/
disk_manager.go
108 lines (94 loc) · 3.31 KB
/
disk_manager.go
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
// Copyright 2018 The Chubao Authors.
//
// Licensed 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 master
import (
"fmt"
"time"
"github.com/chubaofs/chubaofs/util"
"github.com/chubaofs/chubaofs/util/log"
)
func (c *Cluster) scheduleToCheckDiskRecoveryProgress() {
go func() {
for {
if c.partition != nil && c.partition.IsRaftLeader() {
if c.vols != nil {
c.checkDiskRecoveryProgress()
}
}
time.Sleep(time.Second * defaultIntervalToCheckDataPartition)
}
}()
}
func (c *Cluster) checkDiskRecoveryProgress() {
defer func() {
if r := recover(); r != nil {
log.LogWarnf("checkDiskRecoveryProgress occurred panic,err[%v]", r)
WarnBySpecialKey(fmt.Sprintf("%v_%v_scheduling_job_panic", c.Name, ModuleName),
"checkDiskRecoveryProgress occurred panic")
}
}()
c.badPartitionMutex.Lock()
defer c.badPartitionMutex.Unlock()
c.BadDataPartitionIds.Range(func(key, value interface{}) bool {
badDataPartitionIds := value.([]uint64)
newBadDpIds := make([]uint64, 0)
for _, partitionID := range badDataPartitionIds {
partition, err := c.getDataPartitionByID(partitionID)
if err != nil {
Warn(c.Name, fmt.Sprintf("checkDiskRecoveryProgress clusterID[%v],partitionID[%v] is not exist", c.Name, partitionID))
continue
}
vol, err := c.getVol(partition.VolName)
if err != nil {
Warn(c.Name, fmt.Sprintf("checkDiskRecoveryProgress clusterID[%v],partitionID[%v] vol(%s) is not exist",
c.Name, partitionID, partition.VolName))
continue
}
if len(partition.Replicas) == 0 || len(partition.Replicas) < int(vol.dpReplicaNum) {
newBadDpIds = append(newBadDpIds, partitionID)
continue
}
if partition.getMinus() < util.GB {
partition.isRecover = false
partition.RLock()
c.syncUpdateDataPartition(partition)
partition.RUnlock()
Warn(c.Name, fmt.Sprintf("clusterID[%v],partitionID[%v] has recovered success", c.Name, partitionID))
} else {
newBadDpIds = append(newBadDpIds, partitionID)
}
}
if len(newBadDpIds) == 0 {
Warn(c.Name, fmt.Sprintf("clusterID[%v],node:disk[%v] has recovered success", c.Name, key))
c.BadDataPartitionIds.Delete(key)
} else {
c.BadDataPartitionIds.Store(key, newBadDpIds)
log.LogInfof("BadDataPartitionIds key(%s) still have (%d) dp in recover", key, len(newBadDpIds))
}
return true
})
}
func (c *Cluster) decommissionDisk(dataNode *DataNode, badDiskPath string, badPartitions []*DataPartition) (err error) {
msg := fmt.Sprintf("action[decommissionDisk], Node[%v] OffLine,disk[%v]", dataNode.Addr, badDiskPath)
log.LogWarn(msg)
for _, dp := range badPartitions {
if err = c.decommissionDataPartition(dataNode.Addr, dp, diskOfflineErr); err != nil {
return
}
}
msg = fmt.Sprintf("action[decommissionDisk],clusterID[%v] Node[%v] OffLine success",
c.Name, dataNode.Addr)
Warn(c.Name, msg)
return
}