File linstor.changes of Package linstor
-------------------------------------------------------------------
Wed Mar 9 06:07:42 UTC 2022 - Fridrich Strba <fstrba@suse.com>
- Do not hardcode java 8 dependency, since the gradle in Factory
is able to build this one just fine and it works well with
java 11.
-------------------------------------------------------------------
Wed Jan 12 06:15:16 UTC 2022 - nick wang <nwang@suse.com>
- bsc#1194567, update to 1.17.0
* Nodes can now be evacuated on purpose
* Browsing the controller port, will now redirect to the /ui if installed
* Check for all DRBD to be updated before allow resize
* Do not send crypt key before node authentication
* Autoplace do not count evicted resources
* Fixed CRD volume definition loading errors
* Fixed node reconnector deadlock
* Add remote SPDK support
* Fix migratedisk without body content
* Fix toggledisk reuse node-ids
* Fix autosnapshotshipping only work once
* Fix toggledisk missing layerstack bug
* Fix BCache trying to register too often
* Fix Luks resize with newer cryptosetup versions
* Allowed new BCache layer to be used as topmost layer
* Fixed API-Version reported by Controller
* Removed invalid description in rest_v1_openapi.yaml
* Fix controller disable auto verify algorithm
* Update auto-verify algorithm list to not use algorithms with optional keys
* Exos: fixed setting correct properties
* Exos: recache settings
* Exos: support OverrideVlmId property
* Fix creating resource groups using postgresql backend
* Fixed re-auto-placing resource-definition on moving to different rscgrp
* Updated automatic-verify algorithm list for DRBD
* Fix migration on postgresql
* Fixed db migration of resource with suffix volumes (e.g. ext-metadata)
* Fix resource group writing to database(resulting in restart errors)
* Fixed deleting resource definition that had once snapshots created
* Allow c-max-rate 0 as DRBD setting
* Added auto-setting a DRBD verify algorithm on resource-definition
* Support for shared storage pools (same physical device)
* Support for Seagate Exos Enterprise storage products
* Added make resource available API call
* Reworked auto-quorum property handling, now it is properly inherited from RG or Controller
* Removed linstor-common.conf, settings are now stored directly in all res files
* Fixed checking storage pool sizes before running RG spawn (that would fail)
* Check if storage pool has enough space before running resize
* Fixed changing resource group for resource definition to persistent between restarts
* Fixed DRBD resource with external metadata (would always be recreated)
* Fixed incorrect copying of layer data in snapshots with LUKS
* Allow "Undelete" of a resource that is in DELETING state
* API-Version 1.12.0
-------------------------------------------------------------------
Thu Dec 23 09:14:51 UTC 2021 - Dominique Leuenberger <dimstar@opensuse.org>
- Do not use %{python_module…} for recommends, as this would add
recommends to all python flavors. Rather change it to python3-…,
so recommend the default python modules only.
