mmaprototype: add Status to storeState

It is included in the Stringer, but not otherwise hooked up
to anything.

Author: tbg

pkg/kv/kvserver/allocator/mmaprototype/cluster_state.go

@@ -600,6 +600,7 @@ type pendingReplicaChange struct {
 // storeState maintains the complete state about a store as known to the
 // allocator.
 type storeState struct {
+	status Status
 	storeLoad
 	StoreAttributesAndLocality
 	adjusted struct {
@@ -1822,6 +1823,11 @@ func (cs *clusterState) setStore(sal StoreAttributesAndLocality) {
 	if !ok {
 		// This is the first time seeing this store.
 		ss := newStoreState()
+		// At this point, the store's health is unknown. It will need to be marked
+		// as healthy separately. Until we know more, we won't place leases or
+		// replicas on it (nor will we try to shed any that are already reported to
+		// have replicas on it).
+		ss.status = MakeStatus(HealthUnknown, LeaseDispositionRefusing, ReplicaDispositionRefusing)
 		ss.localityTiers = cs.localityTierInterner.intern(sal.locality())
 		ss.overloadStartTime = cs.ts.Now()
 		ss.overloadEndTime = cs.ts.Now()

pkg/kv/kvserver/allocator/mmaprototype/testdata/cluster_state/multiple_ranges

@@ -36,9 +36,9 @@ store-id=1
 # The top-k uses CPURate.
 get-load-info
 ----
-store-id=1 node-id=1 reported=[cpu:100, write-bandwidth:40, byte-size:50] adjusted=[cpu:100, write-bandwidth:40, byte-size:50] node-reported-cpu=100 node-adjusted-cpu=100 seq=1
+store-id=1 node-id=1 status=unknown, refusing=leases,replicas reported=[cpu:100, write-bandwidth:40, byte-size:50] adjusted=[cpu:100, write-bandwidth:40, byte-size:50] node-reported-cpu=100 node-adjusted-cpu=100 seq=1
   top-k-ranges (local-store-id=1) dim=CPURate: r4 r3
-store-id=2 node-id=2 reported=[cpu:45, write-bandwidth:40, byte-size:50] adjusted=[cpu:45, write-bandwidth:40, byte-size:50] node-reported-cpu=45 node-adjusted-cpu=45 seq=1
+store-id=2 node-id=2 status=unknown, refusing=leases,replicas reported=[cpu:45, write-bandwidth:40, byte-size:50] adjusted=[cpu:45, write-bandwidth:40, byte-size:50] node-reported-cpu=45 node-adjusted-cpu=45 seq=1
   top-k-ranges (local-store-id=1) dim=CPURate: r3 r2
 
 
@@ -67,9 +67,9 @@ store-id=1
 # The top-2 ranges for store 2 are based on ByteSize.
 get-load-info
 ----
-store-id=1 node-id=1 reported=[cpu:100, write-bandwidth:40, byte-size:50] adjusted=[cpu:100, write-bandwidth:40, byte-size:50] node-reported-cpu=100 node-adjusted-cpu=100 seq=1
+store-id=1 node-id=1 status=unknown, refusing=leases,replicas reported=[cpu:100, write-bandwidth:40, byte-size:50] adjusted=[cpu:100, write-bandwidth:40, byte-size:50] node-reported-cpu=100 node-adjusted-cpu=100 seq=1
   top-k-ranges (local-store-id=1) dim=CPURate: r4 r3
-store-id=2 node-id=2 reported=[cpu:45, write-bandwidth:40, byte-size:50] adjusted=[cpu:45, write-bandwidth:40, byte-size:50] node-reported-cpu=45 node-adjusted-cpu=45 seq=2
+store-id=2 node-id=2 status=unknown, refusing=leases,replicas reported=[cpu:45, write-bandwidth:40, byte-size:50] adjusted=[cpu:45, write-bandwidth:40, byte-size:50] node-reported-cpu=45 node-adjusted-cpu=45 seq=2
   top-k-ranges (local-store-id=1) dim=ByteSize: r1 r2
 
