Add cluster.raft.leader.transfer command for graceful leader change (#7819)
* proto: add RaftLeadershipTransfer RPC for forced leader change Add new gRPC RPC and messages for leadership transfer: - RaftLeadershipTransferRequest: optional target_id and target_address - RaftLeadershipTransferResponse: previous_leader and new_leader This enables graceful leadership transfer before master maintenance, reducing errors in filers during planned maintenance windows. Ref: https://github.com/seaweedfs/seaweedfs/issues/7527 * proto: regenerate Go files for RaftLeadershipTransfer Generated from master.proto changes. * master: implement RaftLeadershipTransfer gRPC handler Add gRPC handler for leadership transfer with support for: - Transfer to any eligible follower (when target_id is empty) - Transfer to a specific server (when target_id and target_address are provided) Uses hashicorp/raft LeadershipTransfer() and LeadershipTransferToServer() APIs. Returns the previous and new leader in the response. * shell: add cluster.raft.leader.transfer command Add weed shell command for graceful leadership transfer: - Displays current cluster status before transfer - Supports auto-selection of target (any eligible follower) - Supports targeted transfer with -id and -address flags - Provides clear feedback on success/failure with troubleshooting tips Usage: cluster.raft.leader.transfer cluster.raft.leader.transfer -id <server_id> -address <grpc_address> * master: add unit tests for raft gRPC handlers Add tests covering: - RaftLeadershipTransfer with no raft initialized - RaftLeadershipTransfer with target_id but no address - RaftListClusterServers with no raft initialized - RaftAddServer with no raft initialized - RaftRemoveServer with no raft initialized These tests verify error handling when raft is not configured. * shell: add tests for cluster.raft.leader.transfer command Add tests covering: - Command name and help text validation - HasTag returns false for ResourceHeavy - Validation of -id without -address - Argument parsing with unknown flags * master: clarify that leadership transfer requires -raftHashicorp The default raft implementation (seaweedfs/raft, a goraft fork) does not support graceful leadership transfer. This feature is only available when using hashicorp raft (-raftHashicorp=true). Update error messages and help text to make this requirement clear: - gRPC handler returns specific error for goraft users - Shell command help text notes the requirement - Added test for goraft case * test: use strings.Contains instead of custom helper Replace custom contains/containsHelper functions with the standard library strings.Contains for better maintainability. * shell: return flag parsing errors instead of swallowing them - Return the error from flag.Parse() instead of returning nil - Update test to explicitly assert error for unknown flags * test: document integration test scenarios for Raft leadership transfer Add comments explaining: - Why these unit tests only cover 'Raft not initialized' scenarios - What integration tests should cover (with multi-master cluster) - hashicorp/raft uses concrete types that cannot be easily mocked * fix: address reviewer feedback on tests and leader routing - Remove misleading tests that couldn't properly validate their documented behavior without a real Raft cluster: - TestRaftLeadershipTransfer_GoraftNotSupported - TestRaftLeadershipTransfer_ValidationTargetIdWithoutAddress - Change WithClient(false) to WithClient(true) for RaftLeadershipTransfer RPC to ensure the request is routed to the current leader * Improve cluster.raft.transferLeader command - Rename command from cluster.raft.leader.transfer to cluster.raft.transferLeader - Add symmetric validation: -id and -address must be specified together - Handle case where same leader is re-elected after transfer - Add test for -address without -id validation - Add docker compose file for 5-master raft cluster testing
This commit is contained in:
144
weed/shell/command_cluster_raft_leader_transfer.go
Normal file
144
weed/shell/command_cluster_raft_leader_transfer.go
Normal file
@@ -0,0 +1,144 @@
|
||||
package shell
|
||||
|
||||
import (
|
||||
"context"
|
||||
"flag"
|
||||
"fmt"
|
||||
"io"
|
||||
"time"
|
||||
|
||||
"github.com/seaweedfs/seaweedfs/weed/pb/master_pb"
|
||||
)
|
||||
|
||||
func init() {
|
||||
Commands = append(Commands, &commandRaftLeaderTransfer{})
|
||||
}
|
||||
|
||||
type commandRaftLeaderTransfer struct{}
|
||||
|
||||
func (c *commandRaftLeaderTransfer) Name() string {
|
||||
return "cluster.raft.transferLeader"
|
||||
}
|
||||
|
||||
func (c *commandRaftLeaderTransfer) Help() string {
|
||||
return `transfer raft leadership to another master server
|
||||
|
||||
This command initiates a graceful leadership transfer from the current
|
||||
leader to another server. Use this before performing maintenance on
|
||||
the current leader to reduce errors in filers and other components.
