GUI Bugfixes: * BUGFIX: (#21283) Excel-Export of MEDIA did not work properly * BUGFIX: (#21283) Excel-Export of CLIENTS did not work properly * BUGFIX: (#21580) Server 2016 RDX action 'introduce' requires volume type and media pool or media label. Problem: '-m' switch was lost when starting 'introduce' on a Server 2016 * bugfix: NDMP access panel not shown for NDMP backup task * bugfix: MySQL access credentials not save in task properties dialog * BUGFIX: (#21672) - Password repeat field checked even if not visible BUGFIX: (#21580) '-m' switch was lost when starting 'introduce' on a Server 2016 bugfix: (#21610) - UI parameter -S problem if sesam srv have short name in DB bugfix: Fix "localFullAccess" and "*@" policy permission handling bugfix: #19059 (Add an automatically generated file name, if restore 'into file' fails is requested) - correct display of file string BUGFIX: (#19059) generates output file names(*.dmp,*.omg) if the user selects a folder. bugfix: VM attach failed;. Problem: unitnumber not unique for added VMDKs BUGFIX: (#20869) VMware Instant Recovery of specific VM fails with "VM Exception: [No scsi controller (VirtualLsiLogicSASController) defined, stop attach vmdk procedure]". Problem: wrong handling of unitnumber, when create VMDK´s (merge:5759f32) BUGFIX: (#21570) Migrated backups at restorewizard are not shown in GUI anymore. Problem: results.sessionID is blank instead of null. Solution: fill table with filter 'results.sessionID=""' (empty string) (merge:068149b) bugfix: Newday event leads to empty "Backups" view if left open from the previous day BUGFIX: sbc_vadp with param -A add_nfs_store -o RO' does not create with read only. Problem: argument "-o RO" will be given as one param. Solution: trim param (merge:0a343ab) BUGFIX: (#21555) Restore of huge vSphere VM fails with "VM Exception: [For input string: ...]". Problem: restored VMDK size is > 2 TB. Solution: use type 'long' instead of 'int' as VMDK.summary config. BUGFIX: (#21543) The interfaces combobox was filled with inappropriate values in migration event window BUGFIX: (#13819) Start date of migration task must not be greater than end date; checkbox 'relative backup date' will not be selected, after change relative date (merge:57a1d38) bugfix: (#21493) cli 'sql -F export' has wrong output;. Problem: table name of export sql query will be cutted at char '_' of table name BUGFIX: (#19222) Permission settings in Java policy file are completely ignored change: Show "access denied" page if server respond with 403 when getting the server info BUGFIX: (#21418) - Dashboard should not ask for authentication if opened from the Java UI BUGFIX: (#20210 ) Error message is missing when enable CBT fail (merge:cbb5694) BUGFIX: (#21501) Client filter always applies previous selection for the "by Status" views BUGFIX: (#20793) Freezed GUI when (from)-date in migrationEvents is increased by buttons (merge:fb0eaf3) BUGFIX: (#20534) - Auto update of 'Calendar Sheet' is not working BUGFIX: (#21487) Client show error 'Invalid Value: Client used as own datamover for mode access mode PROXY' although client use no 'accessmode.PROXY' (merge:9b6118b) BUGFIX: (#21481) Setting the "Days" in "For a duration of" in a schedule changes "ends at" seconds (merge:c971b3a) BUGFIX: (#21482) When I set a media EOL-free in the gui, the loaderinformation is deleted (only in the gui) - refreshproblem (merge:5ebaf37) change: Fix NPE if media pool is not set for media (error line visible in media view) change: (#21388) fix layout of the drive dialog BUGFIX: (#21388) drive option 'options' will not be taken, when modify only this one. change: (#21210) Add a busy indicator when filling the users or the groups table bugfix: (#21210) Avoid a SQLException (constraint violation) when changing a password or a comment in the UserSettingsPanel bugfix: (#21210) The label 'Password' was not visible in the UserSettingsPanel