 # StoreLeaseholderMsg not containing r1 and r4 since no longer the leaseholder.
@@ -86,7 +86,7 @@ store-id=1
 # r1 and r4 no longer mentioned in the top-k.
 get-load-info
 ----
-store-id=1 node-id=1 reported=[cpu:100, write-bandwidth:40, byte-size:50] adjusted=[cpu:100, write-bandwidth:40, byte-size:50] node-reported-cpu=100 node-adjusted-cpu=100 seq=1
+store-id=1 node-id=1 status=unknown, refusing=leases,replicas reported=[cpu:100, write-bandwidth:40, byte-size:50] adjusted=[cpu:100, write-bandwidth:40, byte-size:50] node-reported-cpu=100 node-adjusted-cpu=100 seq=1
   top-k-ranges (local-store-id=1) dim=CPURate: r3 r2
-store-id=2 node-id=2 reported=[cpu:45, write-bandwidth:40, byte-size:50] adjusted=[cpu:45, write-bandwidth:40, byte-size:50] node-reported-cpu=45 node-adjusted-cpu=45 seq=2
+store-id=2 node-id=2 status=unknown, refusing=leases,replicas reported=[cpu:45, write-bandwidth:40, byte-size:50] adjusted=[cpu:45, write-bandwidth:40, byte-size:50] node-reported-cpu=45 node-adjusted-cpu=45 seq=2
   top-k-ranges (local-store-id=1) dim=ByteSize: r2 r3

pkg/kv/kvserver/allocator/mmaprototype/testdata/cluster_state/rebalance_replica

@@ -13,8 +13,8 @@ store-load-msg
 
 get-load-info
 ----
-store-id=1 node-id=1 reported=[cpu:80, write-bandwidth:80, byte-size:80] adjusted=[cpu:80, write-bandwidth:80, byte-size:80] node-reported-cpu=80 node-adjusted-cpu=80 seq=1
-store-id=2 node-id=2 reported=[cpu:0, write-bandwidth:0, byte-size:0] adjusted=[cpu:0, write-bandwidth:0, byte-size:0] node-reported-cpu=0 node-adjusted-cpu=0 seq=0
+store-id=1 node-id=1 status=unknown, refusing=leases,replicas reported=[cpu:80, write-bandwidth:80, byte-size:80] adjusted=[cpu:80, write-bandwidth:80, byte-size:80] node-reported-cpu=80 node-adjusted-cpu=80 seq=1
+store-id=2 node-id=2 status=unknown, refusing=leases,replicas reported=[cpu:0, write-bandwidth:0, byte-size:0] adjusted=[cpu:0, write-bandwidth:0, byte-size:0] node-reported-cpu=0 node-adjusted-cpu=0 seq=0
 
 store-leaseholder-msg 
 store-id=1
@@ -30,9 +30,9 @@ range-id=1 load=[cpu:80, write-bandwidth:80, byte-size:80] raft-cpu=20
 
 get-load-info
 ----
-store-id=1 node-id=1 reported=[cpu:80, write-bandwidth:80, byte-size:80] adjusted=[cpu:80, write-bandwidth:80, byte-size:80] node-reported-cpu=80 node-adjusted-cpu=80 seq=1
+store-id=1 node-id=1 status=unknown, refusing=leases,replicas reported=[cpu:80, write-bandwidth:80, byte-size:80] adjusted=[cpu:80, write-bandwidth:80, byte-size:80] node-reported-cpu=80 node-adjusted-cpu=80 seq=1
   top-k-ranges (local-store-id=1) dim=CPURate: r1
-store-id=2 node-id=2 reported=[cpu:0, write-bandwidth:0, byte-size:0] adjusted=[cpu:0, write-bandwidth:0, byte-size:0] node-reported-cpu=0 node-adjusted-cpu=0 seq=0
+store-id=2 node-id=2 status=unknown, refusing=leases,replicas reported=[cpu:0, write-bandwidth:0, byte-size:0] adjusted=[cpu:0, write-bandwidth:0, byte-size:0] node-reported-cpu=0 node-adjusted-cpu=0 seq=0
 
 make-pending-changes range-id=1
   rebalance-replica: remove-store-id=1 add-store-id=2
@@ -52,9 +52,9 @@ range-id=1 load=[cpu:80, write-bandwidth:80, byte-size:80] raft-cpu=20
 
