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

[release-1.30] fix: refine check disk lun collision logic #2240

Merged
Show file tree
Hide file tree
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
14 changes: 9 additions & 5 deletions pkg/azuredisk/azuredisk.go
Original file line number Diff line number Diff line change
Expand Up @@ -114,6 +114,8 @@ type Driver struct {
volumeLocks *volumehelper.VolumeLocks
// a timed cache for throttling
throttlingCache azcache.Resource
// a timed cache for disk lun collision check throttling
checkDiskLunThrottlingCache azcache.Resource
}

// newDriverV1 Creates a NewCSIDriver object. Assumes vendor version is equal to driver version &
Expand Down Expand Up @@ -160,13 +162,15 @@ func newDriverV1(options *DriverOptions) *Driver {
}
topologyKey = fmt.Sprintf("topology.%s/zone", driver.Name)

cache, err := azcache.NewTimedCache(5*time.Minute, func(key string) (interface{}, error) {
return nil, nil
}, false)
if err != nil {
getter := func(key string) (interface{}, error) { return nil, nil }
var err error
if driver.throttlingCache, err = azcache.NewTimedCache(5*time.Minute, getter, false); err != nil {
klog.Fatalf("%v", err)
}
driver.throttlingCache = cache
if driver.checkDiskLunThrottlingCache, err = azcache.NewTimedCache(30*time.Minute, getter, false); err != nil {
klog.Fatalf("%v", err)
}

userAgent := GetUserAgent(driver.Name, driver.customUserAgent, driver.userAgentSuffix)
klog.V(2).Infof("driver userAgent: %s", userAgent)

Expand Down
8 changes: 6 additions & 2 deletions pkg/azuredisk/controllerserver.go
Original file line number Diff line number Diff line change
Expand Up @@ -592,6 +592,10 @@ func (d *Driver) ValidateVolumeCapabilities(ctx context.Context, req *csi.Valida
func (d *Driver) getOccupiedLunsFromNode(ctx context.Context, nodeName types.NodeName, diskURI string) []int {
var occupiedLuns []int
if d.checkDiskLUNCollision && !d.isCheckDiskLunThrottled() {
timer := time.AfterFunc(checkDiskLunThrottleLatency, func() {
klog.Warningf("checkDiskLun(%s) on node %s took longer than %v, disable disk lun check temporarily", diskURI, nodeName, checkDiskLunThrottleLatency)
d.checkDiskLunThrottlingCache.Set(consts.CheckDiskLunThrottlingKey, "")
})
now := time.Now()
if usedLunsFromVA, err := d.getUsedLunsFromVolumeAttachments(ctx, string(nodeName)); err == nil {
if len(usedLunsFromVA) > 0 {
Expand All @@ -611,9 +615,9 @@ func (d *Driver) getOccupiedLunsFromNode(ctx context.Context, nodeName types.Nod
}
latency := time.Since(now)
if latency > checkDiskLunThrottleLatency {
klog.Warningf("checkDiskLun(%s) on node %s took %v (limit: %v), disable disk lun check temporarily", diskURI, nodeName, latency, checkDiskLunThrottleLatency)
d.throttlingCache.Set(consts.CheckDiskLunThrottlingKey, "")
klog.Warningf("checkDiskLun(%s) on node %s took %v (limit: %v)", diskURI, nodeName, latency, checkDiskLunThrottleLatency)
} else {
timer.Stop() // cancel the timer
klog.V(6).Infof("checkDiskLun(%s) on node %s took %v", diskURI, nodeName, latency)
}
}
Expand Down
1 change: 1 addition & 0 deletions pkg/azuredisk/fake_azuredisk.go
Original file line number Diff line number Diff line change
Expand Up @@ -139,6 +139,7 @@ func newFakeDriverV1(ctrl *gomock.Controller) (*fakeDriverV1, error) {
return nil, err
}
driver.throttlingCache = cache
driver.checkDiskLunThrottlingCache = cache
driver.deviceHelper = mockoptimization.NewMockInterface(ctrl)

driver.AddControllerServiceCapabilities(
Expand Down
Loading