Skip to content

GitLab

Commit e993689f authored by Kirill Smelkov's avatar Kirill Smelkov
Browse files

.

parent 7f7169ce
Hide whitespace changes
Inline Side-by-side
Showing with 55 additions and 57 deletions
go/neo/server/cluster_test.go
View file @ e993689f
......@@ -199,9 +199,9 @@ func TestMasterStorage(t *testing.T) {
}
// shortcut for nodetab change
node := func(nt *neo.NodeTable, laddr string, typ neo.NodeType, num int32, state neo.NodeState, idtstamp float64) *traceNode {
node := func(x *neo.NodeCommon, laddr string, typ neo.NodeType, num int32, state neo.NodeState, idtstamp float64) *traceNode {
return &traceNode{
NodeTab: unsafe.Pointer(nt),
NodeTab: unsafe.Pointer(x.NodeTab),
NodeInfo: neo.NodeInfo{
Type: typ,
Addr: xnaddr(laddr),
......@@ -232,8 +232,8 @@ func TestMasterStorage(t *testing.T) {
// M starts listening
tc.Expect(netlisten("m:1"))
tc.Expect(node(M.nodeTab, "m:1", neo.MASTER, 1, neo.RUNNING, 0.0))
tc.Expect(clusterState(&M.clusterState, neo.ClusterRecovering))
tc.Expect(node(&M.node, "m:1", neo.MASTER, 1, neo.RUNNING, 0.0))
tc.Expect(clusterState(&M.node.ClusterState, neo.ClusterRecovering))
// TODO create C; C tries connect to master - rejected ("not yet operational")
......@@ -260,7 +260,7 @@ func TestMasterStorage(t *testing.T) {
IdTimestamp: 0,
}))
tc.Expect(node(M.nodeTab, "s:1", neo.STORAGE, 1, neo.PENDING, 0.01))
tc.Expect(node(&M.node, "s:1", neo.STORAGE, 1, neo.PENDING, 0.01))
tc.Expect(conntx("m:2", "s:2", 1, &neo.AcceptIdentification{
NodeType: neo.MASTER,
......@@ -297,13 +297,13 @@ func TestMasterStorage(t *testing.T) {
exc.Raiseif(err)
})
tc.Expect(node(M.nodeTab, "s:1", neo.STORAGE, 1, neo.RUNNING, 0.01))
tc.Expect(node(&M.node, "s:1", neo.STORAGE, 1, neo.RUNNING, 0.01))
xwait(wg)
// XXX M.partTab <- S1
// M starts verification
tc.Expect(clusterState(&M.clusterState, neo.ClusterVerifying))
tc.Expect(clusterState(&M.node.ClusterState, neo.ClusterVerifying))
tc.Expect(conntx("m:2", "s:2", 1, &neo.NotifyPartitionTable{
PTid: 1,
......@@ -334,7 +334,7 @@ func TestMasterStorage(t *testing.T) {
// TODO M.Stop() while verify
// verification ok; M start service
tc.Expect(clusterState(&M.clusterState, neo.ClusterRunning))
tc.Expect(clusterState(&M.node.ClusterState, neo.ClusterRunning))
// TODO ^^^ should be sent to S
tc.Expect(conntx("m:2", "s:2", 1, &neo.StartOperation{Backup: false}))
......@@ -358,7 +358,7 @@ func TestMasterStorage(t *testing.T) {
IdTimestamp: 0,
}))
tc.Expect(node(M.nodeTab, "", neo.CLIENT, 1, neo.RUNNING, 0.02))
tc.Expect(node(&M.node, "", neo.CLIENT, 1, neo.RUNNING, 0.02))
tc.Expect(conntx("m:3", "c:1", 1, &neo.AcceptIdentification{
NodeType: neo.MASTER,
......
go/neo/server/master.go
View file @ e993689f
......@@ -54,12 +54,12 @@ type Master struct {
// to all nodes in cluster
// XXX dup from .node - kill here
///*
/*
stateMu sync.RWMutex // XXX recheck: needed ?
nodeTab *neo.NodeTable
partTab *neo.PartitionTable
clusterState neo.ClusterState
//*/
*/
// channels controlling main driver
ctlStart chan chan error // request to start cluster
......@@ -101,10 +101,12 @@ func NewMaster(clusterName, serveAddr string, net xnet.Networker) *Master {
ClusterName: clusterName,
Net: net,
MasterAddr: serveAddr, // XXX ok?