BUGFIX: (#21427) view schedules -> events shows not all planed events. Problem: terms with contition: 'next_end == null || next_end.before(next_exec) || next_end.equals(next_exec)' was filtered Deploy bugfix #20601 dashboard part: Adjust default filters BUGFIX: (#21435) - Target node changes when saving a restore task a second time bugfix: Restore wizard changes target panel from dump to path if the browse button is used in the target dump panel BUGFIX: (#21432) - "Execution Options" not available on a saved restore task with backup type "Write saveset into file" change: Adjust default filter in "Last Backup State" to not include active/queued by default bugfix: (#20601) - Last successful backup and last full backup not filled in last backup state window bugfix: columns 'last successful/full' restore are empty at component 'last backup state' bugfix: (#21238) Java exceptions are thrown when accessing VMs during gui server start new: (#21388) driveoption 'options' is hidden under datastore.drive property dialog ; move textfield 'options' from option tab to main tab at drive dialog new: (#21388) param -O 'options' for cli 'add/modify drive' NEW: (#21388) field 'options' for drive dialog (reference to hw_drives.options) Deploy dashboard fixes for #21365 into 4.4.3 branch BUGFiX: (#21375) Loader action 'unload' not possible when no slot is selected BUGFIX: (#20770) - Restore results "Info 2" shows always "null" as first part of the location tree (e.g. "null/VMware") BUGFIX: (#21365) - Show non existing tasks Filter for last backup state does not filter correct BUGFIX: (#21001) - Remove media actions from backup view change: Fix next execution not correctly calculated for user defined calendar BUGFIX: (#21340) - Delete user defined calendar is not possible bugfix: (#21210) The User/Password for RW was requested any time even if UAC was completely switched off (merged from master) NEW: (#21210) Mandatory Login before opening Restore Wizard (merged from feature branch) bugfix: #21328 - Fix another refresh issue of the data store saveset table after changing the EOL BUGFIX: (#21175) Saveset EOL icon does not update correctly if setting the saveset EOL to "Now" BUGFIX: (#21328) - Adjust EOL of the whole chain if the EOL of (for example) a DIFF is changed to the future bugfix: (#18783) run VM diff backup with resetCBT use wrong 'ssmoref' BUGFIX: (#21329) Generational restores are always started even if "generational Restore" button is disabled change: (#20664) - Split GUI jar file BUGFIX: (#21298) - Copying a backup task within a task group deletes all other tasks from the group BUGFIX: (#21295) - Java NullPointerException when opening loader action dialog of virtual loader bugfix: (#21239) - Nach einem Tageswechsel wird die laufende Ruecksicherung unter Job Status Ruecksicherung nicht angezeigt bugfix: (#21239) - Improve fix to re-add check for newday event from timer, but with taking the date cache into account BUGFIX: (#21239) - Nach einem Tageswechsel wird die laufende Ruecksicherung unter Job Status Ruecksicherung nicht angezeigt bugfix: (#21277) NPE, when open schedule. Problem: client.sesamVersion is null BUGFIX: (#21273) - Citrix XEN: cannot change credentials of XEN User in task configuration BUGFIX: (#21244) Reduce EOL of a DIFF backup BUGFIX: (#21247) - Cannot create drive for my newly created loader bugfix: (#20939) Create migration task from tab 'Media'. Problem: retrieve 'resultslbl' get full results/media object. Solution: flat fill for fields resultLbls.result;resultLbls.media bugfix: (#19794) TaskGroup name is displayed in Performance Panel (RunningTasks) bugfix: (#19564) In job state backups with F5 all opened trees now stay expanded (add in master and 4.4.3) BUGFIX: (#21103) Auto update doesn't refresh state of media actions change: loader load/unload commands require the drive number set, not the port CHANGE: Explicitly allow slot