GUI client/REST server: 
*** Commit history: 36e9e2516f..38e71f2710 *** 
NEW: Add an explicit type filter to the "dir" and "generate" CLI commands
BUGFIX: (src#36045) sm_cmd taskgen removes leading "/" from source for Linux PATH backup
NEW: (src#36044) CLI task generation can't activate VSS during task creation for Windows clients
bugfix: Fix "countDynamic" statement to be executed when the "Last Backup State" view is shown.
Merge remote-tracking branch 'origin/develop' into develop
BUGFIX: (src#35964) WebUI must show the correct amount of backups when fitlering for a specific task
new: (src#35955) Trying to delete a restore task where the backup task is already gone leads to an API error
BUGFIX: (src#35936) Saveset detail dialog for migrations are not shown
BUGFIX: (src#35923) NPE when using CLEAN DRIVE for a single tape drive
new: (src#35928) VMware restore does not check if the virtual machine already exist in case the overwrite mode is off
BUGFIX: (src#35787) vSphere restore fails if the VM name ends with '-six digits'
new: (src#35928) VMware restore does not check if the virtual machine already exist in case the overwrite mode is off
BUGFIX: (src#35923) Monitoring -> Restores: Oracle/BSR restores are not shown
BUGFIX: (src#35915) vSphere Restore in folders doesn't work
new: Add specific run configuration for starting Sesam Server without the remote buffer.
change: Allow to skip start of remote buffer via system property.
new: (src#35877) UI shows wrong note: Using single file restore is not possible, when compression or encryption is enabled!
BUGFIX: (src#35900) Report "Jobs Overview" doesn't show running jobs started before report time frame start
BUGFIX: (src#35899) Report "Available Media" shows no barcode if the media isn't in the library
new: (src#35837) Add option to avoid auto add of media pool during data store creation
change: Fix where clause changed to outer where clause for all filter properties.
new: (src#35837) Add option to avoid auto add of media pool during data store creation
change: Fix CLI generic request utility is calling endpoints declared as "GET" using the "POST" method.
new: (src#35837) Add option to avoid auto add of media pool during data store creation
BUGFIX: (src#35784) Command: sm_cmd remove saveset SF20260318021507916@6OejXH3XL0q -f 0 has no function
change: Add datastore example in CLI tests and documentation
change: Fix CLI is composing the GET .../{id} requests wrong.
new: (src#35264) Open api documentation does not display getAll
change: Attempt to unify the AI-related project general rules.
BUGFIX: (src#35859) vSphere: Exception during buffer refresh leads errors during browsing
Fix: Fix NPE when setting SessionId in ServerServiceImpl
BUGFIX: (src#35859) vSphere: Exception during buffer refresh leads errors during browsing
Fix: Fix NPE when setting SessionId in ServerServiceImpl
BUGFIX: (src#35815) Filtering migration/restore tasks results to error 500
CHANGE: (src#35822) Enhance credential management of WebUI and RMI
new: Added CLAUDE AI skill for generating changelogs
change: Update AI guidelines to avoid automatic commit of changes
new: Improve CLI output by adding converters to convert Double values in human-readable format
NEW: (src#35818) Allow to select if date/time values are either shown in server time or local time
change: Aktualisiere die Claude Statuszeilenanzeige
CHANGE: (src#35799) Reduce REST server response when requesting server information without authentication
new: AI - Added status line to Claude Code
new: AI - Added and enabled Claude superpowers plugin for project.
CHANGE: (src#35798) Enhance error handing of REST server when getting API requests
NEW: (src#35835) Allow to get any type of object by primary key (name) or UUID if the primary key is a string
CHANGE: (src#35829) Remove "executeSql" from public API
CHANGE: (src#35798) Enhance error handing of REST server when getting API requests
change: Update AI agent global guidelines
new: Add basic AI coding instructions and guidelines (Claude Code)
new: Add AI code review guidelines
CHANGE: (src#35798) Enhance error handing of REST server when getting API requests
BUGFIX: (src#35786) CLI help: sm_cmd help start - object analyzer is missing
Metadata Analyzer: Fix failing unit test
BUGFIX: (src#35788) REST server can be forced into shutdown by unauthorized user
BUGFIX: (src#35777) UI Server redirect allows host injection via headers
Metadata Analyzer: Fix NPE
new: Add "import schedule" command to import schedules from a JSON file
change: Keep custom sort order for "Tapes" table
change: Set the sort direction for orderBy clause for "Tapes" table
CHANGE: (src#35776) Introduce external library to generate UUID V7 in Java code
bugfix: Fix join of media and loader_contents table
NEW: (src#35749) Add API to generate integrity check report
change: Determine data store type from backend type where feasible. Also, add "Dell Data Domain" data store type to CLI help.
change: Remove obsolete data store type "SEP Si3 Deduplication Store" from CLI help
change: Allow reports to contain plain HTML code
NEW: (src#35184) Integrate CAPS/Keepit backup health indicator into the dashboard
bugfix: Fix immediate schedule filtering
change: Update SQL replacements
change: Fix Regex pattern for UUID in SQL replacements
change: Update SQL replacements
change: Add missing stub to DrivesServiceRestImpl to include count method
change: Fix implementation of "count(filter)" for drive groups and drives.
NEW: (src#35404) Show "last execution" information for tasks of any type
new: Set PostgreSQL dialect mapping
BUGFIX: (src#35680) - Deletion of a loader is not possible
NEW: (src#35699) Add datastore type for DD-Boost
change: Remove pagination logic from AlertsServiceImpl.find as this is already done on SQL level.
CHANGE: (src#35586) sm_cmd should fail if task with unknown task type is added
CHANGE: (src#35665) Foreign key constraint exception when changing name of copy task
CHANGE: For an immediate start, apply provided options to migration task instead of the migration event, if the migration task is newly created.
BUGFIX: (src#35680) - Deletion of a loader is not possible
change: Rename internal 'server' field to 'node' for consistency
bugfix: Correctly sort available reports by name when listing them
bugfix: If printing VMware vSphere data stores, the host systems are referenced by internal IDs instead of using the host system name.
change: Fix auto-generation of standalone REST API documentation
change: Return "list reports" sorted by report name
change: Fix auto-generation of standalone REST API documentation
BUGFIX: (src#35677) Trying to get the main protocol of init action returns stack overflow
change: (src#35662) Update DB:clients.msg, if client is 'activated'
NEW: (src#35468) - Add option to expire media
bugfix: (src#35673) Display more parameter for schedule and events in the Web UI
change: Always check for matches in 'attributes=[...]' also if not found as root attribute
change: Fix filtering and paging of notifications
change: Fix "list reports" returns template without associated meta-data
CHANGE: (src#35662) Update DB:clients.msg, if client is 'activated'
NEW: (src#35631) Add a secondary location for custom rythm report templates
BUGFIX: (src#35637) Schedule with a one-time execution and a command event must not change it execution date when editing it afterwards
BUGFIX: (src#35638) SQL exception in "restores/count" if regex search is used.
new: (src#35639) Notification "find" and "count" get different information
NEW: (src#33103) Handling of restore tasks, results and log files
change: Up-rev project versions and release names to V5.2 Build 5 (Artemis V5)
CHANGE: (src#35616) Deleting a backup task via WebUI may stuck an unnecessary long time
new: (src#35609) Add drives filter to AllResults Filter
new: (src#35432) Get event information for all tasks
change: Do not auto-add power state checks if the start/stop/shutdown actions do wait anyway
change: Create the media pool as read-only media pool if the associated data store is read-only too.
change: Media pool of type 'read' cannot be used as source pool for a replication.
change: (src#35574) Action 'cleanup' of a DataStore/Si3 is being executed, but not shown in DB
NEW: (src#35400) Add support for Virtuozzo virtualization task type to REST server and Web UI
change: Add alias for "getSessions" API
change: Add the mapping for shutdown VM
change: (src#32757) No error message is printed if vCenter is not reachable during browse
bugfix: Fix NPE in "Scheduling -> Next Events" when a term does not have a next execution set.
CHANGE: Rework updateTask method of migration service
NEW: (src#35400) Add support for Virtuozzo virtualization task type to REST server and Web UI
CHANGE: (src#35586) sm_cmd should fail if task with unknown task type is added
NEW: (src#35566) When stopping a virtual machine, a guest shutdown should be performed instead of hard power off
BUGFIX: (src#35584) The CLI command "refresh hypervisor " returns with "Invalid number of parameters. Expected (2) != Provided (1)."
change: Final definition of virtuozzo hybrid server backup type (Virtuozzo-HS).
bugfix: Fix SQL exception in "count" if regex search is used.
BUGFIX: (src#35555) Further migration is not possible if the original Saveset is no longer exists
CHANGE: (src#35568) Change 'check' option parsing in 'sm_robot'
bugfix: Fix NPE in MediaPoolEventServiceImpl::doStart when the event ID or name is given but the event does not exist
change: When configuring the authentication, clarify that the password sent is the plain text password.
change: Update AnalyzerParams to use DefaultsUtil.toBool for boolean parsing in the print output
NEW: (src#35566) When stopping a virtual machine, a guest shutdown should be performed instead of hard power off
change: Update 'start analyzer' example
change: CLI start analyzer parameter '-p' now also accept '1' as 'true'.
CHANGE: (src#35554) Change default quiescence mode to '2' for Hyper-V and VMware modules
bugfix: Set session 'created' time to the current time when cloning a session.
change: Fix exception in "countDynamic" when using search patterns
change: Use specific message numbers for retry and create of snapshots with or without quiescence.
CHANGE: (src#35543) Adds to the help for “sm_cmd import calender”
change: (src#35472) Set property "generated" for auto generated schedules
new: (src#35460) Create single API for activating and deactivating the DB authentication method
CHANGE: (src#35534) vSphere restore fails with: 'Failed to reconfigure the virtual machine'
 
Web restore assistant: 
*** Commit history: fc781816..fe217aec *** 
Merge remote-tracking branch 'remotes/origin/develop' into v5_2_0
change: remove console,log for release
bugfix:  (src#35810) Browsing of mounted Windows VM on Windows RDS must just allow to browse mounted drive letters
NEW: (src#35812) Handling of restore tasks, results and log files
bugfix: (src#35675) WebRA browser must not allow to browse higher than mount point
BUGFIX: (src#35619) Hyper-V restore option "clustered" is not visible if default target Hyper-V Server is already the cluster
change: add client to mount messages with server, rds
BUGFIX: (src#35659) Mismatch of VM name and corresponding Hyper-V task in Restore Assistant
BUGFIX: (src#35619) Hyper-V restore option "clustered" is not visible if default target Hyper-V Server is already the cluster
NEW: (src#35400) Add support for Virtuozzo virtualization task type to REST server and Web UI
bugfix: (src#35563) - Restore VM target setting for Proxmox are ignored
CHANGE: (src#35539) Radio buttons for File plus ACLs Restore Options
bugfix: (src#35531) - SFR from VM not work
BUGFIX: (src#35110) Wrong filtering when adjusting date range selection for Proxmox tasks/backups
change: (src#35540) Missing tooltip for function "Save new directory" if selecting a custom target restore path
 
Web client: 
*** Commit history: f60ed9581..d22de9b0e *** 
Revert "CHANGE: (src#35823) If the last element in an FDI chain is marked as “error,” you cannot delete the chain"
Merge remote-tracking branch 'remotes/origin/develop' into v5_2_0
CHANGE: (src#35823) If the last element in an FDI chain is marked as “error,” you cannot delete the chain
change: (src#35987) When backing up a Citrix VM, the text “Source path” in the '5 Summary' is wrong
Show whole path of Hyper-V disks at WebUI when selecting exclude
BUGFIX: (src#35938) No possibilty to delete a scheduled command (e.g. DB2 command)
bugfix: correct missing partial commit in branch
bugfix: (src#35909) Drive configuration with SiS datastore produces corrupt ini files
bugfix: (src#35904) Dont show comma after last item of "Filter by backup level" and "Schedules"
BUGFIX: (src#35901) "Media Action" - "Close tape" have the possiblity to select a drive (not required)
bugfix: (src#35891) If only a schedule is selected when creating a backup via the backup assistant, there is still a comma after it
bugfix: (src#35887) Backup Assistant: Account and password fields are incorrectly highlighted in red after correcting a duplicate task name
bugfix: (src#35886) SAP ASE Automatically Assigns Backup Level Type C by Default
Use details/summary html widget for gc message.