},
nodeTab: &neo.NodeTable{},
partTab: &neo.PartitionTable{},
NodeTab: &neo.NodeTable{},
PartTab: &neo.PartitionTable{},
ClusterState: -1, // invalid
},
ctlStart: make(chan chan error),
ctlStop: make(chan chan struct{}),
......@@ -116,7 +118,6 @@ func NewMaster(clusterName, serveAddr string, net xnet.Networker) *Master {
monotime: monotime,
}
m.clusterState = -1 // invalid
return m
}
......@@ -148,7 +149,7 @@ func (m *Master) Shutdown() error {
// setClusterState sets .clusterState and notifies subscribers
func (m *Master) setClusterState(state neo.ClusterState) {
m.clusterState.Set(state)
m.node.ClusterState.Set(state)
// TODO notify subscribers
}
......@@ -181,7 +182,7 @@ func (m *Master) Run(ctx context.Context) (err error) {
}
// update nodeTab with self
m.nodeTab.Update(m.node.MyInfo, nil /*XXX ok? we are not connecting to self*/)
m.node.NodeTab.Update(m.node.MyInfo, nil /*XXX ok? we are not connecting to self*/)
// accept incoming connections and pass them to main driver
......@@ -318,7 +319,7 @@ func (m *Master) recovery(ctx context.Context) (err error) {
// start recovery on all storages we are currently in touch with
// XXX close links to clients
for _, stor := range m.nodeTab.StorageList() {
for _, stor := range m.node.NodeTab.StorageList() {
if stor.State > neo.DOWN { // XXX state cmp ok ? XXX or stor.Link != nil ?
inprogress++
wg.Add(1)
......@@ -374,25 +375,25 @@ loop:
// close stor link / update .nodeTab
lclose(ctx, r.stor.Link)
m.nodeTab.SetNodeState(r.stor, neo.DOWN)
m.node.NodeTab.SetNodeState(r.stor, neo.DOWN)
}
} else {
// we are interested in latest partTab
// NOTE during recovery no one must be subscribed to
// partTab so it is ok to simply change whole m.partTab
if r.partTab.PTid > m.partTab.PTid {
m.partTab = r.partTab
if r.partTab.PTid > m.node.PartTab.PTid {
m.node.PartTab = r.partTab
}
}
// update indicator whether cluster currently can be operational or not
var ready bool
if m.partTab.PTid == 0 {
if m.node.PartTab.PTid == 0 {
// new cluster - allow startup if we have some storages passed
// recovery and there is no in-progress recovery running
nup := 0
for _, stor := range m.nodeTab.StorageList() {
for _, stor := range m.node.NodeTab.StorageList() {
if stor.State > neo.DOWN {
nup++
}
......@@ -400,7 +401,7 @@ loop:
ready = (nup > 0 && inprogress == 0)
} else {
ready = m.partTab.OperationalWith(m.nodeTab) // XXX + node state
ready = m.node.PartTab.OperationalWith(m.node.NodeTab) // XXX + node state
}
if readyToStart != ready {
......@@ -462,7 +463,7 @@ loop2:
// close stor link / update .nodeTab
lclose(ctx, r.stor.Link)
m.nodeTab.SetNodeState(r.stor, neo.DOWN)
m.node.NodeTab.SetNodeState(r.stor, neo.DOWN)
}
case <-done:
......@@ -478,24 +479,24 @@ loop2:
// S PENDING -> RUNNING
// XXX recheck logic is ok for when starting existing cluster
for _, stor := range m.nodeTab.StorageList() {
for _, stor := range m.node.NodeTab.StorageList() {
if stor.State == neo.PENDING {
m.nodeTab.SetNodeState(stor, neo.RUNNING)
m.node.NodeTab.SetNodeState(stor, neo.RUNNING)
}
}
// if we are starting for new cluster - create partition table
if m.partTab.PTid == 0 {
if m.node.PartTab.PTid == 0 {
log.Infof(ctx, "creating new partition table")
// XXX -> m.nodeTab.StorageList(State > DOWN)
storv := []*neo.Node{}
for _, stor := range m.nodeTab.StorageList() {
for _, stor := range m.node.NodeTab.StorageList() {
if stor.State > neo.DOWN {
storv = append(storv, stor)
}
}
m.partTab = neo.MakePartTab(1 /* XXX hardcoded */, storv)
m.partTab.PTid = 1
m.node.PartTab = neo.MakePartTab(1 /* XXX hardcoded */, storv)
m.node.PartTab.PTid = 1
}
return nil
......@@ -583,13 +584,13 @@ func (m *Master) verify(ctx context.Context) (err error) {
// XXX (= py), rationale=?
