ceph-csi/internal/cephfs/fsjournal.go

384 lines
12 KiB
Go
Raw Normal View History

Make CephFS plugin stateless reusing RADOS based journal scheme This is a part of the stateless set of commits for CephCSI. This commit removes the dependency on config maps to store cephFS provisioned volumes, and instead relies on RADOS based objects and keys, and required CSI VolumeID encoding to detect the provisioned volumes. Changes: - Provide backward compatibility to provisioned volumes by older plugin versions (1.0.0 or older) - Remove Create/Delete support for statically provisioned volumes (fixes #382) - Added namespace support to RADOS OMaps and used the same to store RADOS CSI objects and keys in the CephFS metadata pool - Added support to mention fsname for CephFS provisioning (fixes #359) - Changed field name in CSI Identifier to 'location', to denote a pool or fscid - Updated mounter cache to use new scheme - Required Helm manifests are updated - Required documentation and other manifests are updated - Made driver option 'metadatastorage' as optional, as fresh installs do not need to specify the same Testing done: - Create/Mount/Delete PVC - Create/Delete 5 PVCs - Mount version 1.0.0 PVC - Delete version 1.0.0 PV - Mount Statically defined PV/PVC/Pod - Mount Statically defined version 1.0.0 PV/PVC/Pod - Delete Statically defined version 1.0.0 PV/PVC/Pod - Node restart when mounted to test mountcache - Use InstanceID other than 'default' - RBD basic round of tests, as namespace is added to OMaps - csitest against ceph-fs plugin - NOTE: CephFS plugin still does not detect and address already created volumes but of a different size - Test not providing any value to the metadata storage parameter Signed-off-by: ShyamsundarR <srangana@redhat.com>
2019-05-28 19:03:18 +00:00
/*
Copyright 2019 The Ceph-CSI 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 cephfs
import (
"context"
"errors"
"fmt"
"github.com/ceph/ceph-csi/internal/util"
"github.com/golang/protobuf/ptypes/timestamp"
Make CephFS plugin stateless reusing RADOS based journal scheme This is a part of the stateless set of commits for CephCSI. This commit removes the dependency on config maps to store cephFS provisioned volumes, and instead relies on RADOS based objects and keys, and required CSI VolumeID encoding to detect the provisioned volumes. Changes: - Provide backward compatibility to provisioned volumes by older plugin versions (1.0.0 or older) - Remove Create/Delete support for statically provisioned volumes (fixes #382) - Added namespace support to RADOS OMaps and used the same to store RADOS CSI objects and keys in the CephFS metadata pool - Added support to mention fsname for CephFS provisioning (fixes #359) - Changed field name in CSI Identifier to 'location', to denote a pool or fscid - Updated mounter cache to use new scheme - Required Helm manifests are updated - Required documentation and other manifests are updated - Made driver option 'metadatastorage' as optional, as fresh installs do not need to specify the same Testing done: - Create/Mount/Delete PVC - Create/Delete 5 PVCs - Mount version 1.0.0 PVC - Delete version 1.0.0 PV - Mount Statically defined PV/PVC/Pod - Mount Statically defined version 1.0.0 PV/PVC/Pod - Delete Statically defined version 1.0.0 PV/PVC/Pod - Node restart when mounted to test mountcache - Use InstanceID other than 'default' - RBD basic round of tests, as namespace is added to OMaps - csitest against ceph-fs plugin - NOTE: CephFS plugin still does not detect and address already created volumes but of a different size - Test not providing any value to the metadata storage parameter Signed-off-by: ShyamsundarR <srangana@redhat.com>
2019-05-28 19:03:18 +00:00
)
// volumeIdentifier structure contains an association between the CSI VolumeID to its subvolume
// name on the backing CephFS instance.
Make CephFS plugin stateless reusing RADOS based journal scheme This is a part of the stateless set of commits for CephCSI. This commit removes the dependency on config maps to store cephFS provisioned volumes, and instead relies on RADOS based objects and keys, and required CSI VolumeID encoding to detect the provisioned volumes. Changes: - Provide backward compatibility to provisioned volumes by older plugin versions (1.0.0 or older) - Remove Create/Delete support for statically provisioned volumes (fixes #382) - Added namespace support to RADOS OMaps and used the same to store RADOS CSI objects and keys in the CephFS metadata pool - Added support to mention fsname for CephFS provisioning (fixes #359) - Changed field name in CSI Identifier to 'location', to denote a pool or fscid - Updated mounter cache to use new scheme - Required Helm manifests are updated - Required documentation and other manifests are updated - Made driver option 'metadatastorage' as optional, as fresh installs do not need to specify the same Testing done: - Create/Mount/Delete PVC - Create/Delete 5 PVCs - Mount version 1.0.0 PVC - Delete version 1.0.0 PV - Mount Statically defined PV/PVC/Pod - Mount Statically defined version 1.0.0 PV/PVC/Pod - Delete Statically defined version 1.0.0 PV/PVC/Pod - Node restart when mounted to test mountcache - Use InstanceID other than 'default' - RBD basic round of tests, as namespace is added to OMaps - csitest against ceph-fs plugin - NOTE: CephFS plugin still does not detect and address already created volumes but of a different size - Test not providing any value to the metadata storage parameter Signed-off-by: ShyamsundarR <srangana@redhat.com>
2019-05-28 19:03:18 +00:00
type volumeIdentifier struct {
FsSubvolName string
VolumeID string
}
type snapshotIdentifier struct {
FsSnapshotName string
SnapshotID string
RequestName string
CreationTime *timestamp.Timestamp
FsSubvolName string
}
Make CephFS plugin stateless reusing RADOS based journal scheme This is a part of the stateless set of commits for CephCSI. This commit removes the dependency on config maps to store cephFS provisioned volumes, and instead relies on RADOS based objects and keys, and required CSI VolumeID encoding to detect the provisioned volumes. Changes: - Provide backward compatibility to provisioned volumes by older plugin versions (1.0.0 or older) - Remove Create/Delete support for statically provisioned volumes (fixes #382) - Added namespace support to RADOS OMaps and used the same to store RADOS CSI objects and keys in the CephFS metadata pool - Added support to mention fsname for CephFS provisioning (fixes #359) - Changed field name in CSI Identifier to 'location', to denote a pool or fscid - Updated mounter cache to use new scheme - Required Helm manifests are updated - Required documentation and other manifests are updated - Made driver option 'metadatastorage' as optional, as fresh installs do not need to specify the same Testing done: - Create/Mount/Delete PVC - Create/Delete 5 PVCs - Mount version 1.0.0 PVC - Delete version 1.0.0 PV - Mount Statically defined PV/PVC/Pod - Mount Statically defined version 1.0.0 PV/PVC/Pod - Delete Statically defined version 1.0.0 PV/PVC/Pod - Node restart when mounted to test mountcache - Use InstanceID other than 'default' - RBD basic round of tests, as namespace is added to OMaps - csitest against ceph-fs plugin - NOTE: CephFS plugin still does not detect and address already created volumes but of a different size - Test not providing any value to the metadata storage parameter Signed-off-by: ShyamsundarR <srangana@redhat.com>
2019-05-28 19:03:18 +00:00
/*
checkVolExists checks to determine if passed in RequestName in volOptions exists on the backend.