change: set filterMediaPools only when filtering by pool
bugfix: (src#35861) Clicking at the "update intervall" button at WebUI results in strange behavior
CHANGE: (#35855) Do not show full GC details in data store properties tab
change: improve bookmarklet after gitlab upgrade
bugfix: (src#35797) Sorting via message in 'Last 4 media actions by type' table does not work
change: remove some console.log statements
bugfix: (src#35794) WebUI may not show warning box at backups when backup type switched
BUGFIX: (src#35746) Filtering drive group via overflow menu does not work
change: update icons for media actions assistant
change: update icons for media actions assistant (subject to review)
NEW: (src#35761) WebUI Backup Assistent: Additional option for Nutanix backup task
change: use get and not find to get loader object
bugfix: use get and not find to get media/tape object
Revert "wip: (src#35184) Integrate CAPS/Keepit backup health indicator into the dashboard"
bugfix: (src#35514) Kopano mailbox selection -> adding slash to mailboxname and doesn't actually save changes
bugfix: (src#35495) Within the backup copy assistant, enable the parallel migration execution by default
bugfix: (src#35739) Backup assistant must use VSS as default for Windows path backups
bugfix: (src#35741) Task assistant must not show CBSR host name
wip (src#35734): Consolidation task creation
change: comment out info about migration betweeen Si3 pools
CHANGE: (#35700) Change color for locked tape media
CHANGE: (#35500) Advanced options are missing in summary when creating a backup task in Web UI
wip (src#35734): Consolidation task creation
CHANGE: (#35700) Change color for locked tape media
CHANGE: (#35500) Advanced options are missing in summary when creating a backup task in Web UI
change: used mono font for dedeupRate, fix migration tasks filter
fix: (src#35725) Add Info Icon to Medatata Alerts View in the WebUI
CHANGE: (#35700) Change color for locked tape media
bugfix: (src#35514) Kopano mailbox selection -> adding slash to mailboxname and doesn't actually save changes
bugfix: (src#35495) Within the backup copy assistant, enable the parallel migration execution by default
bugfix: (src#35739) Backup assistant must use VSS as default for Windows path backups
bugfix: (src#35741) Task assistant must not show CBSR host name
Revert "wip: (src#35184) Integrate CAPS/Keepit backup health indicator into the dashboard"
wip (src#35734): Consolidation task creation
change: comment out info about migration betweeen Si3 pools
CHANGE: (#35700) Change color for locked tape media
CHANGE: (#35500) Advanced options are missing in summary when creating a backup task in Web UI
wip (src#35734): Consolidation task creation
CHANGE: (#35700) Change color for locked tape media
CHANGE: (#35500) Advanced options are missing in summary when creating a backup task in Web UI
change: used mono font for dedeupRate, fix migration tasks filter
fix: (src#35725) Add Info Icon to Medatata Alerts View in the WebUI
CHANGE: (#35700) Change color for locked tape media
change: rename Logdatei to Protokolldatei or Log-Datei
change: (#35704) Translate the column "position" in media table
change: remove message about Si3 Recovery log for sanity error.
BUGFIX: (src#35415) Storage Assistant shows Si3 size limit warning when creating a Path data store too
new: (src#34019) Add easy way to copy existing objects like schedules or tasks
BUGFIX: (src#35689) - Filter deactivated drives from source/target drive when editing a copy task
bugfix: (src#35651) Handle uninitialized variable in media result detail
BUGFIX: (src#35415) Storage Assistant shows Si3 size limit warning when creating a Path data store too
NEW: (src#33103) Handling of restore tasks, results and log files
bugfix: (#35620) Update license info popup cypress tests
CHANGE: (src#35571) - Remove unnecessary "Unload drives" option when running the archive adjustment by barcode only and add unknown media handling option
 wip: (src#35455) Improve display and edit of copy tasks
Conversion to local time zone does not work correctly
change: make saveset column wider for locations table
change: add labels property to custom filter component
bugfix: (#35548) Show active user sessions in WebUI
CHANGE: (src#35517) Advanced options are missing in summary when creating a backup task in Web UI
BUGFIX: (src#35553) Disable the START button if incorrect values are entered for the backup cancellation time
change: (#35517) Textarea in command task for command is too short
change: upgrade Angular to 19.2.18
NEW: (src#35001) Media/media pool event type is not shown in next events table
BUGFIX: (src#35482) - Time period calculation in copy assistant is not correct when based on sesam days
new: (src#35455) Improve display and edit of copy tasks
Merge remote-tracking branch 'origin/develop' into develop
BUGFIX: (src#35562) - Deleting a saveset from saveset properties is not possible
change: (src#35561) Sort selection lists of tasks, groups, clients and others in Assistants
bugfix: (src#35528) WebUI: "Next" button remains disabled after switching time period mode from an invalid state
bugfix: (src#35529) WebUI:"Delay until next data store purge" checkbox is clickable in Copy Assistant
bugfix: (src#35501) In Backup Client Selection NEXT is sometimes incorrectly grayed out
bugfix: (src#35504) The password verification in the Backup Assistant under Encryption is not reset.
 
Kernel: 
*** Commit history: f464afefdd..acfec04490 *** 
change: Avoid SQL error during Sesam update from 5.1 to 5.2 on Windows, because 'VIEW:all_results.estimated_stop_time' doesn't exist in 5.1 DB, but tried to be accessed by 5.2 binaries during update
BUGFIX: (#36038) BSR Windows: If an excluded file is a hard link, the blocks of the shared inode are not backed up, and the linked files cannot be recoveredFile contents are not backed up if there are excluded hard links to them. Solution: Do not longer set 'ignored_file_pattern' for '*.tmp' '*.omg' and '*.~'.
bugfix: (#32972) add ini entry to enable detailed logs during s3 restore; clean up logs, update docs
bugfix: (#32972) fix bug on skip with caching of last skipped link
bugfix: (#32972) change queue type; extend logs
bugfix: (#32972) print buffer on restore
bugfix: (#32972)  fix race condition on work cancellation
bugfix: (#32972)  fix rewond concurrent issue; add logs in chunk reader
bugfix: (#32972) extend logs; change clean up logic after skip
bugfix: (#32972) extend logs
new: Increase SEP sesam version. Old version is 5.2.0.20, new version is 5.2.0.21.
bugfix: VFS fail with 'Regex compile failed' then MTF find HardLink and set Major/Minor in HEX format
change: (#35933) Update DB:defaults.mtime to valid timestamp before update from Sesam <= 5.1
bugfix: (#36006) Avoid 'division by zero' error in 'remaining_exec_time()' SQL function
change: (#34283) sm_ssh cores sporadically during OpenSSL function calls. File log should be written only on request
bugfix: (#34283) sm_ssh cores sporadically during OpenSSL function calls. -vvv fix
bugfix: (#34283) sm_ssh cores sporadically during OpenSSL function calls. Enable file logging, thread terminate improvements
bugfix: sm_ssh core due to cleanup race condition
BUGFIX: (#35935) sm_lis_items uses version of FULL backup instead of INCR if 'mtime' is identical. If item matches then check with timestamp from saveset
BUGFIX: (#36008) Add missing sbc_vmware_mount.exe to client installation package to enable mount on client.
BUGFIX: (#35892) The default timeout of 600 seconds of sm_break2 is too long when being called by sm_qm_main stop handle function. In this case the timeout is set to 150 seconds.
new: Increase SEP sesam version. Old version is 5.2.0.19, new version is 5.2.0.20.
change: Reduce Si3 logs for S3 storage backend
bugfix: (#35986) Encrypted and compressed path backup fails with 'Error: Invalid MTF stream block checksum'. EH frame header most set ulength to umcompressed size
BUGFIX: (#35892) sm_shutdown including sm_break2 has not been executed when stopping Sesam service. Solution: Execute sm_shutdown before all subprocesses of Sesam service are stopped to ensure that all running jobs are stopped gracefully and no new jobs are started during service stop.
BUGFIX: (#35892) To standardize the workflow of 'sm_main stop' and 'net stop sm_main', 'sm_main stop' will call 'net stop sm_main' from now on. Additionally some error handling was added.
BUGFIX: (#35914) After updating to Artemis V4 (from 5.2.0.12), the customer received a license violation notice due to having too many mailboxes. Bugfix: Use pattern '[{database}]' to locate amount of mailboxes.
bugfix: (#35968) An MSSQL backup task that specifies only the database name results in an 'all' backup. Solution: Reimplement database check. Problem introduced with commit 40296c1 (#23910) Show MS-SQL recovery model in 'dir' output
bugfix: (#35935) sm_lis_items uses version of FULL backup instead of INCR if 'mtime' is identical. Take dirs from latest saveset also
bugfix: (#35935) sm_lis_items uses version of FULL backup instead of INCR if 'mtime' is identical
bugfix: (#35597) simplify PowerShell command building and enhance test coverage for script execution
change: (#35690) split installer package and remove redundant client wrappers
new: (#35690) install sesam ssh authorized_keys from local file
change: (#35690) clean up staticcheck findings
change: (#35690) remove namespace pvc data double-hop
new: (#35690) expose controller ssh port and use stable startup entrypoint
new: (#35690) use snapshots by default for namespace PVC backup
change: (#35690) enhance logging format
new: (#35690) add archive export/import for namespace backups
new: (#35690) fix restore order and sync hang, rewrite backup/restore to use sbc
new: (#35690) add namespace backup and restore with wildcard RBAC
change: (#35690) add better async lifecycle with operation APIs
change: (#35690) add CR-driven cancel, use sesam-cli sbc for archive, and harden helper pod lifecycle - Add spec.cancel field to SnapshotOperation with POST /v1/operations/:name/cancel endpoint - Implement synchronous operation execution with cancel context support - Use sesam-cli-rhel10-x86_64 as helper image with sleep infinity CMD (fixes immediate exit) - Use sbc commands for archive/restore (sbc -b/-r) with copyFromPod/copyToPod transport - Add container-ready check in waitHelperPodRunning to prevent exec race - Prevent double-run of completed operations via Running condition check - Add make redeploy target that builds, loads to minikube, and reinstalls - Add CLI cancel command (snapshot cancel --wait) - Add unit and integration tests for cancel semantics - Update docs to reflect sbc workflow and cancel API
change: (#35690) add CR-driven cancel, switch to busybox+tar, and harden helper pod lifecycle
change: (#35690) update snapshot tests to use operation package functions
change: (#35690) standardize API response envelope and split controller package
change: (#35690) refactor: eliminate hardcoded constants, duplicate functions, and dead code
change: (#35690) centralize hardcoded constants and extract error codes
change: (#35690) use sesam-cli-rhel10 base image and remove registry prefix
change: (#35690) Add RBAC fixes, E2E test improvements, and internal flow docs
new: merge sshd
bugfix: (#27629) Redirected DB2 roll-forward fails, because new DB name is used in query. Bugfix: sqluvint() Avoid to call SET_ERR_REASON(return_code, rc ) if everything was ok
bugfix: (#27629) Redirected DB2 roll-forward fails, because new DB name is used in query. Bugfix: Avoid to set return code with rc=SQLUV_DEV_ERROR (18) in case of BSA_RC_NO_MORE_DATA (18)
bugfix: (#35763) fix occk workers calculation
bugfix: (#35763) fix occk workers calculation; extend docs
bugfix: (#35406) Changed diskpart mount on Windows so that script to assign letters to new volumes is executed only once. Before it was loop with diskpart command appending to the script and the script was executed on every iteration of the loop, That caused diskpart error to be written to the log, without other consequence on the functionality.