// start verification on all storages we are currently in touch with
for _, stor := range m.nodeTab.StorageList() {
for _, stor := range m.node.NodeTab.StorageList() {
if stor.State > neo.DOWN { // XXX state cmp ok ? XXX or stor.Link != nil ?
inprogress++
wg.Add(1)
go func() {
defer wg.Done()
storCtlVerify(ctx, stor, m.partTab, verify)
storCtlVerify(ctx, stor, m.node.PartTab, verify)
}()
}
}
......@@ -617,14 +618,14 @@ loop:
return
}
storCtlVerify(ctx, node, m.partTab, verify)
storCtlVerify(ctx, node, m.node.PartTab, verify)
}()
case n := <-m.nodeLeave:
m.nodeTab.SetNodeState(n.node, neo.DOWN)
m.node.NodeTab.SetNodeState(n.node, neo.DOWN)
// if cluster became non-operational - we cancel verification
if !m.partTab.OperationalWith(m.nodeTab) {
if !m.node.PartTab.OperationalWith(m.node.NodeTab) {
// XXX ok to instantly cancel? or better
// graceful shutdown in-flight verifications?
vcancel()
......@@ -648,12 +649,12 @@ loop:
// mark storage as non-working in nodeTab
lclose(ctx, v.stor.Link)
m.nodeTab.SetNodeState(v.stor, neo.DOWN)
m.node.NodeTab.SetNodeState(v.stor, neo.DOWN)
}
// check partTab is still operational
// if not -> cancel to go back to recovery
if !m.partTab.OperationalWith(m.nodeTab) {
if !m.node.PartTab.OperationalWith(m.node.NodeTab) {
vcancel()
err = errClusterDegraded
break loop
......@@ -700,7 +701,7 @@ loop2:
// close stor link / update .nodeTab
lclose(ctx, v.stor.Link)
m.nodeTab.SetNodeState(v.stor, neo.DOWN)
m.node.NodeTab.SetNodeState(v.stor, neo.DOWN)
}
case <-done:
......@@ -798,7 +799,7 @@ func (m *Master) service(ctx context.Context) (err error) {
wg := &sync.WaitGroup{}
// spawn per-storage service driver
for _, stor := range m.nodeTab.StorageList() {
for _, stor := range m.node.NodeTab.StorageList() {
if stor.State == neo.RUNNING { // XXX note PENDING - not adding to service; ok?
wg.Add(1)
go func() {
......@@ -850,10 +851,10 @@ loop:
// XXX who sends here?