numbers to exceed the maximum number of slots configured in the DB. The DB tells us the number of licensed slots only. change: Port or slot to check for max value cannot be negative at this point of time bugfix: (#19099) Fix max value checks if a range is specified for port or slot BUGFIX: (#19099) No port/slot range possible in loader actions BUGFIX: (#21186) resetcbt ends with Exception 'Invalid Login to com.vmware.vim25.InvalidLogin' BUGFIX: (#21162) Avoid NullPointerException when adding new NDMP client SRC Bugfixes: Git ID 6c6b7fe * change: GetPrivateProfileString() if fopen failed then return message with errno and strerror * BUGFIX: (#21666) Migration started via 'follow up event' from a task_event for a task_group failed with 'E001-COPY No save sets found'. Problem: The migration was called with a task_group but select statement was executed for session. Solution: Distinguish between usage with '-G {session_id}' or '-J {task_group}' / migration_tasks.task_group * bugfix: (#21544) Restores from si3 store fails with '553 RETR Failed. NO ERROR: (0x0,0) (0) - ChunkNotFoundException(java.nio.HeapByteBuffer,...)'. Problem: Message from Si3 dedup is not returned. Solution: In case of Si3 dedup do not set message from errno BUGFIX: (#20946) - sbc cores if a backup task is canceled by schedule setting "Stop task if it runs longer than". Improved the error handling mechanism of dynamical called functions. change: Allow to set SQLITE_BUSY timeout with GLBV gv_sqlite_busy_timeout, default: 120 BUGFIX: (#21559) - after update backup fails with "[No such interface supported/Schnittstelle nicht unterstützt]". Unused interface IVssBackupComponentsEx3 removed. BUGFIX: (#21459) File blocked by a virus scanner must not lead to a backup error. Problem: Even with Volume Shadow Copy some files are not accessible. Solution: New option '-o ignore_vss_access_denied' ignores 'item not accessible' and backup will end with warnings BUGFIX: (#21568) Migration of some savesets fails with 'E001-DATABAS Failure during access to database'. Problem: The original saveset was deleted from table results coz saveset_exist==0 due to warning during migration. Solution: Update results.saveset_exist when migration ends with warning. When original saveset is deleted then keep information in results when a migrated saveset exists. bugfix: (#21377) - slu handles scsi devices without drivers bugfix: (#21187) use scsi H,T,L when using device id bugfix: (#21467) - add density code for LTO-7 bugfix: (#21383) - connect drives to library if library has > 80 drives bugfix: (#21383) - display slots as empty if slots cannot be accessed change: slu makes better error reporting if device not accessible bugfix: (#20086) Authentication: Make some changes in the Permission Management in GUI. Problem: Authentification menu is enabled in GUI in 4.4.3. Solution: Insert 'enable_permission_menu' with value 0 to disable Authenticatio menu after installation or update to 4.4.3 - will not update the value if already set! BUGFIX: (#21241) SBC Crashes when reaching "System Volume Information\DFSR" Folder. Problem: BackupSeek() sometimes failed and the already retrived ADAT stream is not processed. Solution: Improved checking of BackupSeek(): Read only stream name with BackupRead, treat return zero with ERROR_SUCCESS as ok BUGFIX: (#21241) SBC Crashes when reaching "System Volume Information\DFSR" Folder. Problem: BackupSeek() failed after previous item. Solution: Release handle and context after BackupSeek() returned error BUGFIX: (#21241) SBC Crashes when reaching "System Volume Information\DFSR" Folder. Problem: After recent commit backup of alternate streams failed with ERROR_ACCESS_DENIED. Solution: If last BackupSeek() failed then go on with retrieved alternate stream BUGFIX: (#21241) - SBC Crashes when reaching "System Volume Information\DFSR\" Folder. Handle the status of BackupSeek() function