bugfix: remove redundant imports
bugfix: count links in progress logger instead of pages during OCCK instead
bugfix: restrict max concurrent opened peges for object check during OCCK; optimize data usage during OCCK
bugfix: wait for page during OCCK if it's sweep not finished
BUGFIX: do not use page cache during OCCK
BUGFIX: fix cache weighter function; add entry expiration after access; update caffeine library version
change: (#34373) Correct typo in error message for VM locking, align tag removal message with tag addition message
BUGFIX: (#35930) SESAM_BACKUP with LVM snapshots does not remove mount point directory. Use extra list mount_points to memorize directories which must be removed
bugfix: (#35917) Backup with LVM converts exlude pattern '\' into '/'. Set 'o.exclude' to adapted 'sbc_excludes'
bugfix: (#35917) Correct variable name
BUGFIX: (#35917) Backup with LVM converts exlude pattern '\' into '/'. Solution: Do not replace '\' with '/' in exclude patterns
BUGFIX: (#34879): nutanix restore fails after first retry: wait for correct task uuid after retry
bugfix: (#35784) Add Si3 OCCK to DB:media_results
BUGFIX: (#35916) Backup using sub savesets must not be switched to FULL in DB, if sbc config backup switched to FULL - FIX: after return from sbc_pre, config files prepared by a sub-module are backed up by executing SBC backup command by sbc_proxy. Backup level for the SBC command is set to level COPY, regardless of the main backup level for sub-modules: 'XEN', 'PXMX', 'NTX', 'KVM', 'RHEV', 'VZO', 'ONE', 'AZURE'
change: Set backup level always to copy for backup of VM config for Proxmox. Set at the same place in sbc_proxy after return from sbc_pre also for XEN
bugfix: (#32972) use trySuccess to cancel queued promises on skip
bugfix: (#32972) fix race condition when future was canceled in the moment of it's completion
BUGFIX: (#27581) Set STPD DATA timeout by XBSA. Problem: Timeout 59 sec is used as default value. Solution: Default value should be used for KILL only
fix: (#31983) fix race condition during pcck missing page check
fix: add mode to occk and fsck statuses
change: Disable build of DDBoost client in v5_2_0 branch
NEW: (#35784) Add OCCK status to DB:media_results
new: Do not call 'CLEANUP_QUOTA()' in 'DB_ACCESS()' function, if query starts with 'UPDATE' (uppercase)
bugfix: (#27629) Redirected DB2 roll-forward fails, because new DB name is used in query. Bugfix: Function sqluvend() must call BSAEndData() for BACKUP/READ_BEGUN
change: Disable DDBoost build in 5.2.0
bugfix: (#32972) clean up s3 restore logs
bugfix: (#27629) Redirected DB2 roll-forward fails, because new DB name is used in query. Bugfix: Function sqluvend() must call BSAEndData() for BACKUP/RESTORE
BUGFIX: (#27629) Redirected DB2 roll-forward fails, because new DB name is used in query. Bugfix: Function sqluvend() must not call BSAEndData() if read not yet begun. sm_sbc_com must not return backup with state 'a'
new: (#31983) add docs for fsckCheckMissingPages option
bugfix: (#34879) Nutanix 7.3 Prism Central - removeVM fix for Prism Control
NEW: (#35763) Reset manually adapted Java heap memory setting for Si3-NG
change: Build RPM release kits without build count again
bugfix: (##32972) add separate method to start restore with disabling buffering
bugfix: (#35854) Si3-NG replication over https doesn't work. Secured replication init fail
new: (#35790) added a default value to config map
new: (#31983) enable fsck check through config option
new: (#31983) add missing pages to corruption store on startup
new: (#31983) check missing files during incremental fsck
new: (#35790) added a default value to config map
bugfix: fix build warnings
bugfix: (#35763) add workers based memory calculation
fix: (#35676) log only during opening a connection
bugfix: (#35763) fix backups and replications calculation
bugfix: (#35763) add workers based memory calculation; extend docs
bugfix: (#32972) make disable_buffered  field as Option so old client version do not fail restore start
bugfix: (##32972) add separate method to start restore with disabling buffering
bugfix: (#35854) improve response handling in replication http client; optimize connections usage; improve error handling and logs during replication
bugfix: (#32972) allow to choose restore method on fly
BUGFIX: (#27629) Redirected DB2 roll-forward fails, because new DB name is used in query. Bugfix: The last char of the database name was cut off
BUGFIX: (#35857) Long running Proxmox restore fails with AuthenticationError: Couldn't authenticate user - Catch authentication exception after disk restore, on API call to POST VM config. Reconnect on exception and repeat the API call
NEW: (src#35818) Allow to select if date/time values are either shown in server time or local time
new: (#35763) add docs on heap space memory calculation
bugfix: (#35730) Fixed uuid libs
change: Merge 'kernel' sources from 'develop' to 'v5_2_0' branch
Merge commit 'cc4c2906c22852861d689456ca2472e1526951c1' into v5_2_0
new: (#32972) add ini entry to manually disable buffered s3 restore
new: (#35254) Format common ddboost_client_lib trace messages by using DDBoost API ddp_errstr to print human readable error messages
change: (#35408) Changed format of Proxmox RBD snapshot file name to <times>/<short server name>_<task name>.cbt If file with the new name format is not found during INC backup, search is performed for the old style name. After the next FULL backup, only new names are used
new: (#35254) Implemented common interface for listing objects on DDBoost storage unit. Improved trace logs in common interface
change: (#35832) Use '0/1' for DB:*.locked column instead of '+/-': Add new medium with locked=1
new: (#35690) add workflow documentation for control-plane pattern and operation phases
new: (#35690) adopt SnapshotOperation control-plane workflow with dedupe and tests
bugfix: (#35712) Windows Path backup failed with error: cUrl: [HTTP error before end of send, stop sending / Server response: Can't find chunk - not enough data to skip CRLF. Rework chunk size detection
Remove doubled where statement
bugfix: DI over HTTPS with log level >2 fail with error
bugfix: Correct usage of MEDIA_NOT_LOCKED macro
new: (#35690) add --sync wait and fix rollout readiness checks
NEW: (#34196) Migration of savesets on a locked tape media doesn't work if EOL of savesets is in the past. Overrule EOL check, if medium is locked
new: Use 'DB:migration_events.label' during saveset migration too (not only from DB:migration_tasks)
new: (#35790) Add an API call to return all current si3 configuration settings
CHANGE: (#35832) Use '0/1' for DB:*.locked column instead of '+/-': Use macro 'MEDIA_NOT_LOCKED' to check lock condition
CHANGE: (#35832) Use '0/1' for DB:*.locked column instead of '+/-'. Update DB:results and DB:media during Sesam update
CHANGE: (#35832) Use '0/1' for DB:*.locked column instead of '+/-'
change: (#34879) Add Nutanix 7.3 Prism Central support with PE auto-detection (single PE required for execution, same credentials required)
CHANGE: (#35832) Use '0/1' for DB:*.locked column instead of '+/-'
BUGFIX: (#35826) Fix xore dump in 'RESTORE_PROT() function, if called with 'DATE' parameter
Fix compiler format string warnings
change: Remove deprecated 'rcsepsesam' link for SuSE kits
change: (#35380) Log time during RPM update. Escape '%S', because used by RPM build as macro internally
Revert "Revert "change: (#35380) Log time during RPM update""
new: (#35254) Added common ddboost_client_lib function ddboost_get_storage_info(). Simplified ddboost_test program to test calls to ddboost_client_lib functions
Revert "change: (#35380) Log time during RPM update"
change: (#35380) Log time during RPM update
CHANGE: (#35380) Set DB:sesam_events to 'unlogged' for faster import of Sesam DB during update
BUGFIX: (#35225) Throughput and data size is not updated during first execution of a backup task. Correct SQL update statement, if 'DB:results.estimated_size=NULL'
bugfix: (#35817) When performing a BSR backup, Sesam must specify the correct network interface for bootmedium.xml. Bugfix: Use first retrieved interface (requested family or IPv4)
new: (#35790) Add an API call to return all current si3 configuration settings
change: (#35625) Revert update CBFS driver back to 6.1.181
change: (#35790) fixed JSON response format
bugfix: (#35712) Windows Path backup failed with error: cUrl: [HTTP error before end of send, stop sending / Server response: Can't find chunk - not enough data to skip CRLF. Rework chunk size detection
new: (#35690) add install timeout flag
new: (#35690) add check command for prereq-only validation
change: (#35690) rename run command to serve for HTTP API startup
change: (#35690) migrate CLI to Cobra and group commands by folder
change: (#35690) update module path and image references to use go-snapshotter
change: (#35179) set default pagePostCreationTest, gc.sweepPercentage, gc.sweepLimit settings
Use 'M_X_SNPRINTF' macro
BUGFIX: (#35817) When performing a BSR backup, Sesam must specify the correct network interface for bootmedium.xml. Bugfix: Use the IP interface which matches the client hostname
bugfix: (#35712) Windows Path backup failed with error: cUrl: [HTTP error before end of send, stop sending / Server response: Can't find chunk - not enough data to skip CRLF. Rework chunk size detection
bugfix: fix build warnings
change: sm_mount uses DB:results.subtask_flag to determine if VMDK saveset ID is from a config saveset, instead of checking if ID starts with the letter "S"
new: include the sm_integrity_check in the Linux build.
change: (#35228) Create overlay LIS file for FDI chain of BDA backups. LGC file name changed (if -o consolidate) [_SG] -> _CON
change: (#35228) Create overlay LIS file for FDI chain of BDA backups. LIS file name changed [_SG]_GEN -> _CON
bugfix: (#35782) avoid killing process groups for exited workers
bugfix: enable logs in DiskChunkReader
CHANGE: (#35811) Generation restore failed with 'ERROR: Cannot find matching LIS line for item [...]' due to empty SG.lis file. CHANGE: End with message 'empty LIS ....'
bugfix: fix buffer slicing for segment chunk after skip; extend logs
BUGFIX: (#35791)  Created unique Bitmap names during Proxmox NBD backup. Additional fix. Unique bitmap name was not found although it existed, use query_block_data_full in 1st search for existing bitmap
bugfix: fix buffer slicing for segment chunk on restore; extend logs
WIP: (#30371) Synthetic full for file based backups. WIP: Missed '#TRAILER' and last line before '#TRAILER' line in LSC
BUGFIX: (#35791)  Created unique Bitmap names during Proxmox NBD backup. If incremental backups still have old non-unique names for bitmaps, use those old names until the next full backup
change: (#35763) add multiplier on total calculated heap space size. Default values changed - 64 backups and 32 replications
WIP: (#30371) Synthetic full for file based backups. WIP: LSC must always use the sorted LIS order
change: (#35763) add multiplier on total calculated heap space size. Default values changed - 64 backups and 32 replications
bugfix: (#35763) add multiplier on total calculated heap space size. sm_dedup_interface
BUGFIX: (#35763) add multiplier on total calculated heap space size
new: Adapt info message for migration, if only one saveset should be copied and copy already exists on target pool
new: (#35202) build sm_integrity_check for Linux.
BUGFIX: (#35763) increased base size in heap space calculation; restrict maximum count of workers for OCCK, PCCK, GC, Repair
change: get rid of warning due to possible nil value of err.
bugfix: Log original command line of 'sm_lis_items' execution
BUGFIX: (#35764) Mount of XenServer VM backup shows only data from base full - sm_mount did not add BITMERGE_BITMAP to the MAP file. That happened because it did not use migrated saveset ID when parsing LIS file in function write_vhd_map_file()
new: (#35790) tests fix
new: (#35790) added source info (whether default or ini file)
new: (#35790) added new command in rust client
new: (#35790) added new API route to retrieve server config
Fix log of SQL query list
new: Execution of multiple SQL queries in a transaction by Python 'sm_db()' function
change: (#35408) Removed confusing log from Proxmox NBD backup about creating Sesam snapshot name and removed creation of nbd file in snapshot_config_dir.Snapshot are not created for NBD since we developed dirty-bitmap functionality
bugfix: (#20345) Make BSR bootimage usable in a network environment. Substitute all occurences of '&', '<' and '>' with html entities
BUGFIX: (#35774) A failure at a vSphere restore fills up the file system. Problem: Error while single item restore does not abort restore its self, but try to restore next item. Solution: In case of CBT/BDA restores abort it if any item restore fail
WIP: (#30371) Synthetic full for file based backups. Write first line 'Tape ID ...'
