Skip to content
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.

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

Add stopBalancer timeout; Retry setBalancerState; Improve logging #1929

Merged
merged 1 commit into from
Sep 22, 2023
Merged
Changes from all commits
Commits
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
74 changes: 65 additions & 9 deletions pkg/backup.go
Original file line number Diff line number Diff line change
Expand Up @@ -607,6 +607,7 @@ func getPrimaryNSecondaryMember(mongoDSN string) (primary, secondary string, err
}

if secHost != primary {
klog.Infof("Primary %s & Secondary %s found for mongoDSN %s \n", primary, secHost, mongoDSN)
return primary, secHost, nil
}
}
Expand All @@ -623,10 +624,18 @@ func disabelBalancer(mongosHost string) error {
"config",
"--host", mongosHost,
"--quiet",
"--eval", "JSON.stringify(sh.stopBalancer())",
"--eval", "JSON.stringify(sh.stopBalancer(600000,1000))",
}, mongoCreds...)
// disable balancer
if err := sh.Command(MongoCMD, args...).Command("/usr/bin/tail", "-1").UnmarshalJSON(&v); err != nil {
output, err := sh.Command(MongoCMD, args...).Output()
if err != nil {
klog.Errorf("Error while stopping balancer : %s ; output : %s \n", err.Error(), output)
return err
}

err = json.Unmarshal(output, &v)
if err != nil {
klog.Errorf("Unmarshal error while stopping balancer : %s ; output = %s \n", err.Error(), output)
return err
}

Expand All @@ -642,8 +651,10 @@ func disabelBalancer(mongosHost string) error {
"--eval", "while(sh.isBalancerRunning()){ print('waiting for balancer to stop...'); sleep(1000);}",
}, mongoCreds...)
if err := sh.Command(MongoCMD, args...).Command("/usr/bin/tail", "-1").Run(); err != nil {
klog.Errorf("Error while waiting for the balancer to stop : %s \n", err.Error())
return err
}
klog.Info("Balancer successfully Disabled.")
return nil
}

Expand All @@ -659,14 +670,33 @@ func enableBalancer(mongosHost string) error {
"--quiet",
"--eval", "JSON.stringify(sh.setBalancerState(true))",
}, mongoCreds...)
if err := sh.Command(MongoCMD, args...).Command("/usr/bin/tail", "-1").UnmarshalJSON(&v); err != nil {

var (
output []byte
err error
)
cmd := sh.Command(MongoCMD, args...)
for i := 0; i < 10; i++ {
output, err = cmd.Output()
if err != nil {
klog.Errorf("Try #%d : Error on setBalancerState command : %s, output : %s .\n", i, err.Error(), output)
time.Sleep(time.Second)
} else {
break
}
}

err = json.Unmarshal(output, &v)
if err != nil {
klog.Errorf("Unmarshal error while enabling balancer : %+v , output : %s \n", err.Error(), output)
return err
}

if val, ok := v["ok"].(float64); !ok || int(val) != 1 {
return fmt.Errorf("unable to disable balancer. got response: %v", v)
return fmt.Errorf("unable to enable balancer. got response: %v", v)
}

klog.Info("Balancer successfully re-enabled.")
return nil
}

Expand All @@ -685,7 +715,16 @@ func lockConfigServer(configSVRDSN, secondaryHost string) error {
"--quiet",
"--eval", "db.BackupControl.findAndModify({query: { _id: 'BackupControlDocument' }, update: { $inc: { counter : 1 } }, new: true, upsert: true, writeConcern: { w: 'majority', wtimeout: 15000 }});",
}, mongoCreds...)
if err := sh.Command(MongoCMD, args...).Command("/usr/bin/tail", "-1").UnmarshalJSON(&v); err != nil {

output, err := sh.Command(MongoCMD, args...).Output()
if err != nil {
klog.Errorf("Error while running findAndModify to lock configServer : %s ; output : %s \n", err.Error(), output)
return err
}

err = json.Unmarshal(output, &v)
if err != nil {
klog.Errorf("Unmarshal error while running findAndModify to lock configServer : %s \n", err.Error())
return err
}
val, ok := v["counter"].(float64)
Expand Down Expand Up @@ -739,14 +778,23 @@ func lockSecondaryMember(mongohost string) error {
"--quiet",
"--eval", "JSON.stringify(db.fsyncLock())",
}, mongoCreds...)
if err := sh.Command(MongoCMD, args...).Command("/usr/bin/tail", "-1").UnmarshalJSON(&v); err != nil {

output, err := sh.Command(MongoCMD, args...).Output()
if err != nil {
klog.Errorf("Error while running fsyncLock on secondary : %s ; output : %s \n", err.Error(), output)
return err
}

err = json.Unmarshal(output, &v)
if err != nil {
klog.Errorf("Unmarshal error while running fsyncLock on secondary : %s \n", err.Error())
return err
}

if val, ok := v["ok"].(float64); !ok || int(val) != 1 {
return fmt.Errorf("unable to lock the secondary host. got response: %v", v)
}

klog.Infof("secondary %s locked.", mongohost)
return nil
}

Expand All @@ -765,13 +813,21 @@ func unlockSecondaryMember(mongohost string) error {
"--quiet",
"--eval", "JSON.stringify(db.fsyncUnlock())",
}, mongoCreds...)
if err := sh.Command(MongoCMD, args...).Command("/usr/bin/tail", "-1").UnmarshalJSON(&v); err != nil {

output, err := sh.Command(MongoCMD, args...).Output()
if err != nil {
klog.Errorf("Error while running fsyncUnlock on secondary : %s ; output : %s \n", err.Error(), output)
return err
}
err = json.Unmarshal(output, &v)
if err != nil {
klog.Errorf("Unmarshal error while running fsyncUnlock on secondary : %s \n", err.Error())
return err
}

if val, ok := v["ok"].(float64); !ok || int(val) != 1 {
return fmt.Errorf("unable to lock the secondary host. got response: %v", v)
}

klog.Infof("secondary %s unlocked.", mongohost)
return nil
}
Loading