ipfs-cluster/monitor/metrics/store.go

189 lines
4.0 KiB
Go
Raw Normal View History

package metrics
import (
"sort"
"sync"
"github.com/ipfs/ipfs-cluster/api"
peer "github.com/libp2p/go-libp2p-core/peer"
)
// PeerMetrics maps a peer IDs to a metrics window.
type PeerMetrics map[peer.ID]*Window
// Store can be used to store and access metrics.
type Store struct {
mux sync.RWMutex
byName map[string]PeerMetrics
}
// NewStore can be used to create a Store.
func NewStore() *Store {
return &Store{
byName: make(map[string]PeerMetrics),
}
}
// Add inserts a new metric in Metrics.
func (mtrs *Store) Add(m *api.Metric) {
mtrs.mux.Lock()
defer mtrs.mux.Unlock()
name := m.Name
peer := m.Peer
mbyp, ok := mtrs.byName[name]
if !ok {
mbyp = make(PeerMetrics)
mtrs.byName[name] = mbyp
}
window, ok := mbyp[peer]
if !ok {
// We always lock the outer map, so we can use unsafe
// Window.
window = NewWindow(DefaultWindowCap)
mbyp[peer] = window
}
window.Add(m)
}
// RemovePeer removes all metrics related to a peer from the Store.
func (mtrs *Store) RemovePeer(pid peer.ID) {
mtrs.mux.Lock()
for _, metrics := range mtrs.byName {
delete(metrics, pid)
}
mtrs.mux.Unlock()
}
// RemovePeerMetrics removes all metrics of a given name for a given peer ID.
func (mtrs *Store) RemovePeerMetrics(pid peer.ID, name string) {
mtrs.mux.Lock()
metrics := mtrs.byName[name]
delete(metrics, pid)
mtrs.mux.Unlock()
}
Consensus: add new "crdt" consensus component This adds a new "crdt" consensus component using go-ds-crdt. This implies several refactors to fully make cluster consensus-component independent: * Delete mapstate and fully adopt dsstate (after people have migrated). * Return errors from state methods rather than ignoring them. * Add a new "datastore" modules so that we can configure datastores in the main configuration like other components. * Let the consensus components fully define the "state.State". Thus, they do not receive the state, they receive the storage where we put the state (a go-datastore). * Allow to customize how the monitor component obtains Peers() (the current peerset), including avoiding using the current peerset. At the moment the crdt consensus uses the monitoring component to define the current peerset. Therefore the monitor component cannot rely on the consensus component to produce a peerset. * Re-factor/re-implementation of "ipfs-cluster-service state" operations. Includes the dissapearance of the "migrate" one. The CRDT consensus component defines creates a crdt-datastore (with ipfs-lite) and uses it to intitialize a dssate. Thus the crdt-store is elegantly wrapped. Any modifications to the state get automatically replicated to other peers. We store all the CRDT DAG blocks in the local datastore. The consensus components only expose a ReadOnly state, as any modifications to the shared state should happen through them. DHT and PubSub facilities must now be created outside of Cluster and passed in so they can be re-used by different components.
2019-02-20 14:24:25 +00:00
// LatestValid returns all the last known valid metrics of a given type. A metric
// is valid if it has not expired.
func (mtrs *Store) LatestValid(name string) []*api.Metric {
mtrs.mux.RLock()
defer mtrs.mux.RUnlock()
byPeer, ok := mtrs.byName[name]
if !ok {
return []*api.Metric{}
}
metrics := make([]*api.Metric, 0, len(byPeer))
for _, window := range byPeer {
m, err := window.Latest()
// TODO(ajl): for accrual, does it matter if a ping has expired?
if err != nil || m.Discard() {
continue
}
metrics = append(metrics, m)
}
sortedMetrics := api.MetricSlice(metrics)
sort.Stable(sortedMetrics)
return sortedMetrics
}
Consensus: add new "crdt" consensus component This adds a new "crdt" consensus component using go-ds-crdt. This implies several refactors to fully make cluster consensus-component independent: * Delete mapstate and fully adopt dsstate (after people have migrated). * Return errors from state methods rather than ignoring them. * Add a new "datastore" modules so that we can configure datastores in the main configuration like other components. * Let the consensus components fully define the "state.State". Thus, they do not receive the state, they receive the storage where we put the state (a go-datastore). * Allow to customize how the monitor component obtains Peers() (the current peerset), including avoiding using the current peerset. At the moment the crdt consensus uses the monitoring component to define the current peerset. Therefore the monitor component cannot rely on the consensus component to produce a peerset. * Re-factor/re-implementation of "ipfs-cluster-service state" operations. Includes the dissapearance of the "migrate" one. The CRDT consensus component defines creates a crdt-datastore (with ipfs-lite) and uses it to intitialize a dssate. Thus the crdt-store is elegantly wrapped. Any modifications to the state get automatically replicated to other peers. We store all the CRDT DAG blocks in the local datastore. The consensus components only expose a ReadOnly state, as any modifications to the shared state should happen through them. DHT and PubSub facilities must now be created outside of Cluster and passed in so they can be re-used by different components.
