minio/internal/grid"
"github.com/minio/minio/internal/logger"
)
// To abstract a node over network.
type lockRESTServer struct {
ll *localLocker
}
// RefreshHandler - refresh the current lock
func (l *lockRESTServer) RefreshHandler(args *dsync.LockArgs) (*dsync.LockResp, *grid.RemoteErr) {
resp := lockRPCRefresh.NewResponse()
refreshed, err := l.ll.Refresh(context.Background(), *args)
if err != nil {
return l.makeResp(resp, err)
}
if !refreshed {
return l.makeResp(resp, errLockNotFound)
}
return l.makeResp(resp, err)
}
// LockHandler - Acquires a lock.
func (l *lockRESTServer) LockHandler(args *dsync.LockArgs) (*dsync.LockResp, *grid.RemoteErr) {
resp := lockRPCLock.NewResponse()
success, err := l.ll.Lock(context.Background(), *args)
if err == nil && !success {
return l.makeResp(resp, errLockConflict)
}
return l.makeResp(resp, err)
}
// UnlockHandler - releases the acquired lock.
func (l *lockRESTServer) UnlockHandler(args *dsync.LockArgs) (*dsync.LockResp, *grid.RemoteErr) {
resp := lockRPCUnlock.NewResponse()
_, err := l.ll.Unlock(context.Background(), *args)
// Ignore the Unlock() "reply" return value because if err == nil, "reply" is always true
// Consequently, if err != nil, reply is always false
return l.makeResp(resp, err)
}
// RLockHandler - Acquires an RLock.
func (l *lockRESTServer) RLockHandler(args *dsync.LockArgs) (*dsync.LockResp, *grid.RemoteErr) {
resp := lockRPCRLock.NewResponse()
success, err := l.ll.RLock(context.Background(), *args)
if err == nil && !success {
err = errLockConflict
}
return l.makeResp(resp, err)
}
// RUnlockHandler - releases the acquired read lock.
func (l *lockRESTServer) RUnlockHandler(args *dsync.LockArgs) (*dsync.LockResp, *grid.RemoteErr) {
resp := lockRPCRUnlock.NewResponse()
// Ignore the RUnlock() "reply" return value because if err == nil, "reply" is always true.
// Consequently, if err != nil, reply is always false
_, err := l.ll.RUnlock(context.Background(), *args)
return l.makeResp(resp, err)
}
// ForceUnlockHandler - query expired lock status.
func (l *lockRESTServer) ForceUnlockHandler(args *dsync.LockArgs) (*dsync.LockResp, *grid.RemoteErr) {
resp := lockRPCForceUnlock.NewResponse()
_, err := l.ll.ForceUnlock(context.Background(), *args)
return l.makeResp(resp, err)
}
var (
// Static lock handlers.
// All have the same signature.
lockRPCForceUnlock = newLockHandler(grid.HandlerLockForceUnlock)
lockRPCRefresh = newLockHandler(grid.HandlerLockRefresh)
lockRPCLock = newLockHandler(grid.HandlerLockLock)
lockRPCUnlock = newLockHandler(grid.HandlerLockUnlock)
lockRPCRLock = newLockHandler(grid.HandlerLockRLock)
lockRPCRUnlock = newLockHandler(grid.HandlerLockRUnlock)
)
func newLockHandler(h grid.HandlerID) *grid.SingleHandler[*dsync.LockArgs, *dsync.LockResp] {
return grid.NewSingleHandler[*dsync.LockArgs, *dsync.LockResp](h, func() *dsync.LockArgs {
return &dsync.LockArgs{}
}, func() *dsync.LockResp {
return &dsync.LockResp{}
})
}
// registerLockRESTHandlers - register lock rest router.
func registerLockRESTHandlers() {
lockServer := &lockRESTServer{
ll: newLocker(),
}
logger.FatalIf(lockRPCForceUnlock.Register(globalGrid.Load(), lockServer.ForceUnlockHandler), "unable to register handler")
logger.FatalIf(lockRPCRefresh.Register(globalGrid.Load(), lockServer.RefreshHandler), "unable to register handler")
logger.FatalIf(lockRPCLock.Register(globalGrid.Load(), lockServer.LockHandler), "unable to register handler")
logger.FatalIf(lockRPCUnlock.Register(globalGrid.Load(), lockServer.UnlockHandler), "unable to register handler")
logger.FatalIf(lockRPCRLock.Register(globalGrid.Load(), lockServer.RLockHandler), "unable to register handler")
logger.FatalIf(lockRPCRUnlock.Register(globalGrid.Load(), lockServer.RUnlockHandler), "unable to register handler")
globalLockServer = lockServer.ll
go lockMaintenance(GlobalContext)
}
func (l *lockRESTServer) makeResp(dst *dsync.LockResp, err error) (*dsync.LockResp, *grid.RemoteErr) {
*dst = dsync.LockResp{Code: dsync.RespOK}
switch err {
case nil:
case errLockNotInitialized:
dst.Code = dsync.RespLockNotInitialized
case errLockConflict:
dst.Code = dsync.RespLockConflict
case errLockNotFound:
dst.Code = dsync.RespLockNotFound
default:
dst.Code = dsync.RespErr
dst.Err = err.Error()
}
return dst, nil
}
const (
// Lock maintenance interval.
lockMaintenanceInterval = 1 * time.Minute
// Lock validity duration
lockValidityDuration = 1 * time.Minute
)
// lockMaintenance loops over all locks and discards locks
// that have not been refreshed for some time.
func lockMaintenance(ctx context.Context) {
if !globalIsDistErasure {
return
}
// Initialize a new ticker with 1 minute between each ticks.
lkTimer := time.NewTimer(lockMaintenanceInterval)
// Stop the timer upon returning.
defer lkTimer.Stop()
for {
// Verifies every minute for locks held more than 2 minutes.
select {
case <-ctx.Done():
return
case <-lkTimer.C:
globalLockServer.expireOldLocks(lockValidityDuration)
// Reset the timer for next cycle.
lkTimer.Reset(lockMaintenanceInterval)
}
}
}