 get-load-info
 ----
-store-id=1 node-id=1 reported=[cpu:80, write-bandwidth:80, byte-size:80] adjusted=[cpu:0, write-bandwidth:0, byte-size:0] node-reported-cpu=80 node-adjusted-cpu=0 seq=2
+store-id=1 node-id=1 status=unknown, refusing=leases,replicas reported=[cpu:80, write-bandwidth:80, byte-size:80] adjusted=[cpu:0, write-bandwidth:0, byte-size:0] node-reported-cpu=80 node-adjusted-cpu=0 seq=2
   top-k-ranges (local-store-id=1) dim=CPURate: r1
-store-id=2 node-id=2 reported=[cpu:0, write-bandwidth:0, byte-size:0] adjusted=[cpu:88, write-bandwidth:88, byte-size:88] node-reported-cpu=0 node-adjusted-cpu=88 seq=1
+store-id=2 node-id=2 status=unknown, refusing=leases,replicas reported=[cpu:0, write-bandwidth:0, byte-size:0] adjusted=[cpu:88, write-bandwidth:88, byte-size:88] node-reported-cpu=0 node-adjusted-cpu=88 seq=1
 
 # TODO: this test is confusing since it is sending StoreLeaseholderMsgs from
 # s1 and s2 to the same clusterState, but s1 and s2 are not on the same node.
@@ -66,9 +66,9 @@ store-id=2
 
 get-load-info
 ----
-store-id=1 node-id=1 reported=[cpu:80, write-bandwidth:80, byte-size:80] adjusted=[cpu:0, write-bandwidth:0, byte-size:0] node-reported-cpu=80 node-adjusted-cpu=0 seq=2
+store-id=1 node-id=1 status=unknown, refusing=leases,replicas reported=[cpu:80, write-bandwidth:80, byte-size:80] adjusted=[cpu:0, write-bandwidth:0, byte-size:0] node-reported-cpu=80 node-adjusted-cpu=0 seq=2
   top-k-ranges (local-store-id=1) dim=CPURate: r1
-store-id=2 node-id=2 reported=[cpu:0, write-bandwidth:0, byte-size:0] adjusted=[cpu:88, write-bandwidth:88, byte-size:88] node-reported-cpu=0 node-adjusted-cpu=88 seq=1
+store-id=2 node-id=2 status=unknown, refusing=leases,replicas reported=[cpu:0, write-bandwidth:0, byte-size:0] adjusted=[cpu:88, write-bandwidth:88, byte-size:88] node-reported-cpu=0 node-adjusted-cpu=88 seq=1
   top-k-ranges (local-store-id=2) dim=ByteSize: r1
 
 get-pending-changes
@@ -94,9 +94,9 @@ change-id=2 store-id=1 node-id=1 range-id=1 load-delta=[cpu:-80, write-bandwidth
 
 get-load-info
 ----
-store-id=1 node-id=1 reported=[cpu:80, write-bandwidth:80, byte-size:80] adjusted=[cpu:0, write-bandwidth:0, byte-size:0] node-reported-cpu=80 node-adjusted-cpu=0 seq=2
+store-id=1 node-id=1 status=unknown, refusing=leases,replicas reported=[cpu:80, write-bandwidth:80, byte-size:80] adjusted=[cpu:0, write-bandwidth:0, byte-size:0] node-reported-cpu=80 node-adjusted-cpu=0 seq=2
   top-k-ranges (local-store-id=1) dim=CPURate: r1
-store-id=2 node-id=2 reported=[cpu:80, write-bandwidth:80, byte-size:80] adjusted=[cpu:80, write-bandwidth:80, byte-size:80] node-reported-cpu=80 node-adjusted-cpu=80 seq=2
+store-id=2 node-id=2 status=unknown, refusing=leases,replicas reported=[cpu:80, write-bandwidth:80, byte-size:80] adjusted=[cpu:80, write-bandwidth:80, byte-size:80] node-reported-cpu=80 node-adjusted-cpu=80 seq=2
   top-k-ranges (local-store-id=2) dim=ByteSize: r1
 