WIP: (#30371) Synthetic full for file based backups. Remember Meta data LISv1 of recent saveset and use it in LSC file
BUGFIX: (#20345) Make BSR bootimage usable in a network environment. Problem: Network adapter 'Intel(R) EC1000S 1.0GbE Connection & Intel(R) EC2500S 2.5GbE Connection' with '&'. Solution: Adapt '&' to '&amp;'
change: (#35698) Switch order of parameters for ddboost_close_session() to make code more consistent with ddboost_init_session()
change: (#35349) combine the ps1 and cmd files in the output list of the file check.
BUGFIX: (#35764) Mount of XenServer VM backup shows only data from base full - sm_vfs function bitmap_append_bitmap() did not read bitmap from the main saveset. The reason was that initializeMAPPER_FS() did not set size for TreeNode entry for the bitmap file, it was 0
change: (#16202) Saveset consolidation for vSphere. In case of vSphere and LSC creation FDI chain should be ignored
change: (#35730) Add UUID v7 C function. Merge all uuid files in one
new: List all CLI commands supporting "-f <file>" option in C wrapper.
change: (#35202) also accept -n 'text' as input.
change: (#16202) Saveset consolidation for vSphere. In case of vSphere and LSC creation FDI chain should be ignored
change: (#35202) fix the SetLevel function of log4go.
change: (#35202) set default output format to text and exit with failure if --clean paramter is not valid.
change: (#35202) add sm_integrity check also to client, and gui package.
change: (#35202) combine the three parameters dryrun, move, and delete to one parameter called clean. In addition hide not functioning parameters from help.
BUGFIX: (#35425) If compression is enabled, selective file restore is not possible. Solution: Convert LisV2 to LISv1 in case of compressed saveset
Remove 'Debian Buster' from release build
change: (#16202) Saveset consolidation for vSphere. In case of vSphere and LSC creation FDI chain should be ignored
new: (#35775) Add trigger to set 'DB:sesam_events.client_id/schedule' from 'DB:results'
new: (#35202) Add sm_integrity_check to server, gui, and client Windows kits.
Fix Sesam version file
change: Update Jenkins release build files to current v5_2_0
change: Correct Jenkins build rules for 5.2.0 release
change: Correct Jenkins build targets for 5.2.0 release
Fix merge conflicts from develop -> v5_2_0
new: (#35254) Add DELL DD-Boost support for SEP sesam. Add bin\sms\libDataDomainoncrpc.dll, libDataDomainpthread.dll and libDDBoost.dll to ISM build rules
change: (#16202) Saveset consolidation for vSphere. In case of vSphere and LSC creation FDI chain should be ignored
new: Increase SEP sesam version. Old version is 5.2.0.19, new version is 5.2.0.20.
new: Increase SEP sesam version. Old version is 5.2.0.18, new version is 5.2.0.19.
BUGFIX: (#35656) Fixed creating NBD dirty-bitmap for Proxmox8. Node name was not correctly extraced from json structure
change: automatically added/removed files to signed file list.
new: (#35254) Add DELL DD-Boost support for SEP sesam. Remove ddboost_test
new: (#35254) Add DELL DD-Boost support for SEP sesam. sm_dedup_interface initial commit
bugfix: (#35780) use proper value to calculate total before dedup size during async delete operation; remove AtomicLong usage
new: (#35254) Add DELL DD-Boost support for SEP sesam. sm_dedup_interface initial commit
BUGFIX: (#35778) Selective generation restore from path backup fails if drive used for backup doesn't exist anymore
CHANGE: (#34758) Allow multiple migrations to same media pool for tape drives again
change: (#35202) change the short name and long name as well as the behavior of several command line parameters. Now the global switches can also be passed after the main argument.
new: (#35690) Add E2E Test Guide for go-snapshotter on Minikube
new: (#35690) implement CSI snapshotter
bugfix: (#35730) added UUID v7 generator. Fix Windows build issue
new:(#35730) added UUID v7 generator
new: (#35775) Add trigger to set 'DB:sesam_events.eol' from 'DB:results'
change: (#35254) Build ddboost_test so that it can execute functions from ddboost_client_lib. Temporary defined trace in ddboost_client_lib as a macro that prints to stderr
new: (#35775) Add trigger to set 'DB:sesam_events.client_uuid/task_uuid' from 'DB:results'
change: (#35254) Renamed ddboost_client.o to ddboost_client_lib.o
bugfix: (#35656) Added return from Proxmox NBD function to create dirty-bitmap
BUGFIX: (#35656) fails: Error creating dirty-bitmap for [..] can only concatenate str (not "NoneType") to str - If dirty bitmap cannot be created, log meaningful error message and exit backup with error
bugfix: set default value for pagePostCreationTest in application.conf to be shown in logs during startup
new: (#35254) Add DELL DD-Boost support for SEP sesam. sm_dedup_interface initial commit
change: Log command line by 'sm_cmd' in loglevel 1
NEW: (#35759) Do not use saveset from inactive media pool for restore
NEW: (#35698) Backup and restore using DD-Boost API - implemented common interface functions ddboost_create_file(), ddboost_open_file(), ddboost_close_file(), ddboost_write_data(), ddboost_read_data()
change: spelling correction in messages regarding source-side deduplication
bugfix: (#35690) Fix exit-status handling in notifyClientConnectionRefused: send only status 255 for connection refusal
change: (#35202) extend the parameter description for several parameter, change some parameter names and types.
new: Log command line by 'sm_cmd'
change: add Go generated files to ignore list
new: (#35254) Add DELL DD-Boost support for SEP sesam. Add ddboost_test app for Windows
change: add an enum for the filetype and fix the test cases.
wip (#35698) Add STPD ini configuration for DDBoost
change: keep only one GO build rule
change: (#35349) also consider bat files and make sure that the file ending is compared case insensitive.
change: use the same GO_BUILD_RULE_MAIN build rule for building sm_sshd2.
change: undo the rename of main.go to sm_integrity_check.go and use the general build rule GO_BUILD_RULE_MAIN for building sm_integrity_check
change: switch GO 1.23.4 -> 1.26.1
change: rename main.go -> sm_integrity_check.go to use one GO build rule for all modules
bugfix: (#35111) sbc hangs in sbc_find operation if data sending failes due to connection reset. Solution: Signal sbc_find with kill -9 if operation status already set to Error
bugfix: fix sm_integrity_check build
new: separate msvc-2026 dir for VS2026
bugfix: (#35111) sbc hangs in sbc_find operation if data sending failes due to connection reset. Solution: Signal sbc_find with kill -9 if operation status already set to Error
bugfix: (#35755) A deadlock may occur if sm_backup calls sm_init shortly after starting sm_sms_watch. Solution: Release when reading or writing is complete
new: (#35690) Add Dockerfile and build script for Sesam backup client on AlmaLinux 10
new: (#35202) include version information, an icon and a description for the sm_integrity_check executable during build.
BUGFIX: (#35755) A deadlock may occur if sm_backup calls sm_init shortly after starting sm_sms_watch. Solution: Only occupy the semaphore ctrl_drive# when reading from or writing to a medium.
NEW: (#35698)  Backup and restore using DDBoost API - implemented interface to close DDBoost session. Renamed ddboost_client.c to ddboost_client_lib.c. Added ddboost_client_lib.h
NEW: (#35698)  Backup and restore using DD-Boost API - implemented interface to initialize session to DDBoost
change: (#28875) Added guest_tools and power state to AZURE sbc_dir
BUGFIX: (#31855) Show the right error message in not-log (BSR Windows). Improve error message handling: 'c000000d: An invalid parameter was passed to a service or function.  the drive letter cannot be determined'
new: (#30015) Add information about symbolic links to LISv3 file. Use 'l' for sym-link
NEW: (#35254) ddboost_test executes a test that writes data to storage unit SEP
Use save M_X_SNPRINTF() macro
change: (#26899) Semaphore function with signal handler. Windows: Move sleep to end of loop
bugfix: (#26899) Semaphore function with signal handler. Add 500 ms sleep to avoid CPU-intensive loops.
bugfix: add proper error handling in s3Version cleaner to not fail when retention period is not set
bugfix: fix initial total object size calculation
new: sm_sesam get_file_name: Allow return of LIS file name without checking existance
new: Python get_lisfile_v2() function: Allow return of LIS file name without checking existance
change message text
change: (#35719) Use backend_pools() API to calculate free and used space for storage policies in VZO sbc_dir
bugfix: If an error occurs during postgresql update the error has to be passed to installshield.
bugfix: (#35674) the popup about a required reboot should only appear if the installation/update is executed manually, i.e. when UI level is larger than or equal to 4.
NEW: (#35693) Limiting bandwidth during migrations
bugfix: Only display the postgresql error message if the postgresql update really failed.
bugfix: (#35717) Avoid zombie processes of sm_sms_watch on Linux during IDLE time of drive. Logic improved.
bugfix: (#34815) Correct escape of single quote in SQL query: Allow use of functions in insert statement, e.g. 'sm_uuid()::VARCHAR'
bugfix: remove obsolete code
change: (#35727) Create UUID v7 instead of UUID v1/4 by 'sm_uuid()' SQL function
change: (#35710) semaphore get_free_devnbd implemented for KVM in the same way as for PVE
NEW: (#35724) Export small Sesam DB bootstrap information (no full DB dump), if DB:tasks.backup_srv_options='bootstrap-format-small' is set
bugfix: use atomic value to calculate total before dedup size
bugfix: (#32972) fix issue when restore wait for buffering while file reading finished
bugfix: (#35717) Avoid zombie processes of sm_sms_watch on Linux during IDLE time of drive. Solution: Set iReadPerfReturned = i_READ_PERF_INFO()
BUGFIX: (#35717) Avoid zombie processes of sm_sms_watch on Linux during IDLE time of drive. Solution: In the case of a very short execution, the idle counter was not reset to 0.
new: (#35676) add request method and full uri
new: (#35676) removed static string
new: (#35676) add logRemoteAccess directive to log remote TCP addresses
NEW: (#35254) Separated ddboost_test from ddboost_client, added new file ddboost_test.c that is a copy of demo from Dell-EMC
NEW: (#35254) Added link to DDBoost in rule to link ddboost_test
NEW: (#35254) Added target to build ddboost_test
NEW: (#35254) Added ddboost_client and ddboost_rest_api to common build
NEW: (#35254) Build ddboost_client.o and ddboost_rest_api.o in common/ddboost
Revert "NEW: (#35254) Compile correct common file for DDBoost. ddp_rest_api.c and ddp_rest_api.h link needed APIs exposed by libDDBoost.so and DDBoost client can get access to them via ddboost_rest_api.o"
change: (#35349) use the correct length of the SESAMDEV folder to replace the beginning of the path with the \bin\db folder.
change: (#35349) include also all files of the postgresql zip file in the filelist for Windows during build process.
change: added Virtuozzo network listing in sbc_dir
change: added KVM network listing in sbc_dir
bugfix: do not fail if /tmp/si3validate object not exists after validation; deny object deletion request if corresponding object already in deletion queue
bugfix: (#32972) improve error handling
NEW: (#35254) Compile correct common file for DDBoost. ddp_rest_api.c and ddp_rest_api.h link needed APIs exposed by libDDBoost.so and DDBoost client can get access to them via ddboost_rest_api.o
bugfix: fix the issue with local page uploading to s3 while it used by page body reader; block StartUploading command execution while previous still running
CHANGE: (#35714) The open file sm_dedup_error-{dn}.log must not be removed. Solution: Force the creation of error log files with timestamps by avoiding the <file> tag for the ERROR_APPENDER.
change: (#34071) Groupwise backup without ctrld access. Avoid to print output of check command 'sbc_smdr -h'
CHANGE: (#34071) Groupwise backup without ctrld access. Solution: Use sm_ssh instead. Fall back to obsolete sm_ctrlc
new: Create 'sm_uuid' SQL function before creating tables to allow using it for default value
change: Drop "add uuid" trigger in favor of declaring a default value for the UUID columns.
new: Create the "add uuid" INSERT trigger for more tables
bugfix: (#35565) module libvirtmod has no attribute virDomainListAllCheckpoints fix2
new: (#35713) reload authorized_keys on SIGHUP alongside config
bugfix: (#35713) reload authorized_keys on file change without sshd restart
bugfix: (#35254) Fixed line endings for makefile of ddboost common library
bugfix: (#35712) Windows Path backup failed with error: cUrl: [HTTP error before end of send, stop sending / Server response: Can't find chunk - not enough data to skip CRLF
BUGFIX: (#35712) Windows Path backup failed with error: cUrl: [HTTP error before end of send, stop sending / Server response: Can't find chunk - not enough data to skip CRLF
change: Changed that the migration id is displayed in the delete statement.