-------------------------------------------------------------------
Wed May 5 13:14:26 UTC 2021 - Callum Farmer <gmbr3@opensuse.org>
- Fixes for %_libexecdir changing to /usr/libexec(bsc#1174075)
-------------------------------------------------------------------
Tue Jan 26 06:30:51 UTC 2021 - nick wang <nwang@suse.com>
- bsc#1181397, update to 1.11.1
* Improved error messages for incorrect property values
* Fix incorrect triggers for auto-tiebreaker (adding node/delete node)
* Fix auto-tiebreaker looking for wrong storage pools
* Fix loading of resource if external metadata property is used
* Fixed incorrect loading of property keys in ETCD-database
* Toggle-disk now tries to resolve the storage pool from already deployed resources
* Fixed missing rollback if a resource-group spawn failed
* Added REST-API paths for retrieving property info:
`properties/info`, `properties/info/all`
* Added autoplace for tiebreaker and a new client
notation `--auto-place +1`
* Check if node resource inUse or last diskful before
deleting on node delete
* Do not return error if a resource already exists (now returns an INFO)
* Fixed missing property resolving objects while building
the DRBD config
* Fixed bug with file based storage pools
* Fixed node lost db integrity issues
* Fixed several issues with the auto-evict feature
* Added auto-evict feature
* ETCD prefix is now configurable (migration only works now
starting from version 1.4.3)
* Block IO can now be throttled also by iops
* Fixed REST-API single snapshot filtering
* Fixed drbd-events2 parsing race condition
* Fixed toggle-disk doesn't work if an unrelated node is offline
* Fixed race-condition in auto-tiebreaker
* Fixed usage of wait for snapshot-shipping
* Added auto-"unplace", automatically removes resources if
they exceed the replica count if `DrbdOptions/auto-unplace` is set
* Added error-report deletion API
* Expose DRBD promotion_score and may_promote via REST-API
and server sent events
* Prometheus metrics are now mostly cached and much faster to query
* Send SD_NOTIFY READY to systemd if services are ready
* OpenFlex resource are now correctly limited to 32 characters
* Fix etcd modify net-interface (incorrect serialization)
* Fix autoplacer also toggling diskless resources
* Fix snapshot-shipping with more than 1 volume definitions
* Fix zfs create options handling
* Fixed recaching of lvm-filter
* Fix missing client answer if drbd device doesn't become ready
in reasonable time
* REST-API v1.6.0
-------------------------------------------------------------------
Tue Sep 22 08:06:36 UTC 2020 - nick wang <nwang@suse.com>
- bsc#1176790, update to 1.8.0
* store resource and snapshot creation date in the database
* containers: add full s390x support
* LvmThin: Recreate shipped snapshot after merging
* Snapshot: Resume-IO on error and add timeout to takeSnapshot
[...]
-------------------------------------------------------------------
Mon Jun 1 06:05:35 UTC 2020 - nick wang <nwang@suse.com>
- Update to 1.7.1, match to linstor client/api 1.1.2
* VolumeDefinition: Make create synchronous
* Autoplacer: Fixed decimal regex for weights
* Renamed DrbdProxy/Site to Site
* ParseThinSpace: Fixed NPE due to a race condition
* ParseThinSpaces: return now all requested entries
* AutoPlaceApi: added null-pointer check
* Autoplacer: Fixed "autoplaced on 0 nodes" bug
* ReconnectorTask: Attempt to fix possible deadlock
* DrbdProxy: Disable autoproxy when disabling proxy
* AutoProxy: Added migration to rename Site property
* virter/Makefile: fix hardcoded path
[...]
- 1.4.0-1
* rest-api: go back to version 1.0.11
* errorreport: parameterize comparable
* wiphandler: allow a collection of paths to be specified
* create-storagepool: Allow an onError flux to be given and fix Exc handling
* createdevicepool: cleanup
* scripts/rest-gen: allow null default for string
* genericname: Throw InvalidNameException if string is null
* rest/physicaldevice: Add "atomic" storage pool creation
* rest-api-doc: fix incorrect yaml
* lsblkutils: show full paths for device names
* createdevicepool: do not continue on vdo errors
* args: allow args to be overwritten
[...]
- 0.9.13-1
* taskscheduleservice: catch exception in running task
* retryresourcestask: check if resource is deleted
* database/mariadb: add compatibility with mariadb
* allow node names to be matched with a search domain property
* ctrlauthenticator: fix typo
* ctrlapidataload: fix accessdenied on loadnode
* InvalidKeyException: extend from LinStorRuntimeException
* rest-api-docu: fix missing volume_number in volumes PUT
* CmldArguments: ConfigurationDir is now returned as Path
* grizzly-http: add exception mapper for ApiRcException
* apicallrcimpl: shortcut to create a ApiCallRc with 1 entry
[...]