 
@@ -110,7 +110,7 @@ pending(0)
 
 get-load-info
 ----
-store-id=1 node-id=1 reported=[cpu:0, write-bandwidth:0, byte-size:0] adjusted=[cpu:0, write-bandwidth:0, byte-size:0] node-reported-cpu=0 node-adjusted-cpu=0 seq=3
+store-id=1 node-id=1 status=unknown, refusing=leases,replicas reported=[cpu:0, write-bandwidth:0, byte-size:0] adjusted=[cpu:0, write-bandwidth:0, byte-size:0] node-reported-cpu=0 node-adjusted-cpu=0 seq=3
   top-k-ranges (local-store-id=1) dim=CPURate: r1
-store-id=2 node-id=2 reported=[cpu:80, write-bandwidth:80, byte-size:80] adjusted=[cpu:80, write-bandwidth:80, byte-size:80] node-reported-cpu=80 node-adjusted-cpu=80 seq=2
+store-id=2 node-id=2 status=unknown, refusing=leases,replicas reported=[cpu:80, write-bandwidth:80, byte-size:80] adjusted=[cpu:80, write-bandwidth:80, byte-size:80] node-reported-cpu=80 node-adjusted-cpu=80 seq=2
   top-k-ranges (local-store-id=2) dim=ByteSize: r1

pkg/kv/kvserver/allocator/mmaprototype/testdata/cluster_state/remove_replica

@@ -13,8 +13,8 @@ store-load-msg
 
 get-load-info
 ----
-store-id=1 node-id=1 reported=[cpu:0, write-bandwidth:0, byte-size:0] adjusted=[cpu:0, write-bandwidth:0, byte-size:0] node-reported-cpu=0 node-adjusted-cpu=0 seq=0
-store-id=2 node-id=2 reported=[cpu:20, write-bandwidth:80, byte-size:80] adjusted=[cpu:20, write-bandwidth:80, byte-size:80] node-reported-cpu=20 node-adjusted-cpu=20 seq=1
+store-id=1 node-id=1 status=unknown, refusing=leases,replicas reported=[cpu:0, write-bandwidth:0, byte-size:0] adjusted=[cpu:0, write-bandwidth:0, byte-size:0] node-reported-cpu=0 node-adjusted-cpu=0 seq=0
+store-id=2 node-id=2 status=unknown, refusing=leases,replicas reported=[cpu:20, write-bandwidth:80, byte-size:80] adjusted=[cpu:20, write-bandwidth:80, byte-size:80] node-reported-cpu=20 node-adjusted-cpu=20 seq=1
 
 store-leaseholder-msg 
 store-id=1
@@ -25,9 +25,9 @@ store-id=1
 
 get-load-info
 ----
-store-id=1 node-id=1 reported=[cpu:0, write-bandwidth:0, byte-size:0] adjusted=[cpu:0, write-bandwidth:0, byte-size:0] node-reported-cpu=0 node-adjusted-cpu=0 seq=0
+store-id=1 node-id=1 status=unknown, refusing=leases,replicas reported=[cpu:0, write-bandwidth:0, byte-size:0] adjusted=[cpu:0, write-bandwidth:0, byte-size:0] node-reported-cpu=0 node-adjusted-cpu=0 seq=0
   top-k-ranges (local-store-id=1) dim=CPURate: r1
-store-id=2 node-id=2 reported=[cpu:20, write-bandwidth:80, byte-size:80] adjusted=[cpu:20, write-bandwidth:80, byte-size:80] node-reported-cpu=20 node-adjusted-cpu=20 seq=1
+store-id=2 node-id=2 status=unknown, refusing=leases,replicas reported=[cpu:20, write-bandwidth:80, byte-size:80] adjusted=[cpu:20, write-bandwidth:80, byte-size:80] node-reported-cpu=20 node-adjusted-cpu=20 seq=1
   top-k-ranges (local-store-id=1) dim=CPURate: r1
 
 ranges
@@ -55,9 +55,9 @@ range-id=1 load=[cpu:80, write-bandwidth:80, byte-size:80] raft-cpu=20
 # applied.
 get-load-info
 ----
-store-id=1 node-id=1 reported=[cpu:0, write-bandwidth:0, byte-size:0] adjusted=[cpu:0, write-bandwidth:0, byte-size:0] node-reported-cpu=0 node-adjusted-cpu=0 seq=0
+store-id=1 node-id=1 status=unknown, refusing=leases,replicas reported=[cpu:0, write-bandwidth:0, byte-size:0] adjusted=[cpu:0, write-bandwidth:0, byte-size:0] node-reported-cpu=0 node-adjusted-cpu=0 seq=0
   top-k-ranges (local-store-id=1) dim=CPURate: r1
-store-id=2 node-id=2 reported=[cpu:20, write-bandwidth:80, byte-size:80] adjusted=[cpu:0, write-bandwidth:0, byte-size:0] node-reported-cpu=20 node-adjusted-cpu=0 seq=2
+store-id=2 node-id=2 status=unknown, refusing=leases,replicas reported=[cpu:20, write-bandwidth:80, byte-size:80] adjusted=[cpu:0, write-bandwidth:0, byte-size:0] node-reported-cpu=20 node-adjusted-cpu=0 seq=2
   top-k-ranges (local-store-id=1) dim=CPURate: r1
 
