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

110 lines
3.5 KiB
Go

package vacuum_test
import (
"context"
"fmt"
"testing"
"time"
"github.com/stretchr/testify/require"
"github.com/hanzoai/s3/s3/pb"
"github.com/hanzoai/s3/s3/pb/plugin_pb"
pluginworker "github.com/hanzoai/s3/s3/plugin/worker"
"github.com/hanzoai/s3/s3/worker/tasks/vacuum"
pluginworkers "github.com/hanzoai/s3/test/plugin_workers"
)
func vacuumJobSpec(volumeID uint32, server string) *plugin_pb.JobSpec {
return &plugin_pb.JobSpec{
JobId: fmt.Sprintf("vacuum-job-%d", volumeID),
JobType: "vacuum",
Parameters: map[string]*plugin_pb.ConfigValue{
"volume_id": {
Kind: &plugin_pb.ConfigValue_Int64Value{Int64Value: int64(volumeID)},
},
"server": {
Kind: &plugin_pb.ConfigValue_StringValue{StringValue: server},
},
"collection": {
Kind: &plugin_pb.ConfigValue_StringValue{StringValue: "vac-test"},
},
},
}
}
func TestVacuumExecutionIntegration(t *testing.T) {
volumeID := uint32(202)
dialOption := pb.DialOption{}
handler := vacuum.NewVacuumHandler(dialOption, 1)
harness := pluginworkers.NewHarness(t, pluginworkers.HarnessConfig{
WorkerOptions: pluginworker.WorkerOptions{
GrpcDialOption: dialOption,
},
Handlers: []pluginworker.JobHandler{handler},
})
harness.WaitForJobType("vacuum")
source := pluginworkers.NewVolumeServer(t, "")
source.SetVacuumGarbageRatio(0.6)
ctx, cancel := context.WithTimeout(context.Background(), 30*time.Second)
defer cancel()
result, err := harness.Plugin().ExecuteJob(ctx, vacuumJobSpec(volumeID, source.Address()), nil, 1)
require.NoError(t, err)
require.NotNil(t, result)
require.True(t, result.Success)
checkCalls, compactCalls, commitCalls, cleanupCalls := source.VacuumStats()
require.GreaterOrEqual(t, checkCalls, 2)
require.GreaterOrEqual(t, compactCalls, 1)
require.GreaterOrEqual(t, commitCalls, 1)
// Cleanup is only invoked when Phase 1 (Compact) fails to roll back
// the .cpd/.cpx/.cpldb temp files; on the success path Commit
// consumes them (rename .cpd → .dat, .cpx → .idx, .cpldb → .ldb via
// the leveldb needle map) so no Cleanup call is needed. Matches
// topology.vacuumOneVolumeId which only calls batchVacuumVolumeCleanup
// on the Compact-failure branch.
require.Equal(t, 0, cleanupCalls)
// Phase 3 marks each replica writable so master returns it to the
// writables layout. See upstream hanzo#9685.
require.GreaterOrEqual(t, source.MarkWritableCount(), 1)
}
// A replica that commits still read-only (operator-set, EIO-quarantined,
// disk-space-low) must not be force-marked writable: master built-in vacuum
// skips it via SetVolumeAvailable, and it recovers on its own ReadOnly=false
// heartbeat.
func TestVacuumExecutionSkipsMarkWritableWhenReadOnly(t *testing.T) {
volumeID := uint32(203)
dialOption := pb.DialOption{}
handler := vacuum.NewVacuumHandler(dialOption, 1)
harness := pluginworkers.NewHarness(t, pluginworkers.HarnessConfig{
WorkerOptions: pluginworker.WorkerOptions{
GrpcDialOption: dialOption,
},
Handlers: []pluginworker.JobHandler{handler},
})
harness.WaitForJobType("vacuum")
source := pluginworkers.NewVolumeServer(t, "")
source.SetVacuumGarbageRatio(0.6)
source.SetVacuumCommitReadOnly(true)
ctx, cancel := context.WithTimeout(context.Background(), 30*time.Second)
defer cancel()
result, err := harness.Plugin().ExecuteJob(ctx, vacuumJobSpec(volumeID, source.Address()), nil, 1)
require.NoError(t, err)
require.NotNil(t, result)
require.True(t, result.Success)
_, _, commitCalls, _ := source.VacuumStats()
require.GreaterOrEqual(t, commitCalls, 1)
require.Equal(t, 0, source.MarkWritableCount())
}