Skip to content

fix(redis): 切换同步检查加入自己的心跳机制 #14397 #14399

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Nov 21, 2025
Merged

fix(redis): 切换同步检查加入自己的心跳机制 #14397 #14399

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
29 changes: 28 additions & 1 deletion dbm-services/redis/db-tools/dbactuator/pkg/atomjobs/atomredis/redis_switch.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -929,7 +929,10 @@ func (job *RedisSwitchPreCheck) checkReplicationSync(newMasterConn *myredis.Redi
var masterTime, masterDbsize, slaveTime int64
oldMasterAddr := fmt.Sprintf("%s:%d", storagePair.MasterInfo.IP, storagePair.MasterInfo.Port)
newMasterAddr := fmt.Sprintf("%s:%d", storagePair.SlaveInfo.IP, storagePair.SlaveInfo.Port)

if err := job.trySetHeartBeatWhenDbmonStoped(storagePair); err != nil {
time.Sleep(time.Second * 2)
job.trySetHeartBeatWhenDbmonStoped(storagePair)
}
if err := newMasterConn.SelectDB(1); err != nil {
return fmt.Errorf("[%s] select db 1, exec cmd err:%+v", newMasterAddr, err)
}
Expand Down Expand Up @@ -976,6 +979,30 @@ func (job *RedisSwitchPreCheck) checkReplicationSync(newMasterConn *myredis.Redi
return nil
}

// setHeartBeatWhenDbmonStoped # here we just set on the master heartbeat:
func (job *RedisSwitchPreCheck) trySetHeartBeatWhenDbmonStoped(storagePair InstanceSwitchParam) error {
oldMasterAddr := fmt.Sprintf("%s:%d", storagePair.MasterInfo.IP, storagePair.MasterInfo.Port)
oldMasterConn, err := myredis.NewRedisClientWithTimeout(oldMasterAddr,
job.params.ClusterMeta.StoragePassword, 0, job.params.ClusterMeta.ClusterType, time.Second*10)
if err != nil {
job.runtime.Logger.Warn("[%s] conntect 2 old master failed :%+v", oldMasterAddr, err)
return err
}
defer oldMasterConn.Close()

if err := oldMasterConn.SelectDB(1); err != nil {
job.runtime.Logger.Warn("[%s] select db 1 on master failed :%+v", oldMasterAddr, err)
return err
}
job.runtime.Logger.Info("[%s] try reset dbmon heartbeat key [%s] value 2 current", oldMasterAddr, fmt.Sprintf("%s:time", oldMasterAddr))
rst := oldMasterConn.InstanceClient.Set(context.TODO(), fmt.Sprintf("%s:time", oldMasterAddr), time.Now().Unix(), 0)
if rst.Err() != nil {
job.runtime.Logger.Warn("[%s] set on old master failed :%+v", oldMasterAddr, err)
return err
}
return nil
}

// precheckStorageLogin make sure all todo switch redis can login
func (job *RedisSwitchPreCheck) precheckStorageLogin() error {
wg := &sync.WaitGroup{}
Expand Down
8 changes: 4 additions & 4 deletions dbm-ui/backend/flow/consts.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -57,10 +57,10 @@
DEFAULT_REDIS_INSTANCE_DATABASES = 2
DEFAULT_REDIS_CLUSTER_DATABASES = 1

# 切换时， 默认允许多久心跳
DEFAULT_MASTER_DIFF_TIME = 61
# 切换时， 允许多少秒丢失
DEFAULT_LAST_IO_SECOND_AGO = 100
# 切换时， 默认允许多久心跳 61->11 @20251121
DEFAULT_MASTER_DIFF_TIME = 11
# 切换时， 允许多少秒丢失 100->30 @20251121
DEFAULT_LAST_IO_SECOND_AGO = 30

# 默认Riak端口
DEFAULT_RIAK_PORT = 8087
Expand Down