-------------------------------------------------------------------
Tue May 28 06:36:36 UTC 2019 - nwang@suse.com
- Update to 0.9.10, match to linstor client/api 0.9.5
- linstor-server 0.9.10
* Fixed: lvm not working if the --reportformat option is missing
* Fixed: Possible DRBD split-brain if a resources is readded on a node
* Fixed: HTTP server tries to start on ipv4 if ipv6 bind failed
- linstor-server 0.9.9
* REST-API v1.0.4:
- Added /v1/views/storage-pools overview
- Added UUID for main objects
- Enable gzip compression if requested
* Added WaitTimeoutAfterCreate property
* Fixed: repeatedly created error logs for version mismatch on satellites
* Fixed ZFS volume resize
* Fixed out of date usable size reporting
* Removed DISKLESS provider dependency on DRBD
- linstor-server 0.9.8
* Fixed: Finding device path of NVMe volumes
* Fixed: Local Controller node will now show as ONLINE
* REST: Added /v1/view/resources and set REST-API version to 1.0.3
* REST: Data structures are now generated from the rest documentation
* Allow to pass raid level for LVM
* Passphrase generation changed, old encrypted volumes will not be able to be unlocked
- linstor-server 0.9.7
* Fixed: Volume definitions not deleted on satellite
* Fixed: Race condition while creating volume definitions too fast
* Fixed: Minor bug in NVME layer
* Fixed: NVME issues if drbd-utils have not been installed
* NVME: Added creating and deleting volumes while resource is up
* NVME: Added storage pools to prio props for NVME PrefNic property
* REST: updated to version 1.0.2, added /v1/storage-pool-definitions, minor fixes
* REST: default bind interface is now `::` ipv6 and ipv4
* Use better default JVM memory options
- linstor-server 0.9.6
* Fixed: DRBD handlers written on satellite restart
* Fixed: Migration for resources without volumes
* Fixed: Deletion of resources not updating freespace info
* Fixed: Satellite nullpointer exception on applying already deleted resource
* REST: Undocumented single resource paths will not return arrays anymore
* DRBD handler options are now always quoted.
* Added optional file system creation on DRBD or storage resources
* Support for NVMe over fabrics
-------------------------------------------------------------------
Tue Apr 9 01:11:16 UTC 2019 - nwang@suse.com
- Update to 0.9.5, match to linstor client/api 0.9.2
* Fixed: Multi-volume migration from old versions
* Fixed: Encryption layer didn't correctly decode the passphrase
* Fixed: REST-API 404 is now returned if a resource definition
was not found
* Satellite does now check what layers the host does support
* DRBD/config handlers will now set from the according property values
* REST-API: On create node if only 1 net-interface is given,
it defaults to satellite connector
-------------------------------------------------------------------
Thu Apr 4 07:17:31 UTC 2019 - nwang@suse.com
- Update to 0.9.4, match linstor client 0.9.1
* REST API
* Resource layers allowing volumes without DRBD
* Multiple network paths, configure more than one network path
* Retry on failed resources
* Added migrate disk option to toggle disk command
* Better resource ready detection
-------------------------------------------------------------------
Thu Nov 1 00:51:14 UTC 2018 - nwang@suse.com
- Update to upstream release 0.7.1
* Basic DRBD Proxy support
* Resource can now created in a batch (used for drbdmigrate)
* More synchronous operations (auto-place, storage-pool create)
* linstor-config: improved sql-script command
* Added list API for resource connection
* Fix: drbdsetup events2 parsing will be restarted if process dies
* Fix: resource create check available size
* Fix: Delete snapshots on node lost command
* Fix: SSL configurations files are now resolved from config directory
* Fix: Some race conditions are avoided
* Fix: Deleting thin volumes didn't correctly clean up
free space information
-------------------------------------------------------------------
Wed Oct 31 11:30:56 UTC 2018 - Jan Engelhardt <jengelh@inai.de>
- Replace description by something useful that does not rely
on the term "SDS".
-------------------------------------------------------------------
Wed Sep 26 07:12:43 UTC 2018 - nwang@suse.com
- bsc#1095668. First version in 0.6.5.