BUGFIX: (#35711) Execute Sesam disaster interface during migration of Sesam backup: Set correct backup subtype
change: Proxmox storage was not set for containers during backup. Caused by incomplete merge of changes from develop
BUGFIX: (#35711) Execute Sesam disaster interface during migration of Sesam backup: Set correct backup subtype
bugfix: (#35565) module libvirtmod has no attribute virDomainListAllCheckpoints fix
change: (#28875)  sbc_dir pxmx - added cores_socket and cores reporting, guest agent status/version detection and MAC address reporting
change: (#31928) execute sm_main start after sm_update to ensure that sm_sshd2 is started instead of sm_sshd after install.
change: (#35695) align pub keys naming with python sshd
change: (#35695) restore Python-compatible public key bootstrap for sm_ssh authentication
NEW: (#35699) Add datastore type and OS for 'Dell Data Domain' (DD Boost)
bugfix: (#35708) Print help and exit, if invalid option is set in 'sm_sem' call
NEW: (#35016) Sesam RPM build for SLES16
new: (#33016) Extended LIS file version 3. Restore of LISv3 and mixed backups
NEW: (#35254) Added condition to makefile.common to build DDBoost client library only if DDBoost is present in sesamdev
NEW: (#35254) Added DDBoost client library to common libraries
change: Set 'DB:sesam_events.state=0' during delete of savesets from 'DB:result_lbls'
bugfix: (#35299) vSphere backup on Linux using SAN transport uses one SAN multipath device path only. Remove duplicates
bugfix: (#35299) vSphere backup on Linux using SAN transport uses one SAN multipath device path only
new: (#35299) vSphere backup on Linux using SAN transport uses one SAN multipath device path only. Adapt static_assert_values.h uiSbcDBStructSize after change in db_api.h struct sSBC_DB
bugfix: (#35299) vSphere backup on Linux using SAN transport uses one SAN multipath device path only
NEW: (#35299) vSphere backup on Linux using SAN transport uses one SAN multipath device path only
bugfix: (#35582) Insert a delay for Hyper-V backups with quiescence 2 and 3 for the retries. Wait 10 seconds after first try and 20 seconds after second try.
BUGFIX: (#35627) After COPY backup destroy the snapshot which has been created at the beginning.
BUGFIX: (#35627) COPY backup method of Hyper-V does not create snapshot. Solution: Create snapshot for COPY during Hyper-V backup.
CHANGE: (#35646) Purge DB:migration_results by executing 'sm_purge_migration' only
BUGFIX: (#35670) Set severity to 'CRITICAL' for notification about corrupted Si3 datastore
BUGFIX: (#29902) Some files aren't correctly exposed by VFS - When adding entries to LIS cache, case must not be ignored on Linux but only on Windows
change: Skip detecting volumes for Proxmox containers
change: Check for existing container power state during Proxmox container restore with overwrite
change: (#35653) Proxmnox NBD backup: Changed max wait time to get freen NBD device from 12 hours to 600 seconds
change: (#35653) Exit semaphore loop in PVE backup on termination signal
change: (#35653) Limit number of retires for acquirement of semaphore get_free_devnbd - limited wait time to lock the semaphore to the maximum of 12 hours
change: (#35603): start VM in prelaunch mode using proxmox qm utility: unlock/lock vm before operation: log warning instead of error
change: (#35603): start VM in prelaunch mode using proxmox qm utility: unlock/lock vm before operation: adjust logging
change: (#35603): start VM in prelaunch mode using proxmox qm utility: unlock/lock vm before operation: move lock statement
change: (#35603): start VM in prelaunch mode using proxmox qm utility: unlock/lock vm before operation
change: (#35603): start VM in prelaunch mode using proxmox qm utility
bugfix: (#35325) Proxmox IR: Instant recover failed if TPM/EFI disks were on different type of storage in source VM than was the type of target storage for IR. When sbc_proxy is executed on the PVE node to restore TPM/EFI disks, restore method must not be set based on backup task sub type, but based on the target storage type. This is not a problem for normal VM disks because during IR they are attached from VFS, not restored
BUGFIX: (#29902) Some files aren't correctly exposed by VFS - When adding entries to LIS cache, case must not be ignored on Linux but only on Windows
change: code alignment
bugfix: (#35565) virDomain object has no attribute backupBegin fix
bugfix: (#29358) upload_res error fix
change: Use '$timestamp(I)' to get current timestamp. Furtheron current time is used
new: separate msvc-2026 dir for VS2026
bugfix: Remove unsorted lst file *.tmp. Do not remove unsorted SG.lis file
change: KVM - removed additional vm listing
change: (#35319) Deactivated default installation of CBFS drived with PnP support. Beacuse of it, silent installation/update does not work any more
bugfix: (#35605) When abort time expired then incremental backup may fail with 'ERROR: Saveset ... with failed state c'. Commit 4f0b89c1 accidentally moved header smk-3528 in loop
NEW: (#31928) Activate sm_sshd GO during installation of all develop and master kits.
change: update dockerfile for quay.io to use latest sesam version and alma 9 as base image
change: (#29358) Detect early snapshot failures and classify backend rejection causes
NEW: (#25976) Automatic fail over for devices, RDS and media. Set media pools to 'available', if failed drive becomes available during drive configuration again
change: (#29358) undo upload_volume_to_image changes
bugfix: disable s3 buffered restore
change: (#35319) Support Instant Recovery of Proxmox-VE on Windows Sesam server. Add sbc_proxy/lib to PYTHONPATH
change: (#28931) fix sbc_dir output format - guest_tools_running, mac_addresses
Revert "bugfix: Remove unsorted lst file *.tmp"
bugfix: Remove unsorted lst file *.tmp
bugfix: (#32972) extend logs for error handler
NEW: (#34286) Remove DB:media_results for action 'seed': Replace 'seed' by 'import' in log file name
change: speedup sbc_proxy build (switch from lazy init - every time it's referenced it re-expands)
new: introduce a new variable $version_string_old which can be use to check if the installed sesam version is develop, master or release.
bugfix: (#35406) Windows VFS mount/umount: Added separate temporary diskpart script for assigning driver letters to new mounted disks. Added more clear logs
bugfix: (#35650) Windows VFS mount/umount: Added separate temporary diskpart script for assigning driver letters to new mounted disks. Added more clear logs
NEW: (#31928) Insert a section to activate sm_sshd go and deactivate sm_sshd python during update.
change: (#28931) fix sbc_dir output format
bugfix: (#32972) fix issue when workers count decrease after skip or when future buffered was canceled
NEW: (#34286) Remove DB:media_results entries and log files for deleted tape media and data store action 'seed'
WIP: (#34690) Improve mtfana stream data size processing for compressed/encrypted. WIP: Implement correct frame processing in STPD mtfana_func
bugfix: (#32972) extend logs on skip; change default s3 restore parameters
new: (#35650) Execute manual umount of NFS share on Proxmox node after removal of Proxmox NFS storage
bugfix: (#35319) Proxmox IR: Removal of Linux NFS share was not correctly executed during detach
change: (#35359) Make the error message clearer so the user knows exactly why the restore failed.
NEW: (#35231) Fill DB:media_results.schedule during creation of entry
NEW: (#35319) sm_vfs installs support for PnP driver by default. If option -o nfs=1 is set, it enables its callback functions
NEW: (#35625) Update CBFS driver from 6.1.181 to 6.1.184
bugfix: (#32972) add sync chunk read when buffer is not ready; add priority mailbox with higher priority for restore core messages; increase timeouts; extend logs
NEW: (#35319) Proxmox IR on Windows: Remove of NFS share on Windows RDS client, fixed write of selected volumes and assigned letters to .bat script
bugfix: (#31958) Only prohibit VSS backup from being executed. But do not prohibit the sbc -D "VSS:\Microsoft Hyper-V VSS Writer/" do the dir command still works.
BUGFIX: (#35670) Set severity to 'CRITICAL' for notification about corrupted Si3 datastore
change: (#35597) fix argv construction for quoted PowerShell and Windows paths
change: Check for existing container power state during Proxmox container restore with overwrite
WIP: (#34690) Improve mtfana stream data size processing for compressed/encrypted. WIP: Implement correct frame processing in mtfana tool
change: (#29358) changed upload_volume_to_image - use wait_for_status
change: (#35202) rename the executable from integrity_check.exe to sm_integrity_check.exe.
NEW: (#35628): SLES16: rear recovery fails: add /usr/share/snapper/config-templates/ to recovery image if btrfs is detected
change: (#35653) Proxmnox NBD backup: Changed max wait time to get freen NBD device from 12 hours to 600 seconds
NEW: (#35628): SLES16: rear recovery fails: add /usr/share/snapper/config-templates/ to recovery image if btrfs is detected
new: separate msvc-2026 dir for VS2026
NEW: (#35319) Support Instant Recovery of Proxmox-VE on Windows Sesam server - execute powershell to create NFS share on the mount client, to support Windows RDS
bugfix: (#35359) Prohibit Hyper-V restore from newer Hyper-V version (e.g. 2022) to older Hyper-V version (e.g. 2016/2019), not supported. Problem: missed setting of wrong status and termination function in during versions checking. Solution: added needed code for correct restore end.
change: Set name of main log file in 'sm_robot' after 'DB:media_results.id' has been created
change: Do not write media actio main log by 'sm_cross_check' if no id is set
NEW: (#35649) Add 'cronie' as dependency for SiS package on RHEL
change: (#35624) add relabel to docu SLES16_SELinux.md
change: (#35653) Exit semaphore loop in PVE backup on termination signal
NEW: (#35624) add RHEL10 related SELinux: add README
NEW: (#35624) add RHEL10 related SELinux source files
change: (#35624) Check for SE-Linux on SLES16 only
change: (#36624) Check for SE-Linux on SLES16 only
change: (#36524) Check for SE-Linux on SLES16 only
bugfix: (#35605) When abort time expired then incremental backup may fail with 'ERROR: Saveset ... with failed state c'. Bugfix: Do not overwrite state 'c' or '3'
change: (#35653) Removed some commented code
new: separate msvc-2026 dir for VS2026
new: VS2026 build fixes
change: (#35653) Limit number of retires for acquirement of semaphore get_free_devnbd - limited wait time to lock the semaphore to the maximum of 12 hours
change: (#35624) add SLES16 related SELinux source files: update documentation
change: (#35202) adjust the include list of the integrity_check go files.
change: (#35624) add SLES16 related SELinux source files: remove binary module
NEW: (#35624) add SLES16 related SELinux source files
change: replace code 3500 with 3987 in sbc_proxy snapshot creation messages
new: (#35319) Fixed call to sm_mount -f <file> to work for PATH backup type
CHANGE: (#35646) Purge DB:migration_results by executing 'sm_purge_migration' only
new: Set DB:results.termination_time for backup sub-tasks to termination time of main task
NEW: (#36524) Do not allow Sesam server installation with  activated SE-Linux
change: (#35597) move sshd2 code to cmd+server layout
NEW: (#35649) Add 'cron' as dependency for SiS package
new: separate msvc-2026 die for VS2026
bugfix: (#35597) preserve Windows backslashes when parsing non-PowerShell commands
change: fix projects build dependencies
change: Delete cm_java.py
change: avoid empty parentheses when datastore type is empty v2
change: fix projects build dependencies
change: avoid empty parentheses when datastore type is empty
new: Increase SEP sesam version. Old version is 5.2.0.18, new version is 5.2.0.19.
new: Increase SEP sesam version. Old version is 5.2.0.17, new version is 5.2.0.18.
bugfix: (#35319) Prxmmox SFR failed because path was used as a name for local variable and it overrided import of os.path
NEW: (#35423) Remove old backup directory used for Postgresql version update of Sesam DB
BUGFIX: (#35605) When abort time expired then incremental backup may fail with 'ERROR: Saveset ... with failed state c'. Problem: During adjusting the EOL of large FDI chains the signal was not catched. Bugfix: Send signal to sm_backup after iKILL_STPD_AND_SBC(). Add signal handling during EOL adjusting. Use the fast SQL statement in that case.