 store-leaseholder-msg
@@ -68,9 +68,9 @@ store-id=1
 
 get-load-info
 ----
-store-id=1 node-id=1 reported=[cpu:0, write-bandwidth:0, byte-size:0] adjusted=[cpu:0, write-bandwidth:0, byte-size:0] node-reported-cpu=0 node-adjusted-cpu=0 seq=0
+store-id=1 node-id=1 status=unknown, refusing=leases,replicas reported=[cpu:0, write-bandwidth:0, byte-size:0] adjusted=[cpu:0, write-bandwidth:0, byte-size:0] node-reported-cpu=0 node-adjusted-cpu=0 seq=0
   top-k-ranges (local-store-id=1) dim=CPURate: r1
-store-id=2 node-id=2 reported=[cpu:20, write-bandwidth:80, byte-size:80] adjusted=[cpu:0, write-bandwidth:0, byte-size:0] node-reported-cpu=20 node-adjusted-cpu=0 seq=2
+store-id=2 node-id=2 status=unknown, refusing=leases,replicas reported=[cpu:20, write-bandwidth:80, byte-size:80] adjusted=[cpu:0, write-bandwidth:0, byte-size:0] node-reported-cpu=20 node-adjusted-cpu=0 seq=2
 
 get-pending-changes
 ----
@@ -85,9 +85,9 @@ store-load-msg
 
 get-load-info
 ----
-store-id=1 node-id=1 reported=[cpu:0, write-bandwidth:0, byte-size:0] adjusted=[cpu:0, write-bandwidth:0, byte-size:0] node-reported-cpu=0 node-adjusted-cpu=0 seq=0
+store-id=1 node-id=1 status=unknown, refusing=leases,replicas reported=[cpu:0, write-bandwidth:0, byte-size:0] adjusted=[cpu:0, write-bandwidth:0, byte-size:0] node-reported-cpu=0 node-adjusted-cpu=0 seq=0
   top-k-ranges (local-store-id=1) dim=CPURate: r1
-store-id=2 node-id=2 reported=[cpu:0, write-bandwidth:0, byte-size:0] adjusted=[cpu:0, write-bandwidth:0, byte-size:0] node-reported-cpu=0 node-adjusted-cpu=0 seq=3
+store-id=2 node-id=2 status=unknown, refusing=leases,replicas reported=[cpu:0, write-bandwidth:0, byte-size:0] adjusted=[cpu:0, write-bandwidth:0, byte-size:0] node-reported-cpu=0 node-adjusted-cpu=0 seq=3
 
 get-pending-changes
 ----

pkg/kv/kvserver/allocator/mmaprototype/testdata/cluster_state/transfer_lease

@@ -17,8 +17,8 @@ store-load-msg
 
 get-load-info
 ----
-store-id=1 node-id=1 reported=[cpu:80, write-bandwidth:80, byte-size:80] adjusted=[cpu:80, write-bandwidth:80, byte-size:80] node-reported-cpu=80 node-adjusted-cpu=80 seq=1
-store-id=2 node-id=2 reported=[cpu:20, write-bandwidth:80, byte-size:80] adjusted=[cpu:20, write-bandwidth:80, byte-size:80] node-reported-cpu=20 node-adjusted-cpu=20 seq=1
+store-id=1 node-id=1 status=unknown, refusing=leases,replicas reported=[cpu:80, write-bandwidth:80, byte-size:80] adjusted=[cpu:80, write-bandwidth:80, byte-size:80] node-reported-cpu=80 node-adjusted-cpu=80 seq=1
+store-id=2 node-id=2 status=unknown, refusing=leases,replicas reported=[cpu:20, write-bandwidth:80, byte-size:80] adjusted=[cpu:20, write-bandwidth:80, byte-size:80] node-reported-cpu=20 node-adjusted-cpu=20 seq=1
 