caused the core dump of sbc under local user. BUGFIX: (#21241) - SBC Crashes when reaching "System Volume Information\DFSR\" Folder. Handle the status of BackupSeek() function caused the core dump of sbc. bugfix: (#21288) sbc cores during restore of encrypted and/or compressed path backup with alternate data streams. Problem: If buffer is insufficient stream name is written to the wrong place. Solution: do not proceed with stream data processing until the whole stream name is written. change: (#19593) Windows backup must handle the Alternate Data Streams. Problem: BackupRead() must be called with abort flag set to TRUE to release backup context. Solution: Add extra call to BackupRead() NEW: (#21507) Allow media readability check without having media_read_check license. Problem: Media read check is essential for SEP sesam media recovery. Solution: Set default license value for media_read_check to 1 BUGFIX: (#21456) Backup of a MS SQL DB fails sporadically on a customer system - 15 seconds timeout. Problem: Default timeout is 15 seconds. Solution: Set timeout to 300 seconds with SQLSetConnectAttr(hdbc, SQL_LOGIN_TIMEOUT, 300, ...) BUGFIX: (#21497) Restore of migrated saveset failed with 'ATTENTION: Saveset SF20161224085907277@dTMw8yEeDF1 is not in FDI-List so skip it!'. Problem: In case of generation restore the FULL_DIFF_INCR_CHAIN() must return original_saveset. Solution: Distinguish between generation restore and saveset copy via iGen == 1/0 bugfix: (#20960): 'sm_sbc_com' cores during DB2 restore when a previous restore is incomplete. ('PROT() has been called with a variable including '%' itself) BUGFIX: (#21475) Linux ppc 64: sm_sms_interface getlabel failed with STATUS=ERROR MSG='Error in GetLabel: 1013: Failed to read from tape. Unknown data format.'. Problem: Shared memory arithmetic failed. Solution: Use 'extern size_t xOffset' in all modules to asure use of same global variable BUGFIX: (#21476) DB2 coredump during restore. Problem: XBSA ObjectDescriptor with different sizes due to XBSA preliminary API. Solution: Use new XBSA standard for sdb2 bugfix: (#21323) SM_SSHD sub-process hangs. Problem: In case of high-loaded system sm_sshd sub-process doesn't start corre ctly, but got PID and reported as alive. Solution: After sub-process start, main process should get confirmation otherwise - kill sub-process BUGFIX: (#21468) Increase timeouts in 'sm_ssh' and 'sm_sshd' to 120 sec. to allow more remote connections at same time on heavly loaded systems new: (#21420) Add 'sm_install_saphana' to Sesam client kit on Linux NEW: (#21413) Add property 'enable_permission_menu' to the defaults table. Problem: 'enable_gui_rw_mandatory_login' required too NEW: (#21413) Add property 'enable_permission_menu' to the defaults table. Solution: sesam_insert_start.sql now adds 'enable_permission_menu' with value '0' BUGFIX: (#21428) Exchange 2010: Public folder Backup always fails with "bBuildSourceForDb() failed: Did not find a copy of database [Public Folder]". Handle backup of Public Folder Database on Exchange 2010, backup of Exchange 2007 cluster system; bugfix: (#20656) - Exchange DAG backup does not start if just one node is up and passive DB is preferred, but active also possible. Changed logic around getting passive copies of db by using get-mailboxdatabasecopystatus cmdlet. bugfix: (#21227) Warning: "Setting of standard attributes for item failed" during restore in windows server 2016. Handle ERROR_PATH_NOT_FOUND error for post-processed directories. NEW: (#21128) 'Windows Server 2016' is missing in sesam oper_systems table. Problem: sbc returns 'sbc-2005: Warning: Operating system [Microsoft Windows Server 2016, Build: 9200 (x64)] is not supported!' and ends 'with warnings'. Solution: Accept 'Windows Server 2016' as supported operating system. bugfix: (#21160) VMware CBT restore via Windows proxy stucks at the beginning new: (#16333) Backup of CIFS