NEW: (#35319) Proxmox IR: Read VM config from mounted config saveset by executing sbc_proxy to read config file content from VFS mount. Added new option sm_mount -f <file name>
BUGFIX: (#35627) After COPY backup destroy the snapshot which has been created at the beginning.
bugfix: (#35573) Undefiend variable was causing warning message and skip from removing teporary VFS filesduring VFS umount
change: VZO dir vm - set datastore type to empty value
bugfix: (#34978) Proxmox: Added back host info to dir print, removed by mistake
CHANGE: (#31958) Remove support for non RCT Hyper-V backup. Prohibit the execution of Hyper-V VSS backups since they have limited functionality.
bugfix: (#32972) add force start of buffering; fix issue when buffering finished flag wasn't changed after skip; extend logs
NEW: (#33103) Set DB:restore_results.eol to DB:restore_events.eol if set in DB:restore_events. Overwrites value from DB:restore_tasks.results_eol
NEW: (#34978) Proxmox: If a node cannot be reached by sbc_proxy during sbc_dir for a VM browse, print more clear message. Additional fix: Skip getting network info for an offline node, it resulted in an unhadled exception
NEW: (#35628): SLES16: rear recovery fails: add detection for BTRFS type volumes in sbc_rear module: remove version check
new: (#35202) Add integrity_check executable to the build rules of solution file. It is not included in the Release build process yet.
BUGFIX: (#35627) COPY backup method of Hyper-V does not create snapshot. Solution: Create snapshot for COPY during Hyper-V backup.
NEW: (#35628): SLES16: rear recovery fails: add detection for BTRFS type volumes in sbc_rear module: remove version check
change: Added state value for DirHost KVM
change: Added DirHost for KVM
change: Fill datastores property with used storage pools (KVM and VZO)
bugfix: (#35212) Hyper-V restore must fail without option stop if VM is running. Problem: didn't check the VMs in cluster during restore without cluster flag. Solution: added the logic without cluster flag and the check of VM path.
change: (#35597) cleaning up the test cases
new: (#35597) Fix PowerShell command parsing with full executable paths
new: (#35597) add Hyper-V PowerShell command tests for VM management
bugfix: (#35573) If sbc_proxy umount cannot delete mount directories because they are already removed or are not mount points due to manual umount or VFS error, catch exceptions, log messages and continue with umount operations instead of returning with error
bugfix: (#32972) reset buffered object reader eol when setting new position after skip
bugfix: (#35573) If VFS mount directory fails to be deleted during umount, try to execute OS umount before another try to delete it. This solves tha case when guestmount function does not do the job due to other unhandled errors
NEW: (#33103) Set DB:restore_results.eol to DB:restore_events.eol if set in DB:restore_events. Overwrites value from DB:restore_tasks.results_eol
bugfix: (#32972) clean up actual buffered bytes count for canceled link buffering that finished  after skip
NEW: (#35397) Add 'parted' to RPM dependencies for SLES 15/16 for BSR backup (REAR) on multipath setups
bugfix: (#32972) fix buffering worker start condition; fix buffered chunks calculation
NEW: (#33103) Set DB:restore_results.eol for external restores
change: (#28931) Extend sbc_dir to include additional virtual machine metadata (ips, cpu, memory)
NEW: (#33103) Set DB:restore_results.eol from DB:restore_tasks.results_eol during creation of DB:restore_results entry
NEW: (#33103) DB:restore_tasks.results_eol to set EOL for DB:restore_results
NEW: (#35628): SLES16: rear recovery fails: add detection for BTRFS type volumes in sbc_rear module
change: (#28931) Extend sbc_dir to include additional virtual machine metadata (CPU, memory, guest-related information, ips)
bugfix: (#35319) Fixed errors in Proxmox backup caused by changes for IR on Windows server
change: Use specific error message number instead of '1500'
BUGFIX: (#35623) Overwrite warning message with error message during backup
BUGFIX: (#35475) sm_sshd: fix call to os.makedirs
change: (#35475) sm_sshd: catch exception and log warning if public key copy to stpd work directory fails
change: Fixed unhandled exception in Proxmox restore if target storage is not set -  exit with error message
change: Proxmox: Added function format error to extract friendly ZFS error message
new: add active replications count to server status
bugfix: (#35459) Handle OpenStack login errors and validate connection without exceptions
bugfix: (#35597) Return full path for Windows exec.Command compatibility
bugfix: (#35319) Fixed errors in Proxmox backup caused by changes for IR on Windows server
bugfix: (#35597) fix inline PowerShell commands with backslashes regression
change: Skip detecting volumes for Proxmox containers
NEW: (#35016) add 'postgresql18-server' to dependencies for Sesam server on SLES16
new: (#35435) Start Si3 replication on source side
new: (#35319) Instant Recovery of Proxmox-VE on Windows Sesam server - implemented changes in sbc_proxy and sm_mount
change: (#35608) Change parsing of o.user to user@project@user_domain@project_domain
NEW: (#35207) Add SiS EOL (immutable lock time) as 'key=value' during configuration
change: (#34970) change delimiter in user-defined-keys.si3
CHANGE: (#35379) MS-SQL restore may need more time than necessary due to checks of other instances. Solution: Checking all existing instances and connecting to them is only performed if the '-a DEEP_BROWSE' option is set.
new: (#35435) Start Si3 replication on source side
change: (#35607) Added DirHost to VZO sbc_dir
bugfix: (#35597) fix sm_ssh commands failing with executable paths containing spaces
NEW: (#34760) Run multiple SiS services on one appliance server
change: (#35607) Added hostSystem to VZO sbc_dir
change: (#35435) clean up code
new: update docs - add info how to exclude CHUNK logs
new: (#35435) add logic, route and command to start replication on source server with specifying target server
NEW: (#35016) 'proc-ps' to  dependencies for Sesam client on SLES16
NEW: (#35016) RPM dependencies for Sesam client on SLES16
change: (#35603): start VM in prelaunch mode using proxmox qm utility: unlock/lock vm before operation: log warning instead of error
bugfix: (#35606) Restoring data from a subsequent tape fails if LTO encryption is used. Solution: Set encryption key after EOM tape change.
change: (#35603): start VM in prelaunch mode using proxmox qm utility: unlock/lock vm before operation: adjust logging
change: (#35603): start VM in prelaunch mode using proxmox qm utility: unlock/lock vm before operation: move lock statement
change: enable sles16 build
change: (#35603): start VM in prelaunch mode using proxmox qm utility: unlock/lock vm before operation
bugfix: (#35565) backupBegin API not available fix
bugfix: (#35212) Hyper-V restore must fail without option stop if VM is running. Problem: problems with PS script execution on some cluster nodes. Solution: changed the logic of script to use cluster groups instead of invoke the check on each node.
change: (#35603): start VM in prelaunch mode using proxmox qm utility
new: Increase SEP sesam version. Old version is 5.2.0.17, new version is 5.2.0.18.
new: extend sm_dedup_cli restore with skipping case
bugfix: (#32972) fix issue when last processed object link wasn't stored after skip
NEW: (#34706) Allow multiple Si3 datastores on SiS server. Si3 service counter could be set by '-N <n>' switch
NEW: (#33501) Copy SiS ini file from data store location if not found in Sesam INI directory. This allows migration from RDS to SIS package by remove and install of package
NEW: (#35546) copy new GPG key, add import statement)
BUGFIX: (#35576) sbc_gen output is read byte by byte. Problem: performance losses, especially with large amounts of data. Solution: Read using cache (4K)
Revert "BUGFIX: (#35576) sbc_gen output is read byte by byte. Problem: performance losses, especially with large amounts of data. Solution: Read using cache (4K)"
bugfix: (#35325) Proxmox IR: Instant recover failed if TPM/EFI disks were on different type of storage in source VM than was the type of target storage for IR. When sbc_proxy is executed on the PVE node to restore TPM/EFI disks, restore method must not be set based on backup task sub type, but based on the target storage type. This is not a problem for normal VM disks because during IR they are attached from VFS, not restored
BUGFIX: (#35576) sbc_gen output is read byte by byte. Problem: performance losses, especially with large amounts of data. Solution: Read using cache (4K)
bugfix: Copying of pdb.zip files was wrong. Copy only if PDB file exists
bugfix: Copying of pdb.zip files was wrong. PDB file is located in build root folder
new: (#35597) port Python shlex to Go for proper quote handling
bugfix: (#32972) fix issue when single and repetitive chunk buffered data skipped
bugfix: (#32972) fix restore failing on skip; increase timeouts for s3 restore messages; extend logs; fix issue when wrong buffered chunk was retrieved during restore
change: Refactoring code in 'sm_robot.c' to avoid using of global C variables
BUGFIX: (#33593) Change state from queued to active in DB:media_results, if 'sm_robot' starts
change: Usefile function -t touch now with default timestamp -s 'now'
bugfix: Copying of pdb.zip files was wrong, used ....exe instead of ...-pdb.zip
bugfix: re-add sbc-2045, it is still used by sbc win. Removed in 184c7dc
change: (#29358) storage df added dummy values 1
bugfix: (#35389) Tape encryption is not enabled during initialization. Call CLEAR_ENCYRPTION() before ENABLE_ENCRYPTION() if encypted trailer could not be read and tape media was read from header
bugfix: (#35212) Hyper-V restore must fail without option stop if VM is running. Problem: check the running VMs only on main cluster node on restore start. Solution: added the script to check the running related VMs on each nodes in cluster by running the script on every node.
new: (#35568) Check label on tape, if archive adjustment has been started with 'overwrite' mode
bugfix: (#35293) Plain restore from tape fails if selected items reside on first segment of next tape. Remove debug trace lines
change: (#29358) Rename 'Virtuozzo' to 'Virtuozzo-HS'
NEW: (#35001) Media/media pool event type is not shown in next events table
change: (#29358) Rename 'Virtuozzo' to 'Virtuozzo-HS'
bugfix: (#35582) Insert a delay for Hyper-V backups with quiescence 2 and 3 for the retries. Wait 10 seconds after first try and 20 seconds after second try.
new: (#33016) Extended LIS file version 3. Support sm_mount
bugfix: (#35581) Proxmox IR fails for a VM with an empty disk on LVM-thin storage - BDA chunk "<vdisk size> - 0" was missing for full/copy backups of LVM-thin disks
change: (#29358) sbc_vzo: added datastore select for restore vm
change: (#35293) Plain restore from tape fails if selected items reside on first segment of next tape. Problem: CI test failed due to git replaced LF with CRLF. Solution: Write CRLF before Tape ID line on Windows
new: (#35537) Common quiescence mode for all backup modules - Changed warning message texts for -a qui=2and backup continue
bugfix: (#35563) Fixed parsing of Proxmox restore VM target setting
bugfix: Option -d (diff) to compare binary files must use correct size of amount of read bytes
CHANGE: (#35568) Add new option '-N <0|1|2>' to 'sm_robot' to start archive adjustment with: over barcode only, check unknown media and check all media in drive
new: (#35319) Proxmox IR: restore of VM config during IR is done by mounting the main Proxmox saveset as PATH mount and reading VM configfrom VFS directory. Because it is not possible to restore CPIO on Windows mount client
change: (#35372) sbc_kvm: do not load nbd module, fallback to backup mode copy if not loaded; Revise log message
NEW: (#34823) MSSQL: Support 'ODBC Driver 18 for SQL Server'. Start check with newest driver 'ODBC Driver 18 for SQL Server'
bugfix: (#35565) Check backupBegin availability before use
bugfix: (#34311) Coredump in STPD during restore if INCR saveset used LISv2 but FULL used LISv1. Remove test code
bugfix: (#35567) Log VM Id and exception text when Proxmox browse fails to get agent info for a VM
BUGFIX: (#34311) Coredump in STPD during restore if INCR saveset used LISv2 but FULL used LISv1
new: Implemented qui=0,1,2 parameter support for KVM backups, handles paused VMs and missing QEMU Guest Agent correctly.