 store-leaseholder-msg 
 store-id=1
@@ -29,9 +29,9 @@ store-id=1
 
 get-load-info
 ----
-store-id=1 node-id=1 reported=[cpu:80, write-bandwidth:80, byte-size:80] adjusted=[cpu:80, write-bandwidth:80, byte-size:80] node-reported-cpu=80 node-adjusted-cpu=80 seq=1
+store-id=1 node-id=1 status=unknown, refusing=leases,replicas reported=[cpu:80, write-bandwidth:80, byte-size:80] adjusted=[cpu:80, write-bandwidth:80, byte-size:80] node-reported-cpu=80 node-adjusted-cpu=80 seq=1
   top-k-ranges (local-store-id=1) dim=CPURate: r1
-store-id=2 node-id=2 reported=[cpu:20, write-bandwidth:80, byte-size:80] adjusted=[cpu:20, write-bandwidth:80, byte-size:80] node-reported-cpu=20 node-adjusted-cpu=20 seq=1
+store-id=2 node-id=2 status=unknown, refusing=leases,replicas reported=[cpu:20, write-bandwidth:80, byte-size:80] adjusted=[cpu:20, write-bandwidth:80, byte-size:80] node-reported-cpu=20 node-adjusted-cpu=20 seq=1
   top-k-ranges (local-store-id=1) dim=WriteBandwidth: r1
 
 ranges
@@ -53,9 +53,9 @@ change-id=2 store-id=2 node-id=2 range-id=1 load-delta=[cpu:66, write-bandwidth:
 
 get-load-info
 ----
-store-id=1 node-id=1 reported=[cpu:80, write-bandwidth:80, byte-size:80] adjusted=[cpu:20, write-bandwidth:80, byte-size:80] node-reported-cpu=80 node-adjusted-cpu=20 seq=2
+store-id=1 node-id=1 status=unknown, refusing=leases,replicas reported=[cpu:80, write-bandwidth:80, byte-size:80] adjusted=[cpu:20, write-bandwidth:80, byte-size:80] node-reported-cpu=80 node-adjusted-cpu=20 seq=2
   top-k-ranges (local-store-id=1) dim=CPURate: r1
-store-id=2 node-id=2 reported=[cpu:20, write-bandwidth:80, byte-size:80] adjusted=[cpu:86, write-bandwidth:80, byte-size:80] node-reported-cpu=20 node-adjusted-cpu=86 seq=2
+store-id=2 node-id=2 status=unknown, refusing=leases,replicas reported=[cpu:20, write-bandwidth:80, byte-size:80] adjusted=[cpu:86, write-bandwidth:80, byte-size:80] node-reported-cpu=20 node-adjusted-cpu=86 seq=2
   top-k-ranges (local-store-id=1) dim=WriteBandwidth: r1
 
 ranges
@@ -70,9 +70,9 @@ pending(0)
 
 get-load-info
 ----
-store-id=1 node-id=1 reported=[cpu:80, write-bandwidth:80, byte-size:80] adjusted=[cpu:80, write-bandwidth:80, byte-size:80] node-reported-cpu=80 node-adjusted-cpu=80 seq=3
+store-id=1 node-id=1 status=unknown, refusing=leases,replicas reported=[cpu:80, write-bandwidth:80, byte-size:80] adjusted=[cpu:80, write-bandwidth:80, byte-size:80] node-reported-cpu=80 node-adjusted-cpu=80 seq=3
   top-k-ranges (local-store-id=1) dim=CPURate: r1
-store-id=2 node-id=2 reported=[cpu:20, write-bandwidth:80, byte-size:80] adjusted=[cpu:20, write-bandwidth:80, byte-size:80] node-reported-cpu=20 node-adjusted-cpu=20 seq=3
+store-id=2 node-id=2 status=unknown, refusing=leases,replicas reported=[cpu:20, write-bandwidth:80, byte-size:80] adjusted=[cpu:20, write-bandwidth:80, byte-size:80] node-reported-cpu=20 node-adjusted-cpu=20 seq=3
   top-k-ranges (local-store-id=1) dim=WriteBandwidth: r1
 
 ranges