|
||||
|
||||
Examples:
|
||||
# Transfer to any eligible follower (auto-selection)
|
||||
cluster.raft.transferLeader
|
||||
|
||||
# Transfer to a specific server
|
||||
cluster.raft.transferLeader -id <server_id> -address <server_grpc_address>
|
||||
|
||||
Notes:
|
||||
- Requires hashicorp raft (-raftHashicorp=true on master)
|
||||
- This command must be sent to the current leader
|
||||
- The target server must be a voting member of the raft cluster
|
||||
- Use 'cluster.raft.ps' to list available servers and identify the leader
|
||||
`
|
||||
}
|
||||
|
||||
func (c *commandRaftLeaderTransfer) HasTag(CommandTag) bool {
|
||||
return false
|
||||
}
|
||||
|
||||
func (c *commandRaftLeaderTransfer) Do(args []string, commandEnv *CommandEnv, writer io.Writer) error {
|
||||
leaderTransferCommand := flag.NewFlagSet(c.Name(), flag.ContinueOnError)
|
||||
targetId := leaderTransferCommand.String("id", "", "target server id (must be used with -address)")
|
||||
targetAddress := leaderTransferCommand.String("address", "", "target server grpc address (must be used with -id)")
|
||||
|
||||
if err := leaderTransferCommand.Parse(args); err != nil {
|
||||
return err
|
||||
}
|
||||
|
||||
// Validate: id and address must be specified together
|
||||
if *targetId != "" && *targetAddress == "" {
|
||||
return fmt.Errorf("-address is required when -id is specified")
|
||||
}
|
||||
if *targetAddress != "" && *targetId == "" {
|
||||
return fmt.Errorf("-id is required when -address is specified")
|
||||
}
|
||||
|
||||
// First, show current cluster status
|
||||
fmt.Fprintf(writer, "Checking current raft cluster status...\n")
|
||||
|
||||
var currentLeader string
|
||||
err := commandEnv.MasterClient.WithClient(false, func(client master_pb.SeaweedClient) error {
|
||||
ctx, cancel := context.WithTimeout(context.Background(), 10*time.Second)
|
||||
defer cancel()
|
||||
|
||||
resp, err := client.RaftListClusterServers(ctx, &master_pb.RaftListClusterServersRequest{})
|
||||
if err != nil {
|
||||
return fmt.Errorf("failed to list cluster servers: %v", err)
|
||||
}
|
||||
|
||||
if len(resp.ClusterServers) == 0 {
|
||||
fmt.Fprintf(writer, "No raft cluster configured (single master mode)\n")
|
||||
return fmt.Errorf("leadership transfer not available in single master mode")
|
||||
}
|
||||
|
||||
fmt.Fprintf(writer, "Raft cluster has %d servers:\n", len(resp.ClusterServers))
|
||||
for _, server := range resp.ClusterServers {
|
||||
suffix := ""
|
||||
if server.IsLeader {
|
||||
suffix = " <- current leader"
|
||||
currentLeader = server.Id
|
||||
}
|
||||
fmt.Fprintf(writer, " %s %s [%s]%s\n", server.Id, server.Address, server.Suffrage, suffix)
|
||||
}
|
||||
return nil
|
||||
})
|
||||
if err != nil {
|
||||
return err
|
||||
}
|
||||
|
||||
if currentLeader == "" {
|
||||
return fmt.Errorf("no leader found in cluster")
|
||||
}
|
||||
|
||||
// Perform the transfer
|
||||
targetDesc := "any eligible follower"
|
||||
if *targetId != "" {
|
||||
targetDesc = fmt.Sprintf("server %s (%s)", *targetId, *targetAddress)
|
||||
}
|
||||
fmt.Fprintf(writer, "\nTransferring leadership from %s to %s...\n", currentLeader, targetDesc)
|
||||
|
||||
err = commandEnv.MasterClient.WithClient(true, func(client master_pb.SeaweedClient) error {
|
||||
ctx, cancel := context.WithTimeout(context.Background(), 30*time.Second)
|
||||
defer cancel()
|
||||
|
||||
resp, err := client.RaftLeadershipTransfer(ctx, &master_pb.RaftLeadershipTransferRequest{
|
||||
TargetId: *targetId,
|
||||
TargetAddress: *targetAddress,
|
||||
})
|
||||
if err != nil {
|
||||
return fmt.Errorf("leadership transfer failed: %v", err)
|
||||
}
|
||||
|
||||
if resp.PreviousLeader != resp.NewLeader {
|
||||
fmt.Fprintf(writer, "Leadership successfully transferred.\n")
|
||||
fmt.Fprintf(writer, " Previous leader: %s\n", resp.PreviousLeader)
|
||||
fmt.Fprintf(writer, " New leader: %s\n", resp.NewLeader)