2019-02-20 14:24:25 +00:00
// AllMetrics returns the latest metrics for all peers and metrics types. It
// may return expired metrics.
func (mtrs *Store) AllMetrics() []*api.Metric {
mtrs.mux.RLock()
defer mtrs.mux.RUnlock()
result := make([]*api.Metric, 0)
for _, byPeer := range mtrs.byName {
for _, window := range byPeer {
metric, err := window.Latest()
if err != nil || !metric.Valid {
continue
}
result = append(result, metric)
}
}
return result
}
// PeerMetrics returns the latest metrics for a given peer ID for
// all known metrics types. It may return expired metrics.
func (mtrs *Store) PeerMetrics(pid peer.ID) []*api.Metric {
mtrs.mux.RLock()
defer mtrs.mux.RUnlock()
result := make([]*api.Metric, 0)
for _, byPeer := range mtrs.byName {
window, ok := byPeer[pid]
if !ok {
continue
}
metric, err := window.Latest()
Consensus: add new "crdt" consensus component This adds a new "crdt" consensus component using go-ds-crdt. This implies several refactors to fully make cluster consensus-component independent: * Delete mapstate and fully adopt dsstate (after people have migrated). * Return errors from state methods rather than ignoring them. * Add a new "datastore" modules so that we can configure datastores in the main configuration like other components. * Let the consensus components fully define the "state.State". Thus, they do not receive the state, they receive the storage where we put the state (a go-datastore). * Allow to customize how the monitor component obtains Peers() (the current peerset), including avoiding using the current peerset. At the moment the crdt consensus uses the monitoring component to define the current peerset. Therefore the monitor component cannot rely on the consensus component to produce a peerset. * Re-factor/re-implementation of "ipfs-cluster-service state" operations. Includes the dissapearance of the "migrate" one. The CRDT consensus component defines creates a crdt-datastore (with ipfs-lite) and uses it to intitialize a dssate. Thus the crdt-store is elegantly wrapped. Any modifications to the state get automatically replicated to other peers. We store all the CRDT DAG blocks in the local datastore. The consensus components only expose a ReadOnly state, as any modifications to the shared state should happen through them. DHT and PubSub facilities must now be created outside of Cluster and passed in so they can be re-used by different components.
2019-02-20 14:24:25 +00:00
if err != nil || !metric.Valid {
continue
}
result = append(result, metric)
}
return result
}
// PeerMetricAll returns all of a particular metrics for a
// particular peer.
func (mtrs *Store) PeerMetricAll(name string, pid peer.ID) []*api.Metric {
mtrs.mux.RLock()
defer mtrs.mux.RUnlock()
byPeer, ok := mtrs.byName[name]
if !ok {
return nil
}
window, ok := byPeer[pid]
if !ok {
return nil
}
ms := window.All()
return ms
}
// PeerLatest returns the latest of a particular metric for a
// particular peer. It may return an expired metric.
func (mtrs *Store) PeerLatest(name string, pid peer.ID) *api.Metric {
mtrs.mux.RLock()
defer mtrs.mux.RUnlock()
byPeer, ok := mtrs.byName[name]
if !ok {
return nil
}
window, ok := byPeer[pid]
if !ok {
return nil
}
m, err := window.Latest()
if err != nil {
// ignoring error, as nil metric is indicative enough
return nil
}
return m
}
2019-06-23 08:05:32 +00:00
// MetricNames returns all the known metric names
func (mtrs *Store) MetricNames() []string {
mtrs.mux.RLock()
defer mtrs.mux.RUnlock()
list := make([]string, 0, len(mtrs.byName))
for k := range mtrs.byName {
list = append(list, k)
}
return list
}