Files
zeekay e1e0356e75 rip gRPC from 5 services: gRPC-free interfaces over ZAP, deadline-honoring calls
filer, master, iam, plugin and worker no longer use google.golang.org/grpc on
the client path.

- gRPC-free client interfaces: the generated *_grpc.pb.go client interfaces drop
  "opts ...grpc.CallOption" (no caller ever passed one) and their streaming
  methods return the shared seam types in s3/pb/rpc (ServerStream[T] /
  BidiStream[Req,Resp]) instead of grpc.ServerStreamingClient/BidiStreamingClient.
  The gRPC server scaffolding (Unimplemented*Server, Register*Server, ServiceDesc,
  handler trampolines) is deleted; servers serve over the ZAP wire Dispatch.
- The ZAP adapters (filerzap/masterzap/...) implement the gRPC-free interfaces;
  grpc.DialOption becomes pb.DialOption (the dial is the ZAP transport).
- Unary calls honor context deadlines: the wire client threads ctx to
  transport.Conn.CallContext (v1.8.1), so a call to an unresponsive peer aborts
  on the caller's deadline instead of blocking forever.
- Error semantics: the filer tags each failure with its code name ("NotFound: …",
  "Internal: …") in the message, which crosses the ZAP wire and is classified by
  mount/error_classifier.go — replacing grpc status/codes.
- master ZAP listener port: grpcPort+10000 could exceed 65535; made overflow-safe.
- test doubles updated to the gRPC-free interfaces (0 tests skipped).
- dead cmd/s3-db and cmd/s3-sql (postgres-wire / SQL-engine entry points whose
  backends were already ripped) deleted; finished the backend-rip import cleanup.

go build ./... green; go test ./s3/... green (heavy stream benches run under
their existing -short gates).

Still on gRPC (next wave): volume_server + mq (reverted — servers/adapters not
ready), the remaining ~23 plumbing files, and the go.mod dependency itself.
2026-06-25 06:39:01 -07:00

141 lines
4.0 KiB
Go

package volume_balance_test
import (
"context"
"testing"
"time"
"github.com/stretchr/testify/require"
"google.golang.org/protobuf/proto"
"github.com/hanzoai/s3/s3/pb"
"github.com/hanzoai/s3/s3/pb/master_pb"
"github.com/hanzoai/s3/s3/pb/plugin_pb"
"github.com/hanzoai/s3/s3/pb/worker_pb"
pluginworker "github.com/hanzoai/s3/s3/plugin/worker"
"github.com/hanzoai/s3/s3/worker/tasks/balance"
pluginworkers "github.com/hanzoai/s3/test/plugin_workers"
)
func TestVolumeBalanceDetectionIntegration(t *testing.T) {
response := buildBalanceVolumeListResponse(t)
master := pluginworkers.NewMasterServer(t, response)
dialOption := pb.DialOption{}
handler := balance.NewVolumeBalanceHandler(dialOption)
harness := pluginworkers.NewHarness(t, pluginworkers.HarnessConfig{
WorkerOptions: pluginworker.WorkerOptions{
GrpcDialOption: dialOption,
},
Handlers: []pluginworker.JobHandler{handler},
})
harness.WaitForJobType("volume_balance")
ctx, cancel := context.WithTimeout(context.Background(), 10*time.Second)
defer cancel()
proposals, err := harness.Plugin().RunDetection(ctx, "volume_balance", &plugin_pb.ClusterContext{
MasterGrpcAddresses: []string{master.Address()},
}, 10)
require.NoError(t, err)
// With default batch_size=20 and 10 overloaded volumes vs 1 underloaded,
// all moves are grouped into a single batch proposal.
require.Len(t, proposals, 1, "expected exactly one batch proposal")
proposal := proposals[0]
require.Equal(t, "volume_balance", proposal.JobType)
paramsValue := proposal.Parameters["task_params_pb"]
require.NotNil(t, paramsValue)
params := &worker_pb.TaskParams{}
require.NoError(t, proto.Unmarshal(paramsValue.GetBytesValue(), params))
bp := params.GetBalanceParams()
require.NotNil(t, bp, "expected BalanceParams in batch proposal")
require.Greater(t, len(bp.Moves), 1, "batch proposal should contain multiple moves")
for _, move := range bp.Moves {
require.NotZero(t, move.VolumeId)
require.NotEmpty(t, move.SourceNode)
require.NotEmpty(t, move.TargetNode)
}
}
func buildBalanceVolumeListResponse(t *testing.T) *master_pb.VolumeListResponse {
t.Helper()
volumeSizeLimitMB := uint64(100)
volumeModifiedAt := time.Now().Add(-2 * time.Hour).Unix()
overloadedVolumes := make([]*master_pb.VolumeInformationMessage, 0, 10)
for i := 0; i < 10; i++ {
volumeID := uint32(1000 + i)
overloadedVolumes = append(overloadedVolumes, &master_pb.VolumeInformationMessage{
Id: volumeID,
Collection: "balance",
DiskId: 0,
Size: 20 * 1024 * 1024,
DeletedByteCount: 0,
ModifiedAtSecond: volumeModifiedAt,
ReplicaPlacement: 1,
ReadOnly: false,
})
}
underloadedVolumes := []*master_pb.VolumeInformationMessage{
{
Id: 2000,
Collection: "balance",
DiskId: 0,
Size: 20 * 1024 * 1024,
DeletedByteCount: 0,
ModifiedAtSecond: volumeModifiedAt,
ReplicaPlacement: 1,
ReadOnly: false,
},
}
overloadedDisk := &master_pb.DiskInfo{
DiskId: 0,
MaxVolumeCount: 100,
VolumeCount: int64(len(overloadedVolumes)),
VolumeInfos: overloadedVolumes,
}
underloadedDisk := &master_pb.DiskInfo{
DiskId: 0,
MaxVolumeCount: 100,
VolumeCount: int64(len(underloadedVolumes)),
VolumeInfos: underloadedVolumes,
}
overloadedNode := &master_pb.DataNodeInfo{
Id: "10.0.0.1:8080",
Address: "10.0.0.1:8080",
DiskInfos: map[string]*master_pb.DiskInfo{"hdd": overloadedDisk},
}
underloadedNode := &master_pb.DataNodeInfo{
Id: "10.0.0.2:8080",
Address: "10.0.0.2:8080",
DiskInfos: map[string]*master_pb.DiskInfo{"hdd": underloadedDisk},
}
rack := &master_pb.RackInfo{
Id: "rack-1",
DataNodeInfos: []*master_pb.DataNodeInfo{overloadedNode, underloadedNode},
}
return &master_pb.VolumeListResponse{
VolumeSizeLimitMb: volumeSizeLimitMB,
TopologyInfo: &master_pb.TopologyInfo{
DataCenterInfos: []*master_pb.DataCenterInfo{
{
Id: "dc-1",
RackInfos: []*master_pb.RackInfo{rack},
},
},
},
}
}