NEW: (#25976) Automatic fail over for devices, RDS and media. Set media pools to 'available', if failed drive becomes available during drive configuration again
new: Common parsing of argument -a qui in sbc_proxy. If Nutanix application-consistent snaphot creation fails and -a qui=2, log warning and retry by creating crash-consistent snapshot
new: Increase SEP sesam version. Old version is 5.2.0.16, new version is 5.2.0.17.
CHANGE: (#35554) Change default quiescence mode to '2' for Hyper-V. In addition change the number of tries for application consistent backup from 2 to 3.
new: (#35466) Add Si3 online timestamp to status output
bugfix: (#35359) Prohibit Hyper-V restore from newer Hyper-V version (e.g. 2022) to older Hyper-V version (e.g. 2016/2019), not supported. Problem: don't have the check for versions compatibility during restore. Solution: implemented the logic to check the VM version which is backuped in config file and current host supported version
new: (#35466) add server online timestamp to di status; add pages, pages-in-trash, objects and objects-in-trash counter trackers and add them to di status; fix recalc trackers and reconnection logic for Azure; fix dir size trackers recalculation logic on cloud's reconnection
bugfix: (#35450) Sesam GUI package installation fails on RHEL 9/10: add missing ifdef
change: insert missing L macro and change some log messages.
bugfix: (#35542) CLI command “sm_cmd import calendar” fails over curl
new: Common parsing of argument -a qui in sbc_proxy. Changed default value for -a qui to 2
new: Add new specific message numbers for snapshot operations
BUGFIX: (#35475) sm_setup fails to copy ssh keys to stpd/read. Revert change because it doesn't work
bugfix: fix restore skip for s3; fix si3 server crush in case of s3 restore fail; rework skip function for s3 restore
bugfix: fix the issue with local page uploading to s3 while it used by page body reader; block StartUploading command execution while previous still running
bugfix: fix an error log occurrence during non existing /tmp dir si3validation files deletion attempt
new: Common parsing of argument -a qui in sbc_proxy. Fixed NBD backup, there was an extra exit with error
new: Common parsing of argument -a qui in sbc_proxy. Implemented logic for module NTX, backup level COPY. Standard VM snapshot if -a qui=0. If quisce VM snapshot fails and -a qui=1, exit with error, else log warning and continue with standard snapshot.
new: Common parsing of argument -a qui in sbc_proxy. Implemented logic for module NTX. Crash consistent backup if -a qui=0. If there are no guest tools installed, backup with warning if -a qui=2, if -a qui=1 then exit with error. If guest tools are installed and -a qui=1|2, application consistent backup
new: Common parsing of argument -a qui in sbc_proxy. Implemented logic for module PXMX. In case of failure to freeze FS, warning if -a qui=2, backup contiue. Exit backup with error if -a qui=1
new: (#33016) Extended LIS file version 3. Restore of LISv3 and mixed backups
bugfix: (#35551) Hyper-V backup with qui=2 does not fall back to crash consistent backup. Solution: The wrong variable was checked to find out which kind of quiescence mode should be used. Change the correct variable.
change: inserted several trace messages to easen error analysis.
NEW: (#35544) add new GPG public key to skel directory
new: Common parsing of argument -a qui in sbc_proxy. Changed default value from 0 to 1. Cast value toto int is sbc_opts
new: Common parsing of argument -a qui=0|1|2 in sbc_proxy. Fix incorrect parsing
new: Common parsing of argument -a qui=0|1|2 in sbc_proxy
NEW: (#35535) SFR with Proxmox VE VM saved with 'ZFS' sub type not possible - If backup sub type is detected automatically, set it to ZFSBDA instead of ZFS
BUGFIX: (#35533) Backup to Si3 from 5.2.0.12 using STPD 5.2.0.15 is successful, but data file isn't created
bugfix: (#35533) If STPD version is incompatible to Si3 version backups are marked as successful though they were not. Abort backup if any Si3 API call fail
change: (#29358) Added Virtuozzo overwrite vm and stop_vm_at_recover check
change: Change stop vm after restore message id
change: Fix some typos in the README.MD and add a note about documentation issues
bugfix: (#35481) Cleanup temporary Glance images after backup
BUGFIX: (#35293) Plain restore from tape fails if selected items reside on first segment of next tape. Problem: First file on next media was written before and after Tape ID line. Solution: Print after Tape ID line. Additional: New function 'sm_sesam reformat_lis ...'
CHANGE: (#15589) Set backup sub-type '_SES' for all Sesam backup tasks during update from version < '5.2.0.14' and with backup type 'Path'
bugfix: (#35299) vSphere backup on Linux using SAN transport uses one SAN multipath device path only. Remove duplicates
new: Add severity labels description to README.MD
NEW: (#32234) Add backup warning message to 'DB:results.msg'
bugfix: (#35298) Changed the log levels in CreateSnapshot() function to show the main error in the end
change: (#35487) Improved logs
change: Catch signals when waiting for semaphores, e.g. for 'ctrl_drive#'
change: Use x_DumpEx() to trace SCSI in buffer (Win32 ASPI scsi cmd)
NEW: (#35487) Mount of NSS volumes from virtual disks on Xen server. Call NSS umount script during VFS umount
bugfix: (#35444) Proxmox backup: Fixed function to get free NBD device, not to return devices with staled pids as free. Use value of nbds_max to try to get max number of free device, default 16. Execute cleanup after error to get NBD client connection
new: Increase SEP sesam version. Old version is 5.2.0.15, new version is 5.2.0.16.
BUGFIX: (#35479) Hyper-V restore could fail with the error message 'Error: NHyperRCT::NRestore::post_process_item' if the metadata file size was aligned to FLB size. Solution: If the stream is aligned to the FLB size, no additional SPAD stream is created. Avoid signaling file processing twice: sbc-3309 Item successfully processed [...]
bugfix: Since commit edd0c49 Using DEBUG_TRACEW( TRACE_TRACE, ... ) prints only first the character. Problem: A message with code 4001 must be filtered with trace level < SBC_TRACE_LEVEL - like 4000
bugfix: (#35444) Improved log on error during Proxmox NBD backup
BUGFIX: (#30254) Installer shows hyper-v cluster notice on an exchange DAG node. Solution: use different powershell commands to check for the installed Hyper-V feature. Use 'Get-WindowsFeature Hyper-V' and 'Get-WindowsFeature Failover-Clustering' instead of 'Get-WindowsOptionalFeature -Online -FeatureName Microsoft-Hyper-V' and 'Get-WindowsOptionalFeature -Online -FeatureName FailoverCluster-Mgmt'
BUGFIX: (#35475) sm_setup fails to copy ssh keys to stpd/work/read directory due to race condition. Wait for creation of ssh keys duing first start of 'sm_sshd'
bugfix: (#35382) Incremental MS-SQL backup hangs during fetching backup history. Trace messages improved
bugfix: (#35334) Proxmox dir functions orints only networks of type bridge. Other types cannot be set to VMs
bugfix: (#35382) Incremental MS-SQL backup hangs during fetching backup history. Call 'CHECKPOINT' after fetching the backup history
new: add total number of printed items for 'di list' and 'di pageslist'
change: Remove obsolete build rule for Sesam NetApp module, which is included in 'sbc_proxy' now
BUGFIX: Calculate correct estimated end time for replications
bugfix: Typo in log message 'loop count %d missing 'Slot' in output' (missed argument int i)
bugfix: (#35463) Fixed remove of VFS mount entries from JSON file during umount.VMWare backup on Windows had this issue because obsolte argument for sm_rexec_buf_log was set. Added try catch to calling remove_mnt_dir() and remove_sms_mnt()
bugfix: (#35299) vSphere backup on Linux using SAN transport uses one SAN multipath device path only
new: (#35462) Add hint about length mismatch between 'DB:drive_groups.grp_name and 'C DRV_PROP struct'
new: (#35299) vSphere backup on Linux using SAN transport uses one SAN multipath device path only. Adapt static_assert_values.h uiSbcDBStructSize after change in db_api.h struct sSBC_DB
bugfix: (#35299) vSphere backup on Linux using SAN transport uses one SAN multipath device path only
NEW: (#35299) vSphere backup on Linux using SAN transport uses one SAN multipath device path only
CHANGE: (#15589) Set backup sub-type '_SES' for all Sesam backup tasks during update from version < '5.2.0.14'
fix: (#29358) post-freeze cleanup for openstack fix
fix: (#29358) exit if openstack_connect fails
CHANGE: sm_cmd must not change code page for console (since Artemis). Solution: Using wmain() ensures the correct code page, so that explicit code page changes can be avoided.
change: (#29358) Add post-freeze cleanup of .py and __pycache__ for openstack
bugfix: (#20823) reinsert missing column cluster_name to table servers.
Merge remote-tracking branch 'origin/openstacksdk' into develop
bugfix: (#24705) Adjust Hyper-V browsing to enable disk exclude during restore. Problem: The BAT file of excluded disk was not excluded during restore. Solution: Added logic to exclude related to excluded disks BAT files
change: (#29358) add openstack modules to cx_Freeze include_files
bugfix: (#35450) Sesam GUI package installation fails on RHEL 9/10 with: 'nothing provides redhat-lsb-core needed'
bugfix: (#34675) Use correct hostname for Si3 connection in case of mount of saveset from a cloud storage
change: (#29358) Support for Virtuozzo Hybrid Infrastructure. New backup keys 'VIRTUOZZO_SERVER', 'VIRTUOZZO_SERVER_SEATS', 'VIRTUOZZO_Sockets' must not yet be set for edition 'SEP Unit' (sesam_make_lic)
change: (src#20823) Real UI proxy master server
bugfix: (#35453) Port number isn't forwarded by sbc to 'sbc_com' in case of https:// interface. Solution: Always add port if its available but missing in hostname
bugfix: (#35452) Insert missing wide char makro (L) in trace method DEBUG_TRACEW. Use log method DEBUG_TRACEA is no wide char is logged.
bugfix: Using DEBUG_TRACEW( TRACE_TRACE, "..." ) prints only first the character. Solution: Do not use the special SBC message tag 4000, which assumes that the first argument is the format identifier, but use the new tag 4001 instead.
change: (#29358) removed print message
fix: (#29358) OPENSTACK  sys.argv.pop params order fix
new: (#33016) Extended LIS file version 3. Selective file restore
BUGFIX: (#35448) Do not reset value for key='sm_conf_max_sds_rds' to '1' in DB:defaults during update, if value > 1 has already been set
BUGFIX: (#35442) Recover of saveset from datastore fails because 'info' file isn't read correctly. Do not read STDERR during retreiving saveset '.info' file
change: add support backup to NULL:64:64 for sbc_vmware CBT module
change: improve sbc_bda logging
change: (#35424) sbc reports 'Sending data block via FTP' when transferring RAW data, regardless of the protocol used. Remove this log
Fix compiler warning in 'sm_update.c'
bugfix: Do not compare saveset checksums for failed migration and leave original error message untouched
change: (#29358) add multidisk restore support for Virtuozzo
change: calculate execution time with ms for more correct xbsa throughput
bugfix: (#24705) Adjust Hyper-V browsing to enable disk exclude during restore. Problem: Included disks were not added to sbc command. Solution: Added decoded from URL disk path pattern to sbc command. Implemented exclude logic for RCT restore
change: fixed some spelling changes in method names, variable names and log messages.
bugfix: (#35298) In case guest tools are not installed on a Linux VM, the error message will now explain what the user has to check to get the application consistent backup working. The correct consitentLevel variable is used to check if application consistent or crash consistent backup is being executed.
