fix data race that may cause panic and concurrency request limit bug (#1219)

Signed-off-by: crazycs520 <crazycs520@gmail.com>
This commit is contained in:
crazycs 2024-04-02 11:26:21 +08:00 committed by GitHub
parent 356eb45c5d
commit 125a140034
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 54 additions and 16 deletions

View File

@ -588,7 +588,7 @@ func (c *batchCommandsClient) send(forwardedHost string, req *tikvpb.BatchComman
zap.String("forwardedHost", forwardedHost),
zap.Error(err),
)
c.failPendingRequests(err)
c.failRequestsByIDs(err, req.RequestIds) // fast fail requests.
return
}
@ -604,23 +604,50 @@ func (c *batchCommandsClient) send(forwardedHost string, req *tikvpb.BatchComman
zap.Uint64s("requestIDs", req.RequestIds),
zap.Error(err),
)
c.failPendingRequests(err)
c.failRequestsByIDs(err, req.RequestIds) // fast fail requests.
}
}
// `failPendingRequests` must be called in locked contexts in order to avoid double closing channels.
func (c *batchCommandsClient) failPendingRequests(err error) {
// when enable-forwarding is true, the `forwardedHost` maybe not empty.
// failPendingRequests fails all pending requests which req.forwardedHost equals to forwardedHost parameter.
// Why need check `forwardedHost`? Here is an example, when enable-forwarding is true, and this client has network issue with store1:
// - some requests are sent to store1 with forwarding, such as forwardedHost is store2, those requests will succeed.
// - some requests are sent to store1 without forwarding, and may fail then `failPendingRequests` would be called,
// if we don't check `forwardedHost` and fail all pending requests, the requests with forwarding will be failed too. this may cause some issue:
// 1. data race. see https://github.com/tikv/client-go/issues/1222 and TestRandomRestartStoreAndForwarding.
// 2. panic which cause by `send on closed channel`, since failPendingRequests will close the entry.res channel,
// but in another batchRecvLoop goroutine, it may receive the response from forwardedHost store2 and try to send the response to entry.res channel,
// then panic by send on closed channel.
func (c *batchCommandsClient) failPendingRequests(err error, forwardedHost string) {
util.EvalFailpoint("panicInFailPendingRequests")
c.batched.Range(func(key, value interface{}) bool {
id, _ := key.(uint64)
entry, _ := value.(*batchCommandsEntry)
c.batched.Delete(id)
c.sent.Add(-1)
entry.error(err)
if entry.forwardedHost == forwardedHost {
c.failRequest(err, id, entry)
}
return true
})
}
// failRequestsByIDs fails requests by requestID.
func (c *batchCommandsClient) failRequestsByIDs(err error, requestIDs []uint64) {
for _, requestID := range requestIDs {
value, ok := c.batched.Load(requestID)
if !ok {
continue
}
c.failRequest(err, requestID, value.(*batchCommandsEntry))
}
}
func (c *batchCommandsClient) failRequest(err error, requestID uint64, entry *batchCommandsEntry) {
c.batched.Delete(requestID)
c.sent.Add(-1)
entry.error(err)
}
func (c *batchCommandsClient) waitConnReady() (err error) {
state := c.conn.GetState()
if state == connectivity.Ready {
@ -793,7 +820,7 @@ func (c *batchCommandsClient) recreateStreamingClient(err error, streamClient *b
}
*epoch++
c.failPendingRequests(err) // fail all pending requests.
c.failPendingRequests(err, streamClient.forwardedHost) // fail all pending requests.
b := retry.NewBackofferWithVars(context.Background(), math.MaxInt32, nil)
for { // try to re-create the streaming in the loop.
if c.isStopped() {

View File

@ -59,7 +59,6 @@ import (
"github.com/tikv/client-go/v2/internal/client/mockserver"
"github.com/tikv/client-go/v2/internal/logutil"
"github.com/tikv/client-go/v2/tikvrpc"
"github.com/tikv/client-go/v2/util/israce"
"go.uber.org/zap"
"google.golang.org/grpc/connectivity"
"google.golang.org/grpc/metadata"
@ -888,9 +887,6 @@ func TestBatchClientReceiveHealthFeedback(t *testing.T) {
}
func TestRandomRestartStoreAndForwarding(t *testing.T) {
if israce.RaceEnabled {
t.Skip("skip since race bug in issue #1222")
}
store1, port1 := mockserver.StartMockTikvService()
require.True(t, port1 > 0)
require.True(t, store1.IsRunning())
@ -908,6 +904,8 @@ func TestRandomRestartStoreAndForwarding(t *testing.T) {
wg := sync.WaitGroup{}
done := int64(0)
concurrency := 500
addr1 := store1.Addr()
addr2 := store2.Addr()
wg.Add(1)
go func() {
defer wg.Done()
@ -931,7 +929,7 @@ func TestRandomRestartStoreAndForwarding(t *testing.T) {
}
}()
conn, err := client1.getConnArray(store1.Addr(), true)
conn, err := client1.getConnArray(addr1, true)
assert.Nil(t, err)
for j := 0; j < concurrency; j++ {
wg.Add(1)
@ -944,9 +942,9 @@ func TestRandomRestartStoreAndForwarding(t *testing.T) {
req := &tikvpb.BatchCommandsRequest_Request{Cmd: &tikvpb.BatchCommandsRequest_Request_Coprocessor{Coprocessor: &coprocessor.Request{}}}
forwardedHost := ""
if i%2 != 0 {
forwardedHost = store2.Addr()
forwardedHost = addr2
}
_, err := sendBatchRequest(context.Background(), store1.Addr(), forwardedHost, conn.batchConn, req, time.Millisecond*50, 0)
_, err := sendBatchRequest(context.Background(), addr1, forwardedHost, conn.batchConn, req, time.Millisecond*50, 0)
if err == nil ||
err.Error() == "EOF" ||
err.Error() == "rpc error: code = Unavailable desc = error reading from server: EOF" ||
@ -964,11 +962,24 @@ func TestRandomRestartStoreAndForwarding(t *testing.T) {
for _, cli := range conn.batchConn.batchCommandsClients {
require.Equal(t, int64(9223372036854775807), cli.maxConcurrencyRequestLimit.Load())
require.True(t, cli.available() > 0, fmt.Sprintf("sent: %d", cli.sent.Load()))
// TODO(crazycs520): fix me, see https://github.com/tikv/client-go/pull/1219
//require.True(t, cli.sent.Load() >= 0, fmt.Sprintf("sent: %d", cli.sent.Load()))
require.True(t, cli.sent.Load() >= 0, fmt.Sprintf("sent: %d", cli.sent.Load()))
}
}
func TestFastFailRequest(t *testing.T) {
client := NewRPCClient()
defer func() {
err := client.Close()
require.NoError(t, err)
}()
start := time.Now()
unknownAddr := "127.0.0.1:52027"
req := tikvrpc.NewRequest(tikvrpc.CmdGet, &kvrpcpb.GetRequest{Key: []byte("key")})
_, err := client.sendRequest(context.Background(), unknownAddr, req, time.Second*20)
require.Equal(t, "context deadline exceeded", errors.Cause(err).Error())
require.True(t, time.Since(start) < time.Second*6) // fast fail when dial target failed.
}
func TestErrConn(t *testing.T) {
e := errors.New("conn error")
err1 := &ErrConn{Err: e, Addr: "127.0.0.1", Ver: 10}