fetch bucket replication stats across peers in single call (#14956)

current implementation relied on recursively calling one bucket
at a time across all peers, this would be very slow and chatty
when there are 100's of buckets which would mean 100*peerCount
amount of network operations.

This PR attempts to reduce this entire call into `peerCount`
amount of network calls only. This functionality addresses also a
concern where the Prometheus metrics would significantly slow
down when one of the peers is offline.
This commit is contained in:
Harshavardhana
2022-05-23 09:15:30 -07:00
committed by GitHub
parent 90a52a29c5
commit f8650a3493
10 changed files with 411 additions and 6 deletions

View File

@@ -1845,9 +1845,26 @@ func resyncTarget(oi ObjectInfo, arn string, resetID string, resetBeforeDate tim
return
}
// get the most current of in-memory replication stats and data usage info from crawler.
func getLatestReplicationStats(bucket string, u BucketUsageInfo) (s BucketReplicationStats) {
bucketStats := globalNotificationSys.GetClusterBucketStats(GlobalContext, bucket)
func getAllLatestReplicationStats(bucketsUsage map[string]BucketUsageInfo) (bucketsReplicationStats map[string]BucketReplicationStats) {
peerBucketStatsList := globalNotificationSys.GetClusterAllBucketStats(GlobalContext)
bucketsReplicationStats = make(map[string]BucketReplicationStats, len(bucketsUsage))
for bucket, u := range bucketsUsage {
bucketStats := make([]BucketStats, len(peerBucketStatsList))
for i, peerBucketStats := range peerBucketStatsList {
bucketStat, ok := peerBucketStats[bucket]
if !ok {
continue
}
bucketStats[i] = bucketStat
}
bucketsReplicationStats[bucket] = calculateBucketReplicationStats(bucket, u, bucketStats)
}
return bucketsReplicationStats
}
func calculateBucketReplicationStats(bucket string, u BucketUsageInfo, bucketStats []BucketStats) (s BucketReplicationStats) {
// accumulate cluster bucket stats
stats := make(map[string]*BucketReplicationStat)
var totReplicaSize int64
@@ -1916,6 +1933,12 @@ func getLatestReplicationStats(bucket string, u BucketUsageInfo) (s BucketReplic
return s
}
// get the most current of in-memory replication stats and data usage info from crawler.
func getLatestReplicationStats(bucket string, u BucketUsageInfo) (s BucketReplicationStats) {
bucketStats := globalNotificationSys.GetClusterBucketStats(GlobalContext, bucket)
return calculateBucketReplicationStats(bucket, u, bucketStats)
}
const resyncTimeInterval = time.Minute * 10
// periodicResyncMetaSave saves in-memory resync meta stats to disk in periodic intervals