Add new make target 'copy_rpm_rts'
BUGFIX: (#35394) Hyper-V SFR does not work when VDHX file name contains spaces - Replace spaces with unedrscores before executing diskpart commands to attach virtual disks
change: add iphlpapi.lib as its requested by curl 8.17.0
change: move the creation of the filelists from the makefile.win32 to the modifyAndBuildISM.bat since the sm_execute_update.cmd file is being modified in the modifyAndBuildISM.bat and hence the filelist was created at an too early stage.
NEW: (#26288) ACSLS as loader control software. New: Archive adjustment for ACSLS loader by 'barcode only'
bugfix: (#35389) Tape encryption is not enabled during initialization. Do not perform ENABLE_ENCRYPTION() for DISK_HARD/DISK_CHNG
change: (#35429) reverted back the pip monitor and increased cmd completion grace to 20s to avoid race
new: Remove support for SLES12 from RPM spec files
Revert "change: downgrade rust modules as we back from curl 8.11.1 to 7.77.0"
change: downgrade rust modules as we back from curl 8.11.1 to 7.77.0
bugfix: (#35389) Tape encryption is not enabled during initialization. Error handling improved
change: (#34795): Skip freezing the filesystems for powered off RHV or Oracle-VM virtual machines: improve message
change: move purge progress logs from debug to trace level
change: trace curl with data sizes
bugfix: avoid negative dedup ratio
change: Avoid compiler warning about i_TRY_CHANGE_LABEL( int <> enum _eTapeState)
bugfix: (#35389) Tape encryption is not enabled during initialization. Typo: if ( pMP->iEncrypted == 1 )
Fix syntax error in 'sm_get_volume_lib.c'
bugfix: (#35389) Tape encryption is not enabled during initialization. Solution: Use media.crypt_flag to enable tape encryption, use media.encryption_capable only to store the capability
new: (#35429) disable process monitor in SSH command execution for Linux and Windows
change: (#29358) added check - Overwriting VM in Virtuozzo is currently not supported
bugfix: (#35301) Fixed Nutanix fallback to full backup if base snapshot is missing for incremental
new: (#35365) implement SSH connection rejection
change: (#29358) basic backup and restore VZO vm
BUGFIX: (#35416) sm_setup set_java_path for Java21 may fail on Windows with more than 64 CPUs. Skip warning line
change: Write main log file of archive adjustment into Sesam date based sub-directory
change: Add 'DB:media_results.id' to tape media 'readcheck' main log file
bugfix: (#35361) BDA backups create very big logfiles (Proxmox, KVM). Solution: print extra info on log level >= 2
new: Add 'VIRTUOZZO_VHS' to task types using sub tasks
new: (#35365) return an explanatory message to rejected connections
bugfix: (#35361) BDA backups create very big logfiles (Proxmox, KVM). Solution: print extra info on log level >= 2
NEW: (#35423) Remove old backup directory used for Postgresql version update of Sesam DB
change: (#35365) update and add missing dependencies and remove unused packages in go.mod and go.sum
bugfix: (#35365) close excess connections to prevent log flooding
new: (#28315) implement replication for synthetic objects; provide version compatability for replication process between new client/old source and old client/new source
BUGFIX: (#34795): Skip freezing the filesystems for powered off RHV or Oracle-VM virtual machines: fix status call
new: (#29358) Support restore of Virtuozzo VM
new: (#33016) Extended LIS file version 3. Print chksum as HEX
Fix merge conflict
new: (#33016) Extended LIS file version 3. Add header line
bugfix: (#35417) Execute 'dir' operation for MySQL on Windows using 'sbc_proxy'. Typo
change: (#35407) Sort all license queries by default to guarantee a consistent ordering of licensed items. Additonal: Sort Level L2
bugfix: (#35301) Nutanix Copy backup reads base snapshots file to skip reference snapshots from deletion during cleanup
bugfix: (#34937) fix Repair PageTrack issue when repaired page was still marked as deleted; add docs on PageTrack and PageLock
CHANGE: (#35417) Execute 'dir' operation for MySQL on Windows using 'sbc_proxy'
new: (#29358) Set proxy flag to 'Linux' for backup type 'Virtuozzo'
new: (#29358) Support backup of Virtuozzo VM
bugfix: (#35215) Proxmox restore with overwrite was executed even if VM was running and stop flag was not set
new: (#33016) Extended LIS file version 3
change: (#35407) Sort all license queries by default to guarantee a consistent ordering of licensed items. Additonal: 'OpenLDAP' before 'OpenText GroupWise'
CHANGE: (#35407) Sort all license queries by default to guarantee a consistent ordering of licensed items. Additonal: Sort Level 3/2 backup types too
bugfix: (#35385) make Windows process monitor consult cmdDoneChan and handle unknown PID/OpenProcess errors like Linux
CHANGE: (#35413) Rename 'Novell' to 'OpenText'. The names 'Novell GroupWise' and 'Novell SMS' have been changed to 'OpenText GroupWise' and 'OpenText SMS'
bugfix: (#33687) Backup to Si3-NG using 'raw' option
BUGFIX: (#34795): Skip freezing the filesystems for powered off RHV or Oracle-VM virtual machines
CHANGE: (#35417) Execute 'dir' operation for MySQL on Windows using 'sbc_proxy'
change: (#29358) restore VZO vm changes
bugfix: (#35215) Proxmox restore with overwrite was executed even if VM was running and stop flag was not set
new: (#33016) Extended LIS file version 3
change: (#35407) Sort all license queries by default to guarantee a consistent ordering of licensed items. Additonal: 'OpenLDAP' before 'OpenText GroupWise'
CHANGE: (#35407) Sort all license queries by default to guarantee a consistent ordering of licensed items. Additonal: Sort Level 3/2 backup types too
bugfix: (#35385) make Windows process monitor consult cmdDoneChan and handle unknown PID/OpenProcess errors like Linux
CHANGE: (#35413) Rename 'Novell' to 'OpenText'. The names 'Novell GroupWise' and 'Novell SMS' have been changed to 'OpenText GroupWise' and 'OpenText SMS'
bugfix: (#33687) Backup to Si3-NG using 'raw' option
fix merge conflict
BUGFIX: (#34795): Skip freezing the filesystems for powered off RHV or Oracle-VM virtual machines
BUGFIX: (#35301)  Nutanix FULL-DIFF-INC backup leftover more than one 'Recovery Point' - All snapshots except the last snapshot needed for reference are deleted on each backup
change: sbc_nutanix: more meaningful error message if iscsi discovery does not return any targets
bugfix: (#35385) fix race condition by adding cmd-aware ProcessMonitor
change: (#35301) Delete old base snapshot at the emd of Nutanix backup
change: (#29358) Could not increase the EOL fix
change: (#35301) Do not delete full snapshot during cleanup phase of Nutanix backup, leave it to the full base cleanup, otherwise there is warning on the second attempt to delete the same snapshot again
change: (#29358) backup disks using stdin mode
undo sbc http backup changes
BUGFIX: (#35301)  Nutanix FULL-DIFF-INC backup leftover more than one 'Recovery Point' - All snapshots except the last snapshot needed for reference are deleted on each backup
change: sbc_nutanix: more meaningful error message if iscsi discovery does not return any targets
Merge remote-tracking branch 'origin/develop' into openstacksdk
change: (#29358) download disk data to datastore v1
NEW: (#35399) Storage backend drive type for CBSR store
change: improve backend detect in case of -L/-p/-h is set
new: (#29358) Support for Virtuozzo Hybrid Infrastructure. New backup_type 'Virtuozzo' -> license items 'VIRTUOZZO_SERVER', 'VIRTUOZZO_SERVER_SEATS'. ('VIRTUOZZO_Sockets' as draft)
Revert "BUGFIX: (#35301) Nutanix FULL-DIFF-INC backup leftover more than one 'Recovery Point - Snapshot cleanup removes all old snapshots except the last one needed for base"
change: (#29358) auth data fixed
BUGFIX: (#35301) Nutanix FULL-DIFF-INC backup leftover more than one 'Recovery Point - Snapshot cleanup removes all old snapshots except the last one needed for base
change: (#29358) removed data
CHANGE: (#35396) Remove support for Nutanix-AHV level DIFF backup
change: (#29358) renaamed VZ to VZO
New: Increase SEP sesam version for develop branch. Old version is 5.2.0.14, new version is 5.2.0.15.
new: (#33016) Extended LIS file version 3. LISv3 for CPIO with -o no_ea,noacl could be started with -i lis=3
new: (#29358) first backup - metadata
bugfix: Core dump in sm_arch because of wrong format specifier
new: (#35385) add tests for command completion and exit codes without stdin close
bugfix: (#35385) fix race condition causing false "process killed" reports
change: (#13971) Drive hardware encryption. Differ between media.crypt_flag_media -> encryption is requested and media.contains_encrypted_blocks -> volume contains encrypted logical blocks (VCELB)
WIP: (#34671) Optimize EOL calculculation for long FDI chain. With the GLBV 'gv_conf_fdi_chain_rows_limit' the optimized SQL statements are used if the FDI chain consists of at least this number of backups.
WIP: (#34671) Optimize EOL calculculation for long FDI chain. WIP: GLBV 'gv_conf_fdi_chain_rows_limit' allows to set the rows limit for optimized EOL adjustment with recursive statement. Avoid to trace result of 'select ... from media where ... FDIchain ...'
WIP: (#34671) Optimize EOL calculculation for long FDI chain. WIP: If more than 500 ...
WIP: (#34671) Optimize EOL calculculation for long FDI chain. WIP: Set to backup_id returned by FDI_CHAIN() coz may be set to session id in case of sub task
WIP: (#34671) Optimize EOL calculculation for long FDI chain. WIP: If more than 3000 backups in FDI_chain then update EOL for result_lbls, results and media with SQL statements
NEW: (#29358) Add backup task type for 'Virtuozzo Hybrid Infrastructure'
NEW: (#29258) Add backup task type for 'Virtuozzo Hybrid Infrastructure'
NEW: (#35376) Sesam server for SLES: Allow Java 17 and 21 only
new: (#35319) NFS share during Proxmox IR is created by sm_mount. Added new sm_mount argument -x <path>, when added to sm_mount prepare, crates NFS share after VFS starts. NFS export is created by a new common function create_nfs_share() that is reusable for other modules
change: use http to checkout sep-sesam-extern repository
BUGFIX: (#35393) sbc_rhev: finalize transfer service before attempting to stop backup service if data transfer fails
change: add support for gcc15
change: sbc_pxmx: use --nocache option for qemu-nbd during restore
change: sbc_kvm/sbc_pxmx: log messages related to fileystem freeze in sync with other modules
BUGFIX: (#35356): sbc_kvm: attempt to freeze filesystems using guest agent before backup begin
change: sbc_pxmx: change logic
change: sbc_pxmx: flush nbd device buffers before disconnecting
change: sbc_pxmx: use --cache=writethrough for NBD devices mapped for restore
fix: (sshd2) improve TestSepRunSSHKilled and add new session run tests
bugfix: (#35386) Fixed Proxmox vzdump restore. Volume attributes must not be checked for vzdump because volume object is None
fix: (sshd2) add the missing file
new: (sshd2) add test layer for session requests
new: introduce testable sshd2 request session
new: introduce testable sshd2 request instance
fix: Fix process monitor not detecting SIGKILL by checking zombie state from /proc
change: sbc_pxmx: return false if NBD setup fails instead of continuing with restore
bugfix: (#35337) Use the same function as in KVM module also in Proxmox module to get free NBD device. Moved the function to common3
new: (#33016) Extended LIS file version 3. LISv3 for CPIO with -o no_ea,noacl could be started with -i lis=3
change: sbc_pxmx: return false if NBD setup fails instead of continuing with restore
bugfix: (#35386) Proxmox-VE restore fails for ZFS snapshot with EFI disk - do not restore TPM and EFI with ZFS script, just restore disk data vith SBC
bugfix: (#34857) Proxmox module  was using subtype LVTHI for thin LVM, Sesam does not have this sub type. Fixed to use LVM
new: (#33016) Extended LIS file version 3. LISv3 for CPIO with -o no_ea,noacl could be started with -i lis=3