**NOTE:** These functions manipulate the rados omaps that hold information regarding
volume names as requested by the CSI drivers. Hence, these need to be invoked only when the
respective CSI driver generated volume name based locks are held, as otherwise racy
access to these omaps may end up leaving them in an inconsistent state.
These functions also cleanup omap reservations that are stale. I.e when omap entries exist and
backing subvolumes are missing, or one of the omaps exist and the next is missing. This is
because, the order of omap creation and deletion are inverse of each other, and protected by the
request name lock, and hence any stale omaps are leftovers from incomplete transactions and are
hence safe to garbage collect.
*/
// nolint:gocognit:gocyclo // TODO: reduce complexity
func checkVolExists(ctx context.Context,
volOptions,
parentVolOpt *volumeOptions,
pvID *volumeIdentifier,
sID *snapshotIdentifier,
cr *util.Credentials) (*volumeIdentifier, error) {
var vid volumeIdentifier
// Connect to cephfs' default radosNamespace (csi)
j, err := volJournal.Connect(volOptions.Monitors, radosNamespace, cr)
if err != nil {
return nil, err
}
defer j.Destroy()
imageData, err := j.CheckReservation(
ctx, volOptions.MetadataPool, volOptions.RequestName, volOptions.NamePrefix, "", "")
Make CephFS plugin stateless reusing RADOS based journal scheme This is a part of the stateless set of commits for CephCSI. This commit removes the dependency on config maps to store cephFS provisioned volumes, and instead relies on RADOS based objects and keys, and required CSI VolumeID encoding to detect the provisioned volumes. Changes: - Provide backward compatibility to provisioned volumes by older plugin versions (1.0.0 or older) - Remove Create/Delete support for statically provisioned volumes (fixes #382) - Added namespace support to RADOS OMaps and used the same to store RADOS CSI objects and keys in the CephFS metadata pool - Added support to mention fsname for CephFS provisioning (fixes #359) - Changed field name in CSI Identifier to 'location', to denote a pool or fscid - Updated mounter cache to use new scheme - Required Helm manifests are updated - Required documentation and other manifests are updated - Made driver option 'metadatastorage' as optional, as fresh installs do not need to specify the same Testing done: - Create/Mount/Delete PVC - Create/Delete 5 PVCs - Mount version 1.0.0 PVC - Delete version 1.0.0 PV - Mount Statically defined PV/PVC/Pod - Mount Statically defined version 1.0.0 PV/PVC/Pod - Delete Statically defined version 1.0.0 PV/PVC/Pod - Node restart when mounted to test mountcache - Use InstanceID other than 'default' - RBD basic round of tests, as namespace is added to OMaps - csitest against ceph-fs plugin - NOTE: CephFS plugin still does not detect and address already created volumes but of a different size - Test not providing any value to the metadata storage parameter Signed-off-by: ShyamsundarR <srangana@redhat.com>
2019-05-28 19:03:18 +00:00
if err != nil {
return nil, err
}
if imageData == nil {
Make CephFS plugin stateless reusing RADOS based journal scheme This is a part of the stateless set of commits for CephCSI. This commit removes the dependency on config maps to store cephFS provisioned volumes, and instead relies on RADOS based objects and keys, and required CSI VolumeID encoding to detect the provisioned volumes. Changes: - Provide backward compatibility to provisioned volumes by older plugin versions (1.0.0 or older) - Remove Create/Delete support for statically provisioned volumes (fixes #382) - Added namespace support to RADOS OMaps and used the same to store RADOS CSI objects and keys in the CephFS metadata pool - Added support to mention fsname for CephFS provisioning (fixes #359) - Changed field name in CSI Identifier to 'location', to denote a pool or fscid - Updated mounter cache to use new scheme - Required Helm manifests are updated - Required documentation and other manifests are updated - Made driver option 'metadatastorage' as optional, as fresh installs do not need to specify the same Testing done: - Create/Mount/Delete PVC - Create/Delete 5 PVCs - Mount version 1.0.0 PVC - Delete version 1.0.0 PV - Mount Statically defined PV/PVC/Pod - Mount Statically defined version 1.0.0 PV/PVC/Pod - Delete Statically defined version 1.0.0 PV/PVC/Pod - Node restart when mounted to test mountcache - Use InstanceID other than 'default' - RBD basic round of tests, as namespace is added to OMaps - csitest against ceph-fs plugin - NOTE: CephFS plugin still does not detect and address already created volumes but of a different size - Test not providing any value to the metadata storage parameter Signed-off-by: ShyamsundarR <srangana@redhat.com>
2019-05-28 19:03:18 +00:00
return nil, nil
}
imageUUID := imageData.ImageUUID
vid.FsSubvolName = imageData.ImageAttributes.ImageName
