blob: 0c2d1c2ae4b4d7cbe570eca8d408463dacb827a9 [file]
/*
* Licensed to the Apache Software Foundation (ASF) under one
* or more contributor license agreements. See the NOTICE file
* distributed with this work for additional information
* regarding copyright ownership. The ASF licenses this file
* to you under the Apache License, Version 2.0 (the
* "License"); you may not use this file except in compliance
* with the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing,
* software distributed under the License is distributed on an
* "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
* KIND, either express or implied. See the License for the
* specific language governing permissions and limitations
* under the License.
*/
package diskbalancer
import (
"fmt"
"strings"
"time"
"github.com/apache/incubator-pegasus/admin-cli/executor"
"github.com/apache/incubator-pegasus/admin-cli/executor/toolkits"
"github.com/apache/incubator-pegasus/go-client/idl/base"
)
var WaitRunning = time.Second * 10 // time for wait migrate complete
var WaitCleaning = time.Second * 90 // time for wait garbage replica to clean complete
var ShortCleanInterval = "1" // short time wait the disk cleaner clean garbage replica
var LongCleanInterval = "86400" // long time wait the disk cleaner clean garbage replica
// auto balance target node disk usage:
// -1. change the pegasus server disk cleaner internal for clean temp replica to free disk space in time
// -2. get the optimal migrate action to be ready to balance the disk until can't migrate base latest disk space stats
// -3. if current replica is `primary` status, force assign the replica to `secondary` status
// -4. migrate the replica base `getNextMigrateAction` result
// -5. loop query migrate progress using `DiskMigrate`, it will response `ERR_BUSY` if running
// -6. start next loop until can't allow to balance the node
// -7. recover disk cleaner internal if balance complete
// -8. set meta status to `lively` to balance primary and secondary // TODO(jiashuo1)
func BalanceDiskCapacity(client *executor.Client, replicaServer string, minSize int64, interval int, auto bool) error {
WaitCleaning = time.Second * time.Duration(interval)
if err := changeDiskCleanerInterval(client, replicaServer, ShortCleanInterval); err != nil {
return err
}
defer func() {
if err := changeDiskCleanerInterval(client, replicaServer, LongCleanInterval); err != nil {
toolkits.LogWarn("revert disk cleaner failed")
}
}()
for {
action, err := getNextMigrateAction(client, replicaServer, minSize)
if err != nil {
return err
}
err = executor.DiskMigrate(client, replicaServer, action.replica.Gpid, action.from, action.to)
if err != nil && action.replica.Status != "secondary" {
err := forceAssignReplicaToSecondary(client, replicaServer, action.replica.Gpid)
if err != nil {
return err
}
time.Sleep(WaitRunning)
continue
}
if err != nil {
return fmt.Errorf("migrate(%s) start failed[auto=%v] err = %s", action.toString(), auto, err.Error())
}
toolkits.LogInfo(fmt.Sprintf("migrate(%s) has started[auto=%v], wait complete...\n", action.toString(), auto))
for {
// TODO(jiashuo1): using DiskMigrate RPC to query status, consider support queryDiskMigrateStatus RPC
err = executor.DiskMigrate(client, replicaServer, action.replica.Gpid, action.from, action.to)
if err == nil {
time.Sleep(WaitRunning)
continue
}
if strings.Contains(err.Error(), base.ERR_BUSY.String()) {
toolkits.LogInfo(fmt.Sprintf("migrate(%s) is running, msg=%s, wait complete...\n", action.toString(), err.Error()))
time.Sleep(WaitRunning)
continue
}
toolkits.LogInfo(fmt.Sprintf("migrate(%s) is completed,result=%s, wait[%ds] disk cleaner remove garbage...\n\n",
action.toString(), err.Error(), interval))
break
}
if auto {
time.Sleep(WaitCleaning)
continue
}
time.Sleep(WaitCleaning)
toolkits.LogInfo(fmt.Sprintf("you now disable auto-balance[auto=%v], stop and wait manual loop\n\n", auto))
break
}
return nil
}