case n := <-m.nodeLeave:
m.nodeTab.SetNodeState(n.node, neo.DOWN)
m.node.NodeTab.SetNodeState(n.node, neo.DOWN)
// if cluster became non-operational - cancel service
if !m.partTab.OperationalWith(m.nodeTab) {
if !m.node.PartTab.OperationalWith(m.node.NodeTab) {
err = errClusterDegraded
break loop
}
......@@ -942,8 +943,12 @@ func (m *Master) serveClient(ctx context.Context, cli *neo.Node) (err error) {
func (m *Master) serveClient1(ctx context.Context, req neo.Msg) (resp neo.Msg) {
switch req := req.(type) {
case *neo.AskPartitionTable:
// XXX
panic("TODO")
// XXX lock
rpt := &neo.AnswerPartitionTable{
PTid: m.node.PartTab.PTid,
RowList: m.node.PartTab.Dump(),
}
return rpt
default:
......@@ -978,7 +983,7 @@ func (m *Master) identify(ctx context.Context, n nodeCome) (node *neo.Node, resp
// XXX check uuid matches NodeType
node = m.nodeTab.Get(uuid)
node = m.node.NodeTab.Get(uuid)
if node != nil {
// reject - uuid is already occupied by someone else
// XXX check also for down state - it could be the same node reconnecting
......@@ -989,7 +994,7 @@ func (m *Master) identify(ctx context.Context, n nodeCome) (node *neo.Node, resp
// XXX ok to have this logic inside identify? (better provide from outside ?)
switch nodeType {
case neo.CLIENT:
if m.clusterState != neo.ClusterRunning {
if m.node.ClusterState != neo.ClusterRunning {
return &neo.Error{neo.NOT_READY, "cluster not operational"}
}
......@@ -1039,7 +1044,7 @@ func (m *Master) identify(ctx context.Context, n nodeCome) (node *neo.Node, resp
IdTimestamp: m.monotime(),
}
node = m.nodeTab.Update(nodeInfo, n.conn) // NOTE this notifies all nodeTab subscribers
node = m.node.NodeTab.Update(nodeInfo, n.conn) // NOTE this notifies all nodeTab subscribers
return node, accept
}
......@@ -1080,7 +1085,7 @@ func (m *Master) accept(ctx context.Context, conn *neo.Conn, resp neo.Msg) error
func (m *Master) allocUUID(nodeType neo.NodeType) neo.NodeUUID {
for num := int32(1); num < 1<<24; num++ {
uuid := neo.UUID(nodeType, num)
if m.nodeTab.Get(uuid) == nil {
if m.node.NodeTab.Get(uuid) == nil {
return uuid
}
}
......
go/neo/server/storage.go
View file @ e993689f
......@@ -314,6 +314,8 @@ func (stor *Storage) m1initialize(ctx context.Context, Mconn *neo.Conn) (err err
return err
}
// XXX vvv move Send out of reply preparing logic
switch msg.(type) {
default:
return fmt.Errorf("unexpected message: %T", msg)
......@@ -554,11 +556,6 @@ func (stor *Storage) serveClient(ctx context.Context, conn *neo.Conn) {
// serveClient1 prepares response for 1 request from client
func (stor *Storage) serveClient1(ctx context.Context, req neo.Msg) (resp neo.Msg) {
// req, err := conn.Recv()
// if err != nil {
// return err // XXX log / err / send error before closing
// }
switch req := req.(type) {
case *neo.GetObject:
xid := zodb.Xid{Oid: req.Oid}
......@@ -588,8 +585,6 @@ func (stor *Storage) serveClient1(ctx context.Context, req neo.Msg) (resp neo.Ms
// XXX .DataSerial
}
// req.Reply(reply) // XXX err
case *neo.LastTransaction:
lastTid, err := stor.zstor.LastTid(ctx)
if err != nil {
......@@ -598,8 +593,6 @@ func (stor *Storage) serveClient1(ctx context.Context, req neo.Msg) (resp neo.Ms
return &neo.AnswerLastTransaction{lastTid}
// conn.Send(reply) // XXX err
//case *ObjectHistory:
//case *StoreObject:
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment
GitLab Nexedi Edition | About GitLab | About Nexedi | 沪ICP备2021021310号-2 | 沪ICP备2021021310号-7