Make CephFS plugin stateless reusing RADOS based journal scheme This is a part of the stateless set of commits for CephCSI. This commit removes the dependency on config maps to store cephFS provisioned volumes, and instead relies on RADOS based objects and keys, and required CSI VolumeID encoding to detect the provisioned volumes. Changes: - Provide backward compatibility to provisioned volumes by older plugin versions (1.0.0 or older) - Remove Create/Delete support for statically provisioned volumes (fixes #382) - Added namespace support to RADOS OMaps and used the same to store RADOS CSI objects and keys in the CephFS metadata pool - Added support to mention fsname for CephFS provisioning (fixes #359) - Changed field name in CSI Identifier to 'location', to denote a pool or fscid - Updated mounter cache to use new scheme - Required Helm manifests are updated - Required documentation and other manifests are updated - Made driver option 'metadatastorage' as optional, as fresh installs do not need to specify the same Testing done: - Create/Mount/Delete PVC - Create/Delete 5 PVCs - Mount version 1.0.0 PVC - Delete version 1.0.0 PV - Mount Statically defined PV/PVC/Pod - Mount Statically defined version 1.0.0 PV/PVC/Pod - Delete Statically defined version 1.0.0 PV/PVC/Pod - Node restart when mounted to test mountcache - Use InstanceID other than 'default' - RBD basic round of tests, as namespace is added to OMaps - csitest against ceph-fs plugin - NOTE: CephFS plugin still does not detect and address already created volumes but of a different size - Test not providing any value to the metadata storage parameter Signed-off-by: ShyamsundarR <srangana@redhat.com>
2019-05-28 19:03:18 +00:00
if sID != nil || pvID != nil {
cloneState, cloneStateErr := volOptions.getCloneState(ctx, volumeID(vid.FsSubvolName))
if cloneStateErr != nil {
if errors.Is(cloneStateErr, ErrVolumeNotFound) {
if pvID != nil {
err = cleanupCloneFromSubvolumeSnapshot(
ctx, volumeID(pvID.FsSubvolName),
volumeID(vid.FsSubvolName),
parentVolOpt)
if err != nil {
return nil, err
}
}
err = j.UndoReservation(ctx, volOptions.MetadataPool,
volOptions.MetadataPool, vid.FsSubvolName, volOptions.RequestName)
return nil, err
}
return nil, err
}
if cloneState == cephFSCloneInprogress {
return nil, ErrCloneInProgress
}
if cloneState == cephFSClonePending {
return nil, ErrClonePending
}
if cloneState == cephFSCloneFailed {
err = purgeVolume(ctx, volumeID(vid.FsSubvolName), cr, volOptions, true)
if err != nil {
util.ErrorLog(ctx, "failed to delete volume %s: %v", vid.FsSubvolName, err)
return nil, err
}
if pvID != nil {
err = cleanupCloneFromSubvolumeSnapshot(
ctx, volumeID(pvID.FsSubvolName),
volumeID(vid.FsSubvolName),
parentVolOpt)
if err != nil {
return nil, err
}
}
err = j.UndoReservation(ctx, volOptions.MetadataPool,
volOptions.MetadataPool, vid.FsSubvolName, volOptions.RequestName)
return nil, err
}
if cloneState != cephFSCloneComplete {
return nil, fmt.Errorf("clone is not in complete state for %s", vid.FsSubvolName)
}
} else {
_, err = volOptions.getVolumeRootPathCeph(ctx, volumeID(vid.FsSubvolName))
if err != nil {
if errors.Is(err, ErrVolumeNotFound) {
err = j.UndoReservation(ctx, volOptions.MetadataPool,
volOptions.MetadataPool, vid.FsSubvolName, volOptions.RequestName)
return nil, err
}
return nil, err
}
}
// check if topology constraints match what is found
// TODO: we need an API to fetch subvolume attributes (size/datapool and others), based
// on which we can evaluate which topology this belongs to.
// TODO: CephFS topology support is postponed till we get the same
Make CephFS plugin stateless reusing RADOS based journal scheme This is a part of the stateless set of commits for CephCSI. This commit removes the dependency on config maps to store cephFS provisioned volumes, and instead relies on RADOS based objects and keys, and required CSI VolumeID encoding to detect the provisioned volumes. Changes: - Provide backward compatibility to provisioned volumes by older plugin versions (1.0.0 or older) - Remove Create/Delete support for statically provisioned volumes (fixes #382) - Added namespace support to RADOS OMaps and used the same to store RADOS CSI objects and keys in the CephFS metadata pool - Added support to mention fsname for CephFS provisioning (fixes #359) - Changed field name in CSI Identifier to 'location', to denote a pool or fscid - Updated mounter cache to use new scheme - Required Helm manifests are updated - Required documentation and other manifests are updated - Made driver option 'metadatastorage' as optional, as fresh installs do not need to specify the same Testing done: - Create/Mount/Delete PVC - Create/Delete 5 PVCs - Mount version 1.0.0 PVC - Delete version 1.0.0 PV - Mount Statically defined PV/PVC/Pod - Mount Statically defined version 1.0.0 PV/PVC/Pod - Delete Statically defined version 1.0.0 PV/PVC/Pod - Node restart when mounted to test mountcache - Use InstanceID other than 'default' - RBD basic round of tests, as namespace is added to OMaps - csitest against ceph-fs plugin - NOTE: CephFS plugin still does not detect and address already created volumes but of a different size - Test not providing any value to the metadata storage parameter Signed-off-by: ShyamsundarR <srangana@redhat.com>
2019-05-28 19:03:18 +00:00
// TODO: size checks
// found a volume already available, process and return it!
vid.VolumeID, err = util.GenerateVolID(ctx, volOptions.Monitors, cr, volOptions.FscID,
"", volOptions.ClusterID, imageUUID, volIDVersion)
Make CephFS plugin stateless reusing RADOS based journal scheme This is a part of the stateless set of commits for CephCSI. This commit removes the dependency on config maps to store cephFS provisioned volumes, and instead relies on RADOS based objects and keys, and required CSI VolumeID encoding to detect the provisioned volumes. Changes: - Provide backward compatibility to provisioned volumes by older plugin versions (1.0.0 or older) - Remove Create/Delete support for statically provisioned volumes (fixes #382) - Added namespace support to RADOS OMaps and used the same to store RADOS CSI objects and keys in the CephFS metadata pool - Added support to mention fsname for CephFS provisioning (fixes #359) - Changed field name in CSI Identifier to 'location', to denote a pool or fscid - Updated mounter cache to use new scheme - Required Helm manifests are updated - Required documentation and other manifests are updated - Made driver option 'metadatastorage' as optional, as fresh installs do not need to specify the same Testing done: - Create/Mount/Delete PVC - Create/Delete 5 PVCs - Mount version 1.0.0 PVC - Delete version 1.0.0 PV - Mount Statically defined PV/PVC/Pod - Mount Statically defined version 1.0.0 PV/PVC/Pod - Delete Statically defined version 1.0.0 PV/PVC/Pod - Node restart when mounted to test mountcache - Use InstanceID other than 'default' - RBD basic round of tests, as namespace is added to OMaps - csitest against ceph-fs plugin - NOTE: CephFS plugin still does not detect and address already created volumes but of a different size - Test not providing any value to the metadata storage parameter Signed-off-by: ShyamsundarR <srangana@redhat.com>
2019-05-28 19:03:18 +00:00
if err != nil {
return nil, err
}
util.DebugLog(ctx, "Found existing volume (%s) with subvolume name (%s) for request (%s)",
Make CephFS plugin stateless reusing RADOS based journal scheme This is a part of the stateless set of commits for CephCSI. This commit removes the dependency on config maps to store cephFS provisioned volumes, and instead relies on RADOS based objects and keys, and required CSI VolumeID encoding to detect the provisioned volumes. Changes: - Provide backward compatibility to provisioned volumes by older plugin versions (1.0.0 or older) - Remove Create/Delete support for statically provisioned volumes (fixes #382) - Added namespace support to RADOS OMaps and used the same to store RADOS CSI objects and keys in the CephFS metadata pool - Added support to mention fsname for CephFS provisioning (fixes #359) - Changed field name in CSI Identifier to 'location', to denote a pool or fscid - Updated mounter cache to use new scheme - Required Helm manifests are updated - Required documentation and other manifests are updated - Made driver option 'metadatastorage' as optional, as fresh installs do not need to specify the same Testing done: - Create/Mount/Delete PVC - Create/Delete 5 PVCs - Mount version 1.0.0 PVC - Delete version 1.0.0 PV - Mount Statically defined PV/PVC/Pod - Mount Statically defined version 1.0.0 PV/PVC/Pod - Delete Statically defined version 1.0.0 PV/PVC/Pod - Node restart when mounted to test mountcache - Use InstanceID other than 'default' - RBD basic round of tests, as namespace is added to OMaps - csitest against ceph-fs plugin - NOTE: CephFS plugin still does not detect and address already created volumes but of a different size - Test not providing any value to the metadata storage parameter Signed-off-by: ShyamsundarR <srangana@redhat.com>
2019-05-28 19:03:18 +00:00
vid.VolumeID, vid.FsSubvolName, volOptions.RequestName)
if parentVolOpt != nil && pvID != nil {
err = cleanupCloneFromSubvolumeSnapshot(ctx, volumeID(pvID.FsSubvolName), volumeID(vid.FsSubvolName), parentVolOpt)
if err != nil {
return nil, err
}
}
Make CephFS plugin stateless reusing RADOS based journal scheme This is a part of the stateless set of commits for CephCSI. This commit removes the dependency on config maps to store cephFS provisioned volumes, and instead relies on RADOS based objects and keys, and required CSI VolumeID encoding to detect the provisioned volumes. Changes: - Provide backward compatibility to provisioned volumes by older plugin versions (1.0.0 or older) - Remove Create/Delete support for statically provisioned volumes (fixes #382) - Added namespace support to RADOS OMaps and used the same to store RADOS CSI objects and keys in the CephFS metadata pool - Added support to mention fsname for CephFS provisioning (fixes #359) - Changed field name in CSI Identifier to 'location', to denote a pool or fscid - Updated mounter cache to use new scheme - Required Helm manifests are updated - Required documentation and other manifests are updated - Made driver option 'metadatastorage' as optional, as fresh installs do not need to specify the same Testing done: - Create/Mount/Delete PVC - Create/Delete 5 PVCs - Mount version 1.0.0 PVC - Delete version 1.0.0 PV - Mount Statically defined PV/PVC/Pod - Mount Statically defined version 1.0.0 PV/PVC/Pod - Delete Statically defined version 1.0.0 PV/PVC/Pod - Node restart when mounted to test mountcache - Use InstanceID other than 'default' - RBD basic round of tests, as namespace is added to OMaps - csitest against ceph-fs plugin - NOTE: CephFS plugin still does not detect and address already created volumes but of a different size - Test not providing any value to the metadata storage parameter Signed-off-by: ShyamsundarR <srangana@redhat.com>
2019-05-28 19:03:18 +00:00
return &vid, nil
}
// undoVolReservation is a helper routine to undo a name reservation for a CSI VolumeName.
func undoVolReservation(ctx context.Context, volOptions *volumeOptions, vid volumeIdentifier, secret map[string]string) error {
cr, err := util.NewAdminCredentials(secret)
Make CephFS plugin stateless reusing RADOS based journal scheme This is a part of the stateless set of commits for CephCSI. This commit removes the dependency on config maps to store cephFS provisioned volumes, and instead relies on RADOS based objects and keys, and required CSI VolumeID encoding to detect the provisioned volumes. Changes: - Provide backward compatibility to provisioned volumes by older plugin versions (1.0.0 or older) - Remove Create/Delete support for statically provisioned volumes (fixes #382) - Added namespace support to RADOS OMaps and used the same to store RADOS CSI objects and keys in the CephFS metadata pool - Added support to mention fsname for CephFS provisioning (fixes #359) - Changed field name in CSI Identifier to 'location', to denote a pool or fscid - Updated mounter cache to use new scheme - Required Helm manifests are updated - Required documentation and other manifests are updated - Made driver option 'metadatastorage' as optional, as fresh installs do not need to specify the same Testing done: - Create/Mount/Delete PVC - Create/Delete 5 PVCs - Mount version 1.0.0 PVC - Delete version 1.0.0 PV - Mount Statically defined PV/PVC/Pod - Mount Statically defined version 1.0.0 PV/PVC/Pod - Delete Statically defined version 1.0.0 PV/PVC/Pod - Node restart when mounted to test mountcache - Use InstanceID other than 'default' - RBD basic round of tests, as namespace is added to OMaps - csitest against ceph-fs plugin - NOTE: CephFS plugin still does not detect and address already created volumes but of a different size - Test not providing any value to the metadata storage parameter Signed-off-by: ShyamsundarR <srangana@redhat.com>
2019-05-28 19:03:18 +00:00
if err != nil {
return err
}
defer cr.DeleteCredentials()
Make CephFS plugin stateless reusing RADOS based journal scheme This is a part of the stateless set of commits for CephCSI. This commit removes the dependency on config maps to store cephFS provisioned volumes, and instead relies on RADOS based objects and keys, and required CSI VolumeID encoding to detect the provisioned volumes. Changes: - Provide backward compatibility to provisioned volumes by older plugin versions (1.0.0 or older) - Remove Create/Delete support for statically provisioned volumes (fixes #382) - Added namespace support to RADOS OMaps and used the same to store RADOS CSI objects and keys in the CephFS metadata pool - Added support to mention fsname for CephFS provisioning (fixes #359) - Changed field name in CSI Identifier to 'location', to denote a pool or fscid - Updated mounter cache to use new scheme - Required Helm manifests are updated - Required documentation and other manifests are updated - Made driver option 'metadatastorage' as optional, as fresh installs do not need to specify the same Testing done: - Create/Mount/Delete PVC - Create/Delete 5 PVCs - Mount version 1.0.0 PVC - Delete version 1.0.0 PV - Mount Statically defined PV/PVC/Pod - Mount Statically defined version 1.0.0 PV/PVC/Pod - Delete Statically defined version 1.0.0 PV/PVC/Pod - Node restart when mounted to test mountcache - Use InstanceID other than 'default' - RBD basic round of tests, as namespace is added to OMaps - csitest against ceph-fs plugin - NOTE: CephFS plugin still does not detect and address already created volumes but of a different size - Test not providing any value to the metadata storage parameter Signed-off-by: ShyamsundarR <srangana@redhat.com>
2019-05-28 19:03:18 +00:00
// Connect to cephfs' default radosNamespace (csi)
j, err := volJournal.Connect(volOptions.Monitors, radosNamespace, cr)
if err != nil {
return err
}
defer j.Destroy()
err = j.UndoReservation(ctx, volOptions.MetadataPool,
volOptions.MetadataPool, vid.FsSubvolName, volOptions.RequestName)
Make CephFS plugin stateless reusing RADOS based journal scheme This is a part of the stateless set of commits for CephCSI. This commit removes the dependency on config maps to store cephFS provisioned volumes, and instead relies on RADOS based objects and keys, and required CSI VolumeID encoding to detect the provisioned volumes. Changes: - Provide backward compatibility to provisioned volumes by older plugin versions (1.0.0 or older) - Remove Create/Delete support for statically provisioned volumes (fixes #382) - Added namespace support to RADOS OMaps and used the same to store RADOS CSI objects and keys in the CephFS metadata pool - Added support to mention fsname for CephFS provisioning (fixes #359) - Changed field name in CSI Identifier to 'location', to denote a pool or fscid - Updated mounter cache to use new scheme - Required Helm manifests are updated - Required documentation and other manifests are updated - Made driver option 'metadatastorage' as optional, as fresh installs do not need to specify the same Testing done: - Create/Mount/Delete PVC - Create/Delete 5 PVCs - Mount version 1.0.0 PVC - Delete version 1.0.0 PV - Mount Statically defined PV/PVC/Pod - Mount Statically defined version 1.0.0 PV/PVC/Pod - Delete Statically defined version 1.0.0 PV/PVC/Pod - Node restart when mounted to test mountcache - Use InstanceID other than 'default' - RBD basic round of tests, as namespace is added to OMaps - csitest against ceph-fs plugin - NOTE: CephFS plugin still does not detect and address already created volumes but of a different size - Test not providing any value to the metadata storage parameter Signed-off-by: ShyamsundarR <srangana@redhat.com>
2019-05-28 19:03:18 +00:00
return err
}
func updateTopologyConstraints(volOpts *volumeOptions) error {
// update request based on topology constrained parameters (if present)
poolName, _, topology, err := util.FindPoolAndTopology(volOpts.TopologyPools, volOpts.TopologyRequirement)
if err != nil {
return err
}
if poolName != "" {
volOpts.Pool = poolName
volOpts.Topology = topology
}
return nil
}
Make CephFS plugin stateless reusing RADOS based journal scheme This is a part of the stateless set of commits for CephCSI. This commit removes the dependency on config maps to store cephFS provisioned volumes, and instead relies on RADOS based objects and keys, and required CSI VolumeID encoding to detect the provisioned volumes. Changes: - Provide backward compatibility to provisioned volumes by older plugin versions (1.0.0 or older) - Remove Create/Delete support for statically provisioned volumes (fixes #382) - Added namespace support to RADOS OMaps and used the same to store RADOS CSI objects and keys in the CephFS metadata pool - Added support to mention fsname for CephFS provisioning (fixes #359) - Changed field name in CSI Identifier to 'location', to denote a pool or fscid - Updated mounter cache to use new scheme - Required Helm manifests are updated - Required documentation and other manifests are updated - Made driver option 'metadatastorage' as optional, as fresh installs do not need to specify the same Testing done: - Create/Mount/Delete PVC - Create/Delete 5 PVCs - Mount version 1.0.0 PVC - Delete version 1.0.0 PV - Mount Statically defined PV/PVC/Pod - Mount Statically defined version 1.0.0 PV/PVC/Pod - Delete Statically defined version 1.0.0 PV/PVC/Pod - Node restart when mounted to test mountcache - Use InstanceID other than 'default' - RBD basic round of tests, as namespace is added to OMaps - csitest against ceph-fs plugin - NOTE: CephFS plugin still does not detect and address already created volumes but of a different size - Test not providing any value to the metadata storage parameter Signed-off-by: ShyamsundarR <srangana@redhat.com>
2019-05-28 19:03:18 +00:00
// reserveVol is a helper routine to request a UUID reservation for the CSI VolumeName and,
// to generate the volume identifier for the reserved UUID.
func reserveVol(ctx context.Context, volOptions *volumeOptions, secret map[string]string) (*volumeIdentifier, error) {
Make CephFS plugin stateless reusing RADOS based journal scheme This is a part of the stateless set of commits for CephCSI. This commit removes the dependency on config maps to store cephFS provisioned volumes, and instead relies on RADOS based objects and keys, and required CSI VolumeID encoding to detect the provisioned volumes. Changes: - Provide backward compatibility to provisioned volumes by older plugin versions (1.0.0 or older) - Remove Create/Delete support for statically provisioned volumes (fixes #382) - Added namespace support to RADOS OMaps and used the same to store RADOS CSI objects and keys in the CephFS metadata pool - Added support to mention fsname for CephFS provisioning (fixes #359) - Changed field name in CSI Identifier to 'location', to denote a pool or fscid - Updated mounter cache to use new scheme - Required Helm manifests are updated - Required documentation and other manifests are updated - Made driver option 'metadatastorage' as optional, as fresh installs do not need to specify the same Testing done: - Create/Mount/Delete PVC - Create/Delete 5 PVCs - Mount version 1.0.0 PVC - Delete version 1.0.0 PV - Mount Statically defined PV/PVC/Pod - Mount Statically defined version 1.0.0 PV/PVC/Pod - Delete Statically defined version 1.0.0 PV/PVC/Pod - Node restart when mounted to test mountcache - Use InstanceID other than 'default' - RBD basic round of tests, as namespace is added to OMaps - csitest against ceph-fs plugin - NOTE: CephFS plugin still does not detect and address already created volumes but of a different size - Test not providing any value to the metadata storage parameter Signed-off-by: ShyamsundarR <srangana@redhat.com>
2019-05-28 19:03:18 +00:00
var (
vid volumeIdentifier
imageUUID string
err error
Make CephFS plugin stateless reusing RADOS based journal scheme This is a part of the stateless set of commits for CephCSI. This commit removes the dependency on config maps to store cephFS provisioned volumes, and instead relies on RADOS based objects and keys, and required CSI VolumeID encoding to detect the provisioned volumes. Changes: - Provide backward compatibility to provisioned volumes by older plugin versions (1.0.0 or older) - Remove Create/Delete support for statically provisioned volumes (fixes #382) - Added namespace support to RADOS OMaps and used the same to store RADOS CSI objects and keys in the CephFS metadata pool - Added support to mention fsname for CephFS provisioning (fixes #359) - Changed field name in CSI Identifier to 'location', to denote a pool or fscid - Updated mounter cache to use new scheme - Required Helm manifests are updated - Required documentation and other manifests are updated - Made driver option 'metadatastorage' as optional, as fresh installs do not need to specify the same Testing done: - Create/Mount/Delete PVC - Create/Delete 5 PVCs - Mount version 1.0.0 PVC - Delete version 1.0.0 PV - Mount Statically defined PV/PVC/Pod - Mount Statically defined version 1.0.0 PV/PVC/Pod - Delete Statically defined version 1.0.0 PV/PVC/Pod - Node restart when mounted to test mountcache - Use InstanceID other than 'default' - RBD basic round of tests, as namespace is added to OMaps - csitest against ceph-fs plugin - NOTE: CephFS plugin still does not detect and address already created volumes but of a different size - Test not providing any value to the metadata storage parameter Signed-off-by: ShyamsundarR <srangana@redhat.com>
2019-05-28 19:03:18 +00:00
)
cr, err := util.NewAdminCredentials(secret)
Make CephFS plugin stateless reusing RADOS based journal scheme This is a part of the stateless set of commits for CephCSI. This commit removes the dependency on config maps to store cephFS provisioned volumes, and instead relies on RADOS based objects and keys, and required CSI VolumeID encoding to detect the provisioned volumes. Changes: - Provide backward compatibility to provisioned volumes by older plugin versions (1.0.0 or older) - Remove Create/Delete support for statically provisioned volumes (fixes #382) - Added namespace support to RADOS OMaps and used the same to store RADOS CSI objects and keys in the CephFS metadata pool - Added support to mention fsname for CephFS provisioning (fixes #359) - Changed field name in CSI Identifier to 'location', to denote a pool or fscid - Updated mounter cache to use new scheme - Required Helm manifests are updated - Required documentation and other manifests are updated - Made driver option 'metadatastorage' as optional, as fresh installs do not need to specify the same Testing done: - Create/Mount/Delete PVC - Create/Delete 5 PVCs - Mount version 1.0.0 PVC - Delete version 1.0.0 PV - Mount Statically defined PV/PVC/Pod - Mount Statically defined version 1.0.0 PV/PVC/Pod - Delete Statically defined version 1.0.0 PV/PVC/Pod - Node restart when mounted to test mountcache - Use InstanceID other than 'default' - RBD basic round of tests, as namespace is added to OMaps - csitest against ceph-fs plugin - NOTE: CephFS plugin still does not detect and address already created volumes but of a different size - Test not providing any value to the metadata storage parameter Signed-off-by: ShyamsundarR <srangana@redhat.com>
2019-05-28 19:03:18 +00:00
if err != nil {
return nil, err
}
defer cr.DeleteCredentials()
Make CephFS plugin stateless reusing RADOS based journal scheme This is a part of the stateless set of commits for CephCSI. This commit removes the dependency on config maps to store cephFS provisioned volumes, and instead relies on RADOS based objects and keys, and required CSI VolumeID encoding to detect the provisioned volumes. Changes: - Provide backward compatibility to provisioned volumes by older plugin versions (1.0.0 or older) - Remove Create/Delete support for statically provisioned volumes (fixes #382) - Added namespace support to RADOS OMaps and used the same to store RADOS CSI objects and keys in the CephFS metadata pool - Added support to mention fsname for CephFS provisioning (fixes #359) - Changed field name in CSI Identifier to 'location', to denote a pool or fscid - Updated mounter cache to use new scheme - Required Helm manifests are updated - Required documentation and other manifests are updated - Made driver option 'metadatastorage' as optional, as fresh installs do not need to specify the same Testing done: - Create/Mount/Delete PVC - Create/Delete 5 PVCs - Mount version 1.0.0 PVC - Delete version 1.0.0 PV - Mount Statically defined PV/PVC/Pod - Mount Statically defined version 1.0.0 PV/PVC/Pod - Delete Statically defined version 1.0.0 PV/PVC/Pod - Node restart when mounted to test mountcache - Use InstanceID other than 'default' - RBD basic round of tests, as namespace is added to OMaps - csitest against ceph-fs plugin - NOTE: CephFS plugin still does not detect and address already created volumes but of a different size - Test not providing any value to the metadata storage parameter Signed-off-by: ShyamsundarR <srangana@redhat.com>
2019-05-28 19:03:18 +00:00
err = updateTopologyConstraints(volOptions)
Make CephFS plugin stateless reusing RADOS based journal scheme This is a part of the stateless set of commits for CephCSI. This commit removes the dependency on config maps to store cephFS provisioned volumes, and instead relies on RADOS based objects and keys, and required CSI VolumeID encoding to detect the provisioned volumes. Changes: - Provide backward compatibility to provisioned volumes by older plugin versions (1.0.0 or older) - Remove Create/Delete support for statically provisioned volumes (fixes #382) - Added namespace support to RADOS OMaps and used the same to store RADOS CSI objects and keys in the CephFS metadata pool - Added support to mention fsname for CephFS provisioning (fixes #359) - Changed field name in CSI Identifier to 'location', to denote a pool or fscid - Updated mounter cache to use new scheme - Required Helm manifests are updated - Required documentation and other manifests are updated - Made driver option 'metadatastorage' as optional, as fresh installs do not need to specify the same Testing done: - Create/Mount/Delete PVC - Create/Delete 5 PVCs - Mount version 1.0.0 PVC - Delete version 1.0.0 PV - Mount Statically defined PV/PVC/Pod - Mount Statically defined version 1.0.0 PV/PVC/Pod - Delete Statically defined version 1.0.0 PV/PVC/Pod - Node restart when mounted to test mountcache - Use InstanceID other than 'default' - RBD basic round of tests, as namespace is added to OMaps - csitest against ceph-fs plugin - NOTE: CephFS plugin still does not detect and address already created volumes but of a different size - Test not providing any value to the metadata storage parameter Signed-off-by: ShyamsundarR <srangana@redhat.com>
2019-05-28 19:03:18 +00:00
if err != nil {
return nil, err
}
// Connect to cephfs' default radosNamespace (csi)
j, err := volJournal.Connect(volOptions.Monitors, radosNamespace, cr)
if err != nil {
return nil, err
}
defer j.Destroy()
imageUUID, vid.FsSubvolName, err = j.ReserveName(
ctx, volOptions.MetadataPool, util.InvalidPoolID,
volOptions.MetadataPool, util.InvalidPoolID, volOptions.RequestName,
volOptions.NamePrefix, "", "", volOptions.ReservedID)
if err != nil {
return nil, err
Make CephFS plugin stateless reusing RADOS based journal scheme This is a part of the stateless set of commits for CephCSI. This commit removes the dependency on config maps to store cephFS provisioned volumes, and instead relies on RADOS based objects and keys, and required CSI VolumeID encoding to detect the provisioned volumes. Changes: - Provide backward compatibility to provisioned volumes by older plugin versions (1.0.0 or older) - Remove Create/Delete support for statically provisioned volumes (fixes #382) - Added namespace support to RADOS OMaps and used the same to store RADOS CSI objects and keys in the CephFS metadata pool - Added support to mention fsname for CephFS provisioning (fixes #359) - Changed field name in CSI Identifier to 'location', to denote a pool or fscid - Updated mounter cache to use new scheme - Required Helm manifests are updated - Required documentation and other manifests are updated - Made driver option 'metadatastorage' as optional, as fresh installs do not need to specify the same Testing done: - Create/Mount/Delete PVC - Create/Delete 5 PVCs - Mount version 1.0.0 PVC - Delete version 1.0.0 PV - Mount Statically defined PV/PVC/Pod - Mount Statically defined version 1.0.0 PV/PVC/Pod - Delete Statically defined version 1.0.0 PV/PVC/Pod - Node restart when mounted to test mountcache - Use InstanceID other than 'default' - RBD basic round of tests, as namespace is added to OMaps - csitest against ceph-fs plugin - NOTE: CephFS plugin still does not detect and address already created volumes but of a different size - Test not providing any value to the metadata storage parameter Signed-off-by: ShyamsundarR <srangana@redhat.com>
2019-05-28 19:03:18 +00:00
}
// generate the volume ID to return to the CO system
vid.VolumeID, err = util.GenerateVolID(ctx, volOptions.Monitors, cr, volOptions.FscID,
"", volOptions.ClusterID, imageUUID, volIDVersion)
Make CephFS plugin stateless reusing RADOS based journal scheme This is a part of the stateless set of commits for CephCSI. This commit removes the dependency on config maps to store cephFS provisioned volumes, and instead relies on RADOS based objects and keys, and required CSI VolumeID encoding to detect the provisioned volumes. Changes: - Provide backward compatibility to provisioned volumes by older plugin versions (1.0.0 or older) - Remove Create/Delete support for statically provisioned volumes (fixes #382) - Added namespace support to RADOS OMaps and used the same to store RADOS CSI objects and keys in the CephFS metadata pool - Added support to mention fsname for CephFS provisioning (fixes #359) - Changed field name in CSI Identifier to 'location', to denote a pool or fscid - Updated mounter cache to use new scheme - Required Helm manifests are updated - Required documentation and other manifests are updated - Made driver option 'metadatastorage' as optional, as fresh installs do not need to specify the same Testing done: - Create/Mount/Delete PVC - Create/Delete 5 PVCs - Mount version 1.0.0 PVC - Delete version 1.0.0 PV - Mount Statically defined PV/PVC/Pod - Mount Statically defined version 1.0.0 PV/PVC/Pod - Delete Statically defined version 1.0.0 PV/PVC/Pod - Node restart when mounted to test mountcache - Use InstanceID other than 'default' - RBD basic round of tests, as namespace is added to OMaps - csitest against ceph-fs plugin - NOTE: CephFS plugin still does not detect and address already created volumes but of a different size - Test not providing any value to the metadata storage parameter Signed-off-by: ShyamsundarR <srangana@redhat.com>
2019-05-28 19:03:18 +00:00
if err != nil {
return nil, err
}
util.DebugLog(ctx, "Generated Volume ID (%s) and subvolume name (%s) for request name (%s)",
Make CephFS plugin stateless reusing RADOS based journal scheme This is a part of the stateless set of commits for CephCSI. This commit removes the dependency on config maps to store cephFS provisioned volumes, and instead relies on RADOS based objects and keys, and required CSI VolumeID encoding to detect the provisioned volumes. Changes: - Provide backward compatibility to provisioned volumes by older plugin versions (1.0.0 or older) - Remove Create/Delete support for statically provisioned volumes (fixes #382) - Added namespace support to RADOS OMaps and used the same to store RADOS CSI objects and keys in the CephFS metadata pool - Added support to mention fsname for CephFS provisioning (fixes #359) - Changed field name in CSI Identifier to 'location', to denote a pool or fscid - Updated mounter cache to use new scheme - Required Helm manifests are updated - Required documentation and other manifests are updated - Made driver option 'metadatastorage' as optional, as fresh installs do not need to specify the same Testing done: - Create/Mount/Delete PVC - Create/Delete 5 PVCs - Mount version 1.0.0 PVC - Delete version 1.0.0 PV - Mount Statically defined PV/PVC/Pod - Mount Statically defined version 1.0.0 PV/PVC/Pod - Delete Statically defined version 1.0.0 PV/PVC/Pod - Node restart when mounted to test mountcache - Use InstanceID other than 'default' - RBD basic round of tests, as namespace is added to OMaps - csitest against ceph-fs plugin - NOTE: CephFS plugin still does not detect and address already created volumes but of a different size - Test not providing any value to the metadata storage parameter Signed-off-by: ShyamsundarR <srangana@redhat.com>
2019-05-28 19:03:18 +00:00
vid.VolumeID, vid.FsSubvolName, volOptions.RequestName)
return &vid, nil
}
// reserveSnap is a helper routine to request a UUID reservation for the CSI SnapName and,
// to generate the snapshot identifier for the reserved UUID.
func reserveSnap(ctx context.Context, volOptions *volumeOptions, parentSubVolName string, snap *cephfsSnapshot, cr *util.Credentials) (*snapshotIdentifier, error) {
var (
vid snapshotIdentifier
imageUUID string
err error
)
// Connect to cephfs' default radosNamespace (csi)
j, err := snapJournal.Connect(volOptions.Monitors, radosNamespace, cr)
if err != nil {
return nil, err
}
defer j.Destroy()
imageUUID, vid.FsSnapshotName, err = j.ReserveName(
ctx, volOptions.MetadataPool, util.InvalidPoolID,
volOptions.MetadataPool, util.InvalidPoolID, snap.RequestName,
snap.NamePrefix, parentSubVolName, "", snap.ReservedID)
if err != nil {
return nil, err
}
// generate the snapshot ID to return to the CO system
vid.SnapshotID, err = util.GenerateVolID(ctx, volOptions.Monitors, cr, volOptions.FscID,
"", volOptions.ClusterID, imageUUID, volIDVersion)
if err != nil {
return nil, err
}
util.DebugLog(ctx, "Generated Snapshot ID (%s) for request name (%s)",
vid.SnapshotID, snap.RequestName)
return &vid, nil
}
// undoSnapReservation is a helper routine to undo a name reservation for a CSI SnapshotName.
func undoSnapReservation(ctx context.Context, volOptions *volumeOptions, vid snapshotIdentifier, snapName string, cr *util.Credentials) error {
// Connect to cephfs' default radosNamespace (csi)
j, err := snapJournal.Connect(volOptions.Monitors, radosNamespace, cr)
if err != nil {
return err
}
defer j.Destroy()
err = j.UndoReservation(ctx, volOptions.MetadataPool,
volOptions.MetadataPool, vid.FsSnapshotName, snapName)
return err
}
/*
checkSnapExists checks to determine if passed in RequestName in volOptions exists on the backend.
**NOTE:** These functions manipulate the rados omaps that hold information regarding
volume names as requested by the CSI drivers. Hence, these need to be invoked only when the
respective CSI driver generated volume name based locks are held, as otherwise racy
access to these omaps may end up leaving them in an inconsistent state.
These functions also cleanup omap reservations that are stale. I.e when omap entries exist and
backing subvolumes are missing, or one of the omaps exist and the next is missing. This is
because, the order of omap creation and deletion are inverse of each other, and protected by the
request name lock, and hence any stale omaps are leftovers from incomplete transactions and are
hence safe to garbage collect.
*/
func checkSnapExists(
ctx context.Context,
volOptions *volumeOptions,
parentSubVolName string,
snap *cephfsSnapshot,
cr *util.Credentials) (*snapshotIdentifier, *snapshotInfo, error) {
// Connect to cephfs' default radosNamespace (csi)
j, err := snapJournal.Connect(volOptions.Monitors, radosNamespace, cr)
if err != nil {
return nil, nil, err
}
defer j.Destroy()
snapData, err := j.CheckReservation(
ctx, volOptions.MetadataPool, snap.RequestName, snap.NamePrefix, parentSubVolName, "")
if err != nil {
return nil, nil, err
}
if snapData == nil {
return nil, nil, nil
}
sid := &snapshotIdentifier{}
snapUUID := snapData.ImageUUID
snapID := snapData.ImageAttributes.ImageName
sid.FsSnapshotName = snapData.ImageAttributes.ImageName
snapInfo, err := volOptions.getSnapshotInfo(ctx, volumeID(snapID), volumeID(parentSubVolName))
if err != nil {
if errors.Is(err, ErrSnapNotFound) {
err = j.UndoReservation(ctx, volOptions.MetadataPool,
volOptions.MetadataPool, snapID, snap.RequestName)
return nil, nil, err
}
return nil, nil, err
}
defer func() {
if err != nil {
err = volOptions.deleteSnapshot(ctx, volumeID(snapID), volumeID(parentSubVolName))
if err != nil {
util.ErrorLog(ctx, "failed to delete snapshot %s: %v", snapID, err)
return
}
err = j.UndoReservation(ctx, volOptions.MetadataPool,
volOptions.MetadataPool, snapID, snap.RequestName)
if err != nil {
util.ErrorLog(ctx, "removing reservation failed for snapshot %s: %v", snapID, err)
}
}
}()
tm, err := parseTime(ctx, snapInfo.CreatedAt)
if err != nil {
return nil, nil, err
}
sid.CreationTime = tm
// found a snapshot already available, process and return it!
sid.SnapshotID, err = util.GenerateVolID(ctx, volOptions.Monitors, cr, volOptions.FscID,
"", volOptions.ClusterID, snapUUID, volIDVersion)
if err != nil {
return nil, nil, err
}
util.DebugLog(ctx, "Found existing snapshot (%s) with subvolume name (%s) for request (%s)",
snapData.ImageAttributes.RequestName, parentSubVolName, sid.FsSnapshotName)
return sid, &snapInfo, nil
}