HwameiStor Release Notes¶
This page lists the Release Notes related to HwameiStor, providing an overview of the evolution path and feature changes for each version.
2024-09-30¶
v0.15.0¶
- Improved the usage of reflink for formatting the file system (XFS).
- Fixed an issue encountered during LVM execution.
- Fixed an issue where node disk references were not created but were already mounted.
2024-08-30¶
v0.14.9¶
- Added the verbs/patch permissions to ClusterRole.
- Fixed an issue where bound PVCs were not filtered.
2024-07-30¶
v0.14.8¶
- Improved the default reclaim policy changed from delete to retain
- Fixed an issue to output the disk along with the partition table.
- Fixed an issue where empty pvName was not handled
- Fixed an issue with the affinity and taint skip policy
2024-06-30¶
v0.14.7¶
- Improved affinity issues and added an option to skip affinity.
- Improved and cleaned up the dataset manager code.
2024-04-30¶
v0.14.6¶
- Improved Resolve the issue where LVR does not respect pod affinity
- Improved Add an option to enable data validation during migration
2024-03-31¶
v0.14.4¶
- Improved Delete local mount path when unmounting data volumes
- Improved Added PoolHDD FreeCap to the print column
2024-01-31¶
v0.14.1¶
Improvements¶
- Improved Skip volumes that have defined storage classes but no storage class instances when ignoring failure policy
- Improved Check for the existence of local disks before use
- Improved snapshot client
- Improved Added snapshot support to Hwameictl command-line tool
- Improved Added cluster support to Hwameictl command-line tool for snapshots
Fixes¶
- Fixed an issue with getnode API
- Fixed an issue with ctl disk_list
- Fixed an issue with snapshot control error
2023-12-31¶
v0.14.0¶
Improvements¶
- Improved Removed the use of meaningless poolType
- Improved Added VolumeSnapshotClass
- Improved Added support for
snapshot
parameter in Hwameictl command-line tool
Fixes¶
- Fixed Ensured Volume Group always maintains consistent accessibility with all volumes in the group
- Fixed an issue with pending error when cleaning up replicas in localvolumemigrate
- Fixed Added system exit statement when indexer setting fails
- Fixed an issue with StorageNodePoolDiskGet retrieving localdisk incorrectly
- Fixed volume leakage when users delete PVC but do not create PV
2023-11-30¶
v0.13.3¶
New Features¶
- Added support for volume cloning
- Added delayed volume deletion when snapshot is discovered
- Added filtering nodes based on source volume accessibility
Improvements¶
- Improved logging disk claim events only when the disk is available
Fixes¶
- Fixed filtering replicaSnapRestoreName before submitting a job
- Fixed incorrect state transition in LocalVolumeConvert
- Fixed exiting when indexer addition fails
- Fixed using storage node IP when migrating volumes
- Fixed cleaning up replicas after canceling publication
- Fixed error in
apiserver getnodedisk
and added set-diskowner API
2023-10-30¶
v0.13.1¶
- Added compatibility with Kubernetes v1.28 version
- Added LVM data volume snapshot feature
- Improved HwameiStor Operator with additional component resource configuration
- Improved snapshot recovery timeout
- Improved default log level for LDM
2023-8-30¶
v0.12.1¶
New Features¶
- Added Support for
ext
filesystem. - Added New features parameters for Local Disk.
- Added Field to record device history information.
- Added SN/ID Path recognition for disks.
- Added Automatic disk expansion functionality.
- Added System resource auditing for cluster, storage nodes, disks, and data volumes
Improvements¶
- Fixed Display issue of NVME disks in LocalStoragePool nodes.
- Fixed an issue with lost replica status during migration operation.
- Fixed Disallow disk allocation when Disk Owner is empty.
- Fixed Issues with Failover functionality in
deploy
andMakefile
2023-7-05¶
v0.11.1¶
- Added Support for automatic detection of
cgroup
version.
2023-6-25¶
v0.11.0¶
- Added Implementation of IO limitation or QoS.
- Added Use of
/virtual/
detection to identify virtual block devices. - Fixed Inconsistency in creation time field for StorageClass.
2023-5-26¶
v0.10.3¶
- Improved Permissions-related content in Helm templates.
2023-7-05¶
v0.11.1¶
- Added support for automatic detection of
cgroup
version
2023-6-25¶
v0.11.0¶
- Added implementation of IO limitation or QoS
- Added identification of virtual block devices using /virtual/
- Fixed an inconsistency issue with StorageClass creation time field
2023-5-26¶
v0.10.3¶
- Improved permissions-related content in Helm Charts
2023-5-25¶
v0.10.2¶
- Improved automatic conversion of Node Name
.
to-
, affecting LDM heartbeat detection - Added Admin permission functionality
2023-5-17¶
v0.10.0¶
Improvements¶
- Added identification and setting of local disk Manager Owner information
- Improved marking disk status as 'inactive' when removal time is received
- Improved dashboard for disk Smart indicators
- Added local StorageClass
- Improved display of UI component statuses
- Improved separation of disk allocation and disk status update processes
- Improved renaming of Exporter port to http-metrics
- Improved addition of port name in exporter service
Bug Fixes¶
- Fixed an issue where LD is bound but has no capacity in LSN
- Fixed Metrics port listening issue
- Fixed an issue that may cause "not found" errors
- Fixed an issue with UI tag in Helm
v0.9.3¶
Improvements¶
- Improved filling of disk owner information during usage
- Improved merging of disk self attributes when triggered by udev events
- Improved adding labels to svc
- Improved separation of disk allocation and disk status update processes
- Improved renaming of Exporter port to http-metrics
- Improved addition of port name in exporter service
Bug Fixes¶
- Fixed an issue where LD is bound but has no capacity in LSN
- Fixed an issue when listening the metrics port
- Fixed an issue that may cause "not found" errors
- Fixed an issue with UI tag in Helm
2023-3-30¶
v0.9.2¶
- Added UI relok8s
v0.9.1¶
- Added volume status monitoring Issue #741
- Fixed Local Storage deployment parameters Issue #742
v0.9.0¶
New Features¶
- Added disk Owner Issue #681
- Added Grafana Dashboard Issue #733
- Added Operator installation method, automatically pulling UI during installation Issue #679
- Added UI application Label Issue #710
Improvements¶
- Added used capacity of disks Issue #681
- Improved skipping scoring mechanism when no available disks are found Issue #724
- Set default DRBD port to 43001 Issue #723
2023-1-30¶
v0.8.0¶
- Improved Chinese documentation
- Improved value.yaml file
- Updated Roadmap
- Improved default failure strategy when installation fails
2022-12-30¶
v0.7.1¶
- Added HwameiStor Dashboard UI, which displays storage resource and storage node usage status
- Added interface for managing HwameiStor storage nodes, local disks, and migration records
- Added StorageClass management functionality, displaying basic information about StorageClass and corresponding nodes
- Added local volume management functionality, supporting data volume migration and high availability conversion
- Improved unnecessary logging before data migration and avoided interference from Job execution in other namespaces