|
||||
} else {
|
||||
fmt.Fprintf(writer, "Leadership transfer initiated, but the same leader was re-elected.\n")
|
||||
fmt.Fprintf(writer, " Current leader: %s\n", resp.NewLeader)
|
||||
}
|
||||
return nil
|
||||
})
|
||||
|
||||
if err != nil {
|
||||
fmt.Fprintf(writer, "\nLeadership transfer failed: %v\n", err)
|
||||
fmt.Fprintf(writer, "\nTroubleshooting:\n")
|
||||
fmt.Fprintf(writer, " - Ensure you are connected to the current leader\n")
|
||||
fmt.Fprintf(writer, " - Ensure target server is a voting member (use 'cluster.raft.ps')\n")
|
||||
fmt.Fprintf(writer, " - Ensure target server is healthy and reachable\n")
|
||||
return err
|
||||
}
|
||||
|
||||
return nil
|
||||
}
|
||||
|
||||
89
weed/shell/command_cluster_raft_leader_transfer_test.go
Normal file
89
weed/shell/command_cluster_raft_leader_transfer_test.go
Normal file
@@ -0,0 +1,89 @@
|
||||
package shell
|
||||
|
||||
import (
|
||||
"bytes"
|
||||
"strings"
|
||||
"testing"
|
||||
)
|
||||
|
||||
func TestRaftLeaderTransfer_Name(t *testing.T) {
|
||||
cmd := &commandRaftLeaderTransfer{}
|
||||
expected := "cluster.raft.transferLeader"
|
||||
if cmd.Name() != expected {
|
||||
t.Errorf("expected name %q, got %q", expected, cmd.Name())
|
||||
}
|
||||
}
|
||||
|
||||
func TestRaftLeaderTransfer_Help(t *testing.T) {
|
||||
cmd := &commandRaftLeaderTransfer{}
|
||||
help := cmd.Help()
|
||||
|
||||
// Verify help text contains key information
|
||||
expectedPhrases := []string{
|
||||
"transfer raft leadership",
|
||||
"cluster.raft.transferLeader",
|
||||
"-id",
|
||||
"-address",
|
||||
"cluster.raft.ps",
|
||||
"-raftHashicorp",
|
||||
}
|
||||
|
||||
for _, phrase := range expectedPhrases {
|
||||
if !strings.Contains(help, phrase) {
|
||||
t.Errorf("help text should contain %q", phrase)
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
func TestRaftLeaderTransfer_HasTag(t *testing.T) {
|
||||
cmd := &commandRaftLeaderTransfer{}
|
||||
// The command should not have any special tags
|
||||
if cmd.HasTag(ResourceHeavy) {
|
||||
t.Error("expected HasTag to return false for ResourceHeavy")
|
||||
}
|
||||
}
|
||||
|
||||
func TestRaftLeaderTransfer_ValidateTargetIdWithoutAddress(t *testing.T) {
|
||||
cmd := &commandRaftLeaderTransfer{}
|
||||
var buf bytes.Buffer
|
||||
|
||||
// Create a mock command environment - this will fail because no master client
|
||||
// but we can verify argument parsing
|
||||
err := cmd.Do([]string{"-id", "test-server"}, nil, &buf)
|
||||
|
||||
// Should fail because -address is required when -id is specified
|
||||
if err == nil {
|
||||
t.Error("expected error when -id is specified without -address")
|
||||
}
|
||||
if err != nil && !strings.Contains(err.Error(), "-address is required") {
|
||||
t.Errorf("expected error about missing -address, got: %v", err)
|
||||
}
|
||||
}
|
||||
|
||||
func TestRaftLeaderTransfer_ValidateTargetAddressWithoutId(t *testing.T) {
|
||||
cmd := &commandRaftLeaderTransfer{}
|
||||
var buf bytes.Buffer
|
||||
|
||||
// Verify argument parsing - address without id should fail
|
||||
err := cmd.Do([]string{"-address", "localhost:19333"}, nil, &buf)
|
||||
|
||||
// Should fail because -id is required when -address is specified
|
||||
if err == nil {
|
||||
t.Error("expected error when -address is specified without -id")
|
||||
}
|
||||
if err != nil && !strings.Contains(err.Error(), "-id is required") {
|
||||
t.Errorf("expected error about missing -id, got: %v", err)
|
||||
}
|
||||
}
|
||||
|
||||
func TestRaftLeaderTransfer_UnknownFlag(t *testing.T) {
|
||||
cmd := &commandRaftLeaderTransfer{}
|
||||
var buf bytes.Buffer
|
||||
|
||||
// Unknown flag should return an error
|
||||
err := cmd.Do([]string{"-unknown-flag"}, nil, &buf)
|
||||
if err == nil {
|
||||
t.Error("expected error for unknown flag")
|
||||
}
|
||||
}
|
||||
|
||||
Reference in New Issue
Block a user