fix: workaround problems with udevd races

When `udevd` rescans block device partitions while Talos is doing partitions, it might be that Talos can hit the following error while trying to open/mount a partition: ``` no such device or address ``` Previous attempts to fix that were using `ENODEV`, while the proper code is `ENXIO`. Also take exclusive lock while working with user disks to prevent concurrent udevd rescan. Signed-off-by: Andrey Smirnov <[email protected]>
siderolabs · May 24, 2024 · 92a274e · 92a274e
1 parent 31b24ea
commit 92a274e
Show file tree

Hide file tree

Showing 3 changed files with 4 additions and 4 deletions.
diff --git a/cmd/installer/pkg/install/install.go b/cmd/installer/pkg/install/install.go
@@ -445,7 +445,7 @@ func retryBlockdeviceOpen(device string) (*blockdevice.BlockDevice, error) {
 		switch {
 		case os.IsNotExist(openErr):
 			return retry.ExpectedError(openErr)
-		case errors.Is(openErr, syscall.ENODEV):
+		case errors.Is(openErr, syscall.ENODEV), errors.Is(openErr, syscall.ENXIO):
 			return retry.ExpectedError(openErr)
 		default:
 			return nil

diff --git a/internal/app/machined/pkg/runtime/v1alpha1/v1alpha1_sequencer_tasks.go b/internal/app/machined/pkg/runtime/v1alpha1/v1alpha1_sequencer_tasks.go
@@ -850,7 +850,7 @@ func partitionAndFormatDisks(logger *log.Logger, r runtime.Runtime) error {
 
 	for _, disk := range r.Config().Machine().Disks() {
 		if err := func() error {
-			bd, err := blockdevice.Open(disk.Device(), blockdevice.WithMode(blockdevice.ReadonlyMode))
+			bd, err := blockdevice.Open(disk.Device(), blockdevice.WithMode(blockdevice.ReadonlyMode), blockdevice.WithExclusiveLock(true))
 			if err != nil {
 				return err
 			}
@@ -921,7 +921,7 @@ func mountDisks(logger *log.Logger, r runtime.Runtime) (err error) {
 	mountpoints := mount.NewMountPoints()
 
 	for _, disk := range r.Config().Machine().Disks() {
-		bd, err := blockdevice.Open(disk.Device(), blockdevice.WithMode(blockdevice.ReadonlyMode))
+		bd, err := blockdevice.Open(disk.Device(), blockdevice.WithMode(blockdevice.ReadonlyMode), blockdevice.WithExclusiveLock(true))
 		if err != nil {
 			return err
 		}

diff --git a/internal/pkg/mount/mount.go b/internal/pkg/mount/mount.go
@@ -148,7 +148,7 @@ func mountRetry(f RetryFunc, p *Point, isUnmount bool) (err error) {
 			switch err {
 			case unix.EBUSY:
 				return retry.ExpectedError(err)
-			case unix.ENOENT, unix.ENODEV:
+			case unix.ENOENT, unix.ENXIO:
 				// if udevd triggers BLKRRPART ioctl, partition device entry might disappear temporarily
 				return retry.ExpectedError(err)
 			case unix.EUCLEAN, unix.EIO: