Ussuri Series Release Notes

16.1.0-10

Bug Fixes

  • Fixed bug #1859652 to allow retyping an attached volume to SolidFire.

  • Cinder no longer allows an incremental backup to be created while having the parent backup in another project.

  • Fix bug #1874134, allowing an iSCSI or FCP volume to be extended to a size up to 16TB regardless of its original size, even if it’s attached to an instance.

  • Fix HPE 3PAR driver issue where volumes that were live migrated to it would end up being inaccessible. We would no longer be able to use the volume for any operation, such as attach, detach, delete, snapshot, etc. (bug 1697422)

  • PowerMax Driver - Concurrent live migrations can sometimes fail when one thread deletes a storage group that another thread may need.

16.1.0

Upgrade Notes

  • The fix for Bug #1823200 requires os-brick >= 3.0.2 but < 3.1.0.

Security Issues

Bug Fixes

  • Bug #1823200: This release contains an updated Dell EMC VxFlex OS driver. It must be used with os-brick >= 3.0.2 but < 3.1.0. and requires that a new configuration file be deployed on compute nodes, cinder nodes, and anywhere you would perform a volume attachment in your deployment. See the Dell EMC VxFlex OS (ScaleIO) Storage driver documentation for details about the configuration file, and see OSSN-0086 for more information about the security vulnerability.

  • Bug #1875570 <https://bugs.launchpad.net/cinder/+bug/1875570>: Fixed issue with NFS backend where the image-volume cache was never used to create a volume, even when the cache was enabled.

  • Fix the HPE 3PAR driver’s attempt to rename the backend volume after it was migrated. If the original volume resides on the same 3PAR backend then the pre and post migration volume names are swapped. Otherwise, the newly migrated volume is renamed to match the original name. (bug 1858119)

16.0.0

Prelude

Welcome to the Ussuri release of the OpenStack Block Storage service (cinder). The cinder team would like to bring the following points to your attention. Details may be found below.

  • With this release, the Block Storage API version 3 has reached microversion 3.60.

  • Python 2 is no longer supported. The minimum version of Python that may be used with this release is Python 3.6.

  • The unsupported driver removal policy has been revised. See the “Known Issues” section of this document for more information.

New Features

  • Huawei Cinder Driver Support Dorado V6 Storage.(iSCSI, FC)

  • Added the image_conversion_cpu_limit and image_conversion_address_space_limit as configurable parameters. This adds configurability to the image conversion process to prevent the process from timing out when converting larger images.

  • Added the property src_backup_id to the volume’s metadata, to record from which backup the new volume was created. If the src_backup_id exists in the volume’s metadata, it will be updated. When restoring from a chain of incremental backups, src_backup_id is set to the last incremental backup used for the restore.

  • The Datera driver now supports API v2.2, IOPS/GB and BW/GB settings, LDAP and CHAP, extended metadata attributes during volume creation and attachment. Most retype operations do not detach volumes anymore. Manageable Snapshots can be listed. Flash and Hybrid capacity information added.

  • The LINSTOR Driver for Cinder now supports LINSTOR 0.9.12.

  • Fujitsu Driver: Added multiple pools support.

  • Ability to add minimum and maximum volume size restrictions which can be set on a per volume-type granularity. New volume type keys of ‘provisioning:min_vol_size’ and ‘provisioning:max_vol_size’.

  • PowerMax Driver - Support to allow the use of multiple replication modes on one backend array.

  • Dell EMC PowerMax driver now supports Unisphere storage group and array tagging to allow the user to specify a user defined tag to facilitate easy access and classification.

  • Dell EMC PowerMax driver now faciliates the user to override the short host name and port group name seen in PowerMax masking view and storage view terminology. This means the user can give more meaningful names, especially when the short host name exceeds 16 characters and the port group name exceeds 12 characters, which is the condition where the driver truncates these values.

  • RBD driver: support added for reverting a volume to the most recent snapshot taken.

    Please be aware of the following known issues with this operation and the Ceph storage backend:

    • Rolling back a volume to a snapshot overwrites the current volume with the data from the snapshot, and the time it takes to complete this operation increases with the size of the volume.

      It is faster to create a new volume from a snapshot. You may wish to recommend this option to your users whose use cases do not strictly require revert-to-snapshot.

    • The efficiency of revert-to-snapshot is also dependent upon the Ceph storage backend in use, namely, whether or not BlueStore is being used in your Ceph installation.

    Please consult the Ceph documentation for details.

  • Added SandStone driver that allows cinder to manage volumes in ISCSI environment.

  • NetApp SolidFire driver: Enabled support for Active/Active (including replication) to the SolidFire driver. This allows users to configure SolidFire backends in clustered environments.

  • This release includes support for Glance multiple stores. An operator may now specify which Glance store will be used when a volume is uploaded to Glance as an image. Some details about this feature:

    • This feature is not directly user-facing. To enable it, an operator must add the field image_service:store_id in the volume-type extra-specs. The value of the field is a valid store identifier (id) configured in Glance, which may be discovered by making a GET /v2/info/stores call to the Image Service API.

    • If image_service:store_id is not set in the extra-specs for a volume-type, then any volume of that type uploaded as an image will be uploaded to the default store in Glance.

    • The image_service:store_id can only be set in the extra-specs for a volume-type when multiple glance stores are configured.

    • Cinder validates proposed Glance store identifiers by contacting Glance at the time the image_service:store_id is added to a volume-type’s extra-specs. Thus the Image Service API must be available when a volume-type is updated.

  • Beginning with microversion 3.60, users may apply time comparison filters to the volume summary list and volume detail list requests by using the created_at or updated_at fields. Time must be expressed in ISO 8601 format. See the Block Storage API v3 Reference for details.

  • Dell EMC Unity Driver: Added consistency group replication support. The storops library version 1.2.3 or newer is required.

  • Dell EMC Unity driver: Add efficient retype support when new type uses the same Unity device.

  • Dell EMC Unity driver: Add tiering policy configuration support for volume.

  • This release includes support for Glance automatic image colocation. When a volume originally created from an image is uploaded to the Image service, Cinder passes Glance a reference to the original image. Glance may use this information to colocate the new image data in the same image store(s) as the original image data. Consult the Glance documentation for more information.

  • VxFlex OS driver now supports VxFlex OS 3.5.x.

  • VxFlex OS driver now supports storage-assisted volume migration.

  • VxFlex OS driver now supports OpenStack volume replication v2.1 for VxFlex OS v3.5.0 storage backends.

  • VxFlex OS driver now supports storage-assisted revert volume to snapshot.

Known Issues

  • HPE 3PAR driver now supports networks with duplicated FQDNs via configuration option unique_fqdn_network so attaching in these networks will work (bug #1834695).

  • PowerMax driver - Disabling inuse storage assisted migration to a metro or asynchronous replicated volume type as this operation will not facilitate FC scanning or iSCSI login of the target array.

  • RBD driver: There are some known issues concerning the revert-to-snapshot support added in this release.

    • The time it takes to complete the revert-to-snapshot operation increases with the size of the volume. It is faster to create a new volume from a snapshot.

    • The efficiency of revert-to-snapshot depends upon the Ceph storage backend in use, particularly whether or not BlueStore is being used in your Ceph installation.

    Please consult the Ceph documentation for details.

  • For the convenience of operators, the unsupported driver removal policy was changed during the Ussuri development cycle to allow unsupported drivers to remain in the Cinder source code repository at the discretion of the Cinder team (basically, this means the vendor has not explicitly declared the driver EOL or the vendor has expressed interest in getting the third-party CI system working again). Be aware, however, that such drivers are subject to immediate removal if they begin failing the general Cinder gating tests.

    The following drivers were classified as unsupported in the Train release and continue as unspported in this release:

    • IBM FlashSystem drivers: FC and iSCSI

    • ProphetStor drivers: FC and iSCSI

    • Veritas ACCESS iSCSI driver

    • Virtuozzo Storage driver

    For completeness, here is the list of drivers first marked unsupported in the Ussuri release. See the “Deprecation Notes” section of this document for details.

    • Brocade Fibre Channel Zone Manager driver

    • Veritas Clustered NFS driver

    In order to use an unsupported driver, enable_unsupported_driver must be set to True in the driver’s section in the cinder.conf file. If you are the consumer of such a driver, we encourage you to contact the vendor to make them aware of your concerns.

  • A key aspect of supported drivers is that there is a fully functioning third-party CI system that reports on all proposed changes to the cinder code. However, the fact that there are ‘unsupported’ drivers at all indicates that third-party CI systems are unfortunately not always fully functioning. You may consult the report at the following link to learn the extent to which a particular driver’s CI system is reporting on the development branch of cinder:

    http://cinderstats.ivehearditbothways.com/cireport.txt

Upgrade Notes

  • IBM GPFS drivers had been previously marked unsupported. Testing requirements have been addressed and they are now fully supported again. IBM GPFS drivers allow cinder to manage volumes.

  • The LINSTOR driver for Cinder supports LINSTOR 0.9.12. The driver supports LINSTOR backend using REST API.

    The new driver adds ‘linstor_autoplace_count’ configuration option that specifies the number of volume replicas.

  • Python 2.7 support has been dropped. The last release of Cinder to support py2.7 is OpenStack Train. The minimum version of Python now supported by Cinder is Python 3.6.

  • The Huawei FusionStorage driver had previously been marked unsupported. The 3rd Party CI has been restored, and the driver is now fully supported in the Ussuri release.

  • IBM DS8000 drivers had been previously marked unsupported. Testing requirements have been addressed and they are now fully supported again. IBM DS8000 drivers allow Cinder to manage volumes in FC environment.

  • IBM Spectrum Virtualize Family (previously known as Storwize) drivers had been previously marked unsupported. Testing requirements have been addressed and they are now fully supported again. IBM Spectrum Virtualize Family drivers allow Cinder to manage volumes both in iSCSI and FC environments.

  • The IET iSCSI target driver has been marked deprecated and will be removed in the “V” release. The IET iSCSI target project is no longer active and is not supported by all distributions. It is recommended to migrate to a supported distribution and iSCSI target prior to upgrading.

  • The Brocade Fibre Channel Zone Manager driver has been marked as unsupported and is now deprecated. It is subject to removal during the “V” development cycle, following the standard OpenStack deprecation policy. The config option enable_unsupported_driver will need to be set to True in the driver’s section in cinder.conf to continue to use it in this release.

  • SolidFire supports Synchronous, Asynchronous and SnapshotsOnly replication modes. This adds the config option solidfire:replication_mode to specify the mode to be used by Cinder. Its value can be Sync, Async or SnapshotsOnly.

  • HPE Nimble Storage drivers had been previously marked unsupported. Testing requirements have been addressed and they are now fully supported again. HPE Nimble Storage drivers allow cinder to manage volumes both in iSCSI and FC environment.

  • Dell EMC PS Series storage driver is not supported and removed starting from the Ussuri release. It was marked as deprecated in the Train release.

  • The HPE Lefthand Driver (iSCSI) was marked unsupported in the Train release as the StoreVirtual product line has gone EOL and the LeftHand OS no longer receives upgrades. The driver has been removed in this release. All data on backends powered by HPE LeftHand OS should be migrated to a supported storage backend before upgrading your Cinder installation.

  • If you have backup_driver_status_check_interval option in your cinder.conf we recommend you to use backup_driver_stats_polling_interval to avoid deprecation warnings in logs.

  • SPDK target and volume drivers have been updated with new SPDK specific RPC calls due to deprecation of some old RPC calls. Starting from Ussuri release SPDK release v19.10 or higher is required.

  • The Sheepdog driver was marked unsupported in the Train release and has now been removed. All data on Sheepdog backends should be migrated to a supported backend before upgrading your Cinder installation.

  • The following drivers were classified as unsupported in the Train release and continue as unspported in this release. See the “Known Issues” section of this document for details.

    • IBM FlashSystem drivers: FC and iSCSI

    • ProphetStor drivers: FC and iSCSI

    • Veritas ACCESS iSCSI driver

    • Virtuozzo Storage driver

    In order to use an unsupported driver, enable_unsupported_driver must be set to True in the driver’s section in the cinder.conf file. If you are the consumer of such a driver, we encourage you to contact the vendor to make them aware of your concerns.

  • The Veritas Clustered NFS driver has been marked as unsupported and is now deprecated. enabled_unsupported_driver will need to be set to True in the driver’s section in cinder.conf to continue to use the driver.

  • The Virtuozzo driver has been marked as unsupported and is now deprecated. enable_unsupported_driver will need to be set to True in the driver’s section in cinder.conf to continue to use it.

  • VMware vmdk driver: The vmware vmdk driver had its get_volume_stats removed in a previous release due to a potential performance hit of 20% at a high load. The problem with reporting unknown back to the scheduler, is that it effectively removes cinder’s ability to properly schedule based on capacity utilization. When this driver is enabled in a heterogenous environment without properly reporting utilization statistics, the scheduler’s capacity filter will always allow this driver to service a provisioning request. Without reporting the backend stats, the capacity filter also can’t determine the reserved_percentage as well as the max_over_subscription_ratio. To enable the collection of stats set vmware_enable_volume_stats to True in the driver section of cinder.conf. The default setting is False. Keep in mind that there may be a degradation in performance on the vcenter when enabling this setting.

  • The volume_extension:volume_type_encryption policy, which was deprecated in Stein, has been un-deprecated for the convenience of operators who would like to set the policies for the create, get, update, and delete operations for a volume type’s encryption type in one place. The default value for this policy target has not changed.

    As a reminder, the finer-grained policies are:

    • volume_extension:volume_type_encryption:create

    • volume_extension:volume_type_encryption:get

    • volume_extension:volume_type_encryption:update

    • volume_extension:volume_type_encryption:delete

    The default values for these policies have also not changed. See the sample Cinder policy file for more information.

Deprecation Notes

  • The Datera driver removed v2 API support and the usage of initiator-groups

  • The configuration option rbd_keyring_conf for the Ceph cinder driver presents a security risk and the option is hereby deprecated and scheduled to be removed early in the ‘V’ development cycle, following the standard OpenStack deprecation policy. For more information, see OSSN-0085: Cinder configuration option can leak secret key from Ceph backend.

  • The IET iSCSI target driver has been marked deprecated and will be removed in the “V” release. The IET iSCSI target project is no longer active and is not supported by all distributions. It is recommended to migrate to a supported distribution and iSCSI target prior to upgrading.

  • The Brocade Fibre Channel Zone Manager driver has been marked as unsupported and is now deprecated. It is subject to removal during the “V” development cycle, following the standard OpenStack deprecation policy. The config option enable_unsupported_driver will need to be set to True in the driver’s section in cinder.conf to continue to use it in this release.

  • backup_driver_status_check_interval config option is renamed to backup_driver_stats_polling_interval to be similar with volume drivers configuration. Old option name support will be dropped in U release.

  • In this release, sending os-reset_status notifications to the following nonstandard publisher_ids is DEPRECATED:

    • ‘volumeStatusUpdate’ for volume status resets

    • ‘volumeStatusUpdate’ for snapshot status resets

    • ‘backupStatusUpdate’ for backup status resets

    The notifications continue to be published to the above during the deprecation period. Beginning with this release, the os-reset_status notifications are also sent to the following standard publisher_ids:

    • ‘volume’ for volume status resets

    • ‘snapshot’ for snapshot status resets

    • ‘backup’ for backup status resets

    This will allow consumers of these notifications to make a smooth transition. In the Victoria release, os-reset_status notifications will only be sent to the standard publisher_ids.

  • The following drivers were deprecated in the Train release but have not yet been removed. They continue as unsupported and deprecated in this release. See the “Known Issues” section of this document for details.

    • IBM FlashSystem drivers: FC and iSCSI

    • ProphetStor drivers: FC and iSCSI

    • Veritas ACCESS iSCSI driver

    • Virtuozzo Storage driver

    In order to use an unsupported driver, enable_unsupported_driver must be set to True in the driver’s section in the cinder.conf file. If you are the consumer of such a driver, we encourage you to contact the vendor to make them aware of your concerns.

  • The Veritas Clustered NFS driver has been marked as unsupported and is now deprecated. enable_unsupported_driver will need to be set to True in the driver’s section in cinder.conf to continue to use the driver. It its support status does not change the driver will be removed in the ‘V’ development cycle.

  • The Virtuozzo driver has been marked as unsupported and is now deprecated. enable_unsupported_driver will need to be set to True in the driver’s section in cinder.conf to continue to use it. If its support status does not change, it will be removed in the ‘U’ development cycle.

Security Issues

  • Due to OSSN-0085: Cinder configuration option can leak secret key from Ceph backend, deployers using the rbd_keyring_conf option are advised to stop using it immediately. The option has been deprecated for removal early in the ‘V’ development cycle.

Bug Fixes

  • Fixes a bug that prevented creation of Quobyte volumes from snapshots during snapshot backups. This now allows backing up volumes with existing snapshots. Partially fixes bug 1703405 .

  • NetApp SolidFire driver: Fixed an issue that causes failback to fail after a volume service restart. This change fixes bug 1859653.

  • The volume-transfers list calls (GET /v3/{project_id}/volume-transfers, GET /v3/{project_id}/volume-transfers/detail) were not recognizing name as a filterable attribute. That has been fixed in the current release.

  • Fix volume migration fails in the same ceph RBD pool. Bug 1871524.

  • PowerMax Driver - Allowing for default volume type in group operations where the array serial number is retrieved from the cinder.conf instead of the pool_name on the extra specs.

  • PowerMax Driver - Issue with upgrades from pre Pike to Pike and later. The device is not found when trying to snapshot a legacy volume.

  • Fixes a bug that could cause mount failures with the Quobyte driver if the quobyte_volume_url setting was changed in a running system.

  • Datera driver: fixes in retyping / QoS, Glance interoperability, fast clones, IP pools, volume templates and initiators, unicode character support, scalability issues

  • An incorrect lock in the remotefs code, which is used for the NFS driver, and other similar drivers, resulted in concurrent clone volume operations failing. create_cloned_volume now locks on the source volume id, meaning multiple clone operations from the same source volume are serialized.

    A lock in the volume manager flow generally prevents this on normal clone volume operations, but this clone method in the driver is called for operations such as cloning from the cinder image-volume cache or cloning from a cinder backend used as a glance store.

  • The SolidFire replication was fixed. Several bugs were addressed (creating replicated volumes from snapshots, from volumes, retype a volume to a replicated type, managing a volume to a replicated type, correctly updating portal info on failover/failback and some minor other fixes). Closes bugs #1834013, #1751932.

  • This change fixes bug 1845483 - 3PAR: For Peer Persistence, add policy options in RCG. Following options are added after Remote Copy Group (RCG) is created: autoFailover, pathManagement, autoRecover

  • HPE 3PAR driver: The detach issue for multiattach capability (Bug 1834660) was fixed in the Cinder 15.0.0 (Train) release, but due to an oversight, the fix has not been announced until now.

  • NetApp ONTAP: Fixes bug 1839384 Detaching any instance from multiattached volume terminates connection. Now the connection is terminated only if there’re no other instances using the same initiator.

  • PowerMax driver - fix to eliminate ‘cannot use the device for the function because it is in a Copy Session’ when attempting to delete a volume group that previously had a group snapshot created on and deleted from it.

  • Catch argument exceptions when configuring multiattach for rbd volumes. This allows multiattach images with flags already set to continue instead of raising an exception and failing.

  • os-reset_status notifications for volumes, snapshots and backups will now go to the standard publisher id for volume, snapshot and backup like all other notifications for volume, snapshot and backup.

  • VMware vmdk driver: The collection of volume stats, which had been disabled, may now be turned on by using the vmware_enable_volume_stats configuration option. The default for this option is False (no stats collection). Be aware that enabling volume stats may cause performance issues under high load.

Other Notes

  • The Datera driver went under a major driver revamp/restructure and a new separate Datera Python-SDK requirement has been introduced

  • PowerMax driver - the minimum version of Unisphere for PowerMax for the current release is 9.1.0.5. It is however recommended to install the Security Releases(SR) of Unisphere for PowerMax if they become available.

  • PowerMax Driver - Two new replication specific configuration options sync_interval and sync_retries have been added to PowerMax cinder configuration. These configuration options determine how many times to retry checks to see if a SnapVX copy mode has completed with a replication enabled volume, and how long to wait between retries.