shares with user name/password authentication. Problem: Unable to backup password protected CIFS. Solution: Support backup from CIFS using Username and Password NEW: (#17936) Encrypt deduplicated data for Si3 Dedup Store. Problem: sm_dedup_interface doesnt have any information about encryption state. Solution: request and print information about encryption state change: (#17936) Encrypt deduplicated data for Si3 Dedup Store. Windows: If gv_rw_ini:/stpd_conf/{datastore_drivenum}.pwd file exists then set -Dconfig.passwdfile={pwd_file} new: (#20843) Secure Si3 replication by SSL/TLS. Problem: There is no access control or encryption for replication connection. Solution: Configure a 2nd port for encrypted replications. BUGFIX: (#21401) vSphere restore takes too long coz tape is rewinded and read from beginning. Problem: Start segment is set to 1 if segment file is missing. Solution: Set start segment to 1 only in case of datastore BUGFIX: (#21405) BSR Pro backup on win 32bit fails with 'BSR (DiskImage) registry tree not found'. Problem: The registry hives for x64 have been checked. Solution: Correct platform check with platform.machine()=='x86' BUGFIX: (#19947) Meaningful message must be returned if BSR Pro backup failed due to missing installation. Problem: Currently a Traceback appears with: ''global name 'BSR_Pro_regkey' is not defined''. Solution: Set BSR_Pro_regkey to a defined registry value BUGFIX: (#21386) 'sm_arch' writes barcode as '-u' into 'DB:media.barcode', if called with empty barcode by 'sm_robot'. Use 'i_cm_system()' instead of 'i_cm_exec()' to avoid wrong parsing of empty arguments. BUGFIX: (#21384) Core dump of 'sm_robot', if same barcode used for multiple media. (Sesam log message has been called without correct parameter) bugfix (#21226): Setting shortname fails during path restore on ReFS in windows server 2016. ReFS doesn't support setting short name for file. bugfix: (#21079) Domino Server crashes in case a DB is not existing bugfix: (#18001) SESAM_BACKUP does not backup var path due to Reparse Point. Add informative log about traverse FS in case of -o follow option is set BUGFIX: (#18001) SESAM_BACKUP does not backup var path due to Reparse Point. Support reparse points in the VSS - add INFO trace; bugfix: (#18001) - SESAM_BACKUP does not backup var path due to Reparse Point. Keep original path of the reparse point in the saveset in case of VSS backup bugfix: (#18001) - SESAM_BACKUP does not backup var path due to Reparse Point. Implemented based functionality for working with reparse points; BUGFIX: (#21338) Error from 'sm_arch' during introduction of 1st medium for a data store. Call of 'sm_arch' is protected by a semaphore now, to avoid DB duplicate key error BUGFIX: (#21354) ADJUST EOL reduces media.eol. Problem: The media.eol was not compared to the max(eol) from result_lbls. Solution: Get media.eol for all used media and compare if eol is higher than max(eol) from savesets in result_lbls stored on that media BUGFIX: (#21276) Windows all backup saves to much data, e.g. C:/$Recycle.Bin/.... Solution: Modify default exclude list entries: '?:/$Recycle.Bin/', '?:/recycled/', '?:/recycler/', '?:/found.???/' and add '?:/swapfile.sys' BUGFIX: (#21293) Update 'DB:data_stores.stored' value after 'purge' operation. Value has been updated by 'check_space' only BUGFIX: (#21177) 'sm_data_store check_db/fsck' Do not report missing data file, if saveset filename includes a '.' (like Of_XE_f5rl61j7.dbf.data). Happens especially, if 'sbc_oracle_rman.sh' script is used BUGFIX: (#21366) NDMP backup fails on Windows. Problem: The issue was already fixed on master with commit 1e099d4: Fixed check for spGlobal->sLocalSaveset.iFilDes == ERROR on Windows. Solution: Merge that fix to v4_4_3 to set up XBSA connection BUGFIX: (#21263) Update must adjust result_lbls.eol to result.eol for failed backup to allow purge of failed backup save sets. Solution: Add database type specific statements to [Update 4.4] NEW: (#21360) FrontSide license count must ignore data occupied by not longer configured backup tasks. Solution: If save sets of obsolete backup tasks are protected by EOL then these savesets are not counted in FrontSide. The backup tasks are listed under 'FrontSide / Backup Task is no longer configured' BUGFIX: (#18592) Activate date range filter for replications BUGFIX: (#21361) Generational restore with multiple tapes fails with bad archive as wrong tape is loaded. Problem: The EOM procedure loaded wrong tape due to saveset_used not set for original saveset. Solution: Set restore_results.saveset_used for original saveset too. BUGFIX: (#21296) Archive adjustment for virtual loader fails with 'Failure from bar code reader: unknown error:'. Problem: In case of virtual loader the drive properties must be read to allow further processing. Solution: Get drive properties for virtual loader 'DIR_VIRT' BUGFIX: (#21135) Read of tape trailer of an encrypted LTO tape fails. This caused a new media initialisation for every backup, which is appended to an encrypted medium. BUGFIX: Fix SQL error in license module during calculation of front side volume. (LIC_COUNT_VOLUME_BASED_TB with unexpected result: ERROR: column tasks.client_id must appear in the GROUP BY clause or be used in an aggregate function) BUGFIX: (#21341) EOL free tape in drive with not expected label is initialized with expected label. Problem: The expected label is used for initialization. Solution: Remember label returned grom GET_MEDIUM and use is for initialization BUGFIX: (#20873) Add sm.ini ExcludePattern for 'gv_rw_work:/sem' to avoid backing up of temporary work files. Solution: Add in sm.ini 'ExcludePattern14=gv_rw_work:sem' add logic in sbc to expand gv_... prefix for exclude patterns BUGFIX: (#21306) Saveset tree return success, if a saveset in FDI chain is missing. Problem: Error is already returned in FDI_CHAIN() but i_GET_SAVESET_TREE() did not use it. Solution: Improved error handling: Print available saveset tree information with message 'E017-BACKUP In the saveset chain of '%s' the backup '%s' is missing.' BUFIX: (#21313) Windows: MaxDB Restore not possible: backint module is unable to write to communication pipe. Problem: On Windows the command os.path.exists({pipe}) is closing the pipe immediately. Solution: Check if 'pipe' appears in the target pipe name in case of Windows BUGFIX: (#21308) MaxDB on linux: backup disfunctional 'unknown format in the Backint for MaxDB output file'. Problem: backup_id exceeded the size of 16 chars. Solution: Build backup_id with max. 6 chars from database name and 10 chars from timestamp bugfix: (#20386) - sm_vfs - rework logging for VMDK-CBT to shrink log size bugfix: (#21258) - sm_vfs VMDK-CBT check if name is valid before extracting offset and size parameters bugfix: (#21257) - calculate 64bit file size correctly from CPIO stream NEW: (#21066) sm_cmd: Transfer all connection related options, like server, user and password to the 'java' call. Before password for example wasn't set bugfix: (#21187) use scsi H,T,L when using device id BUGFIX: (#21282) MaxDB Restore not possible: backint module is unable to parse sm_sbc_com inquiry. Problem: Since commit e69eb70 '(#19403) Windows: MaxDB backup failed due to path with blanks' the executable is enclosed in double quotation marks but Windows does not execute commands with further arguments also enclosed with double quotation marks which. Solution: Avoid to set -j "" if caller does not provide job (task) name BUGFIX: 'DB:VIEW:all_results' hasn't been created during Sesam update and installation' BUGFIX: (#21248) If FULL backup fails then result_lbls.eol is not set and purge will not remove that save set. Problem: result_lbls.eol for failed backups must be set in the beginning. Solution: Set eol to 3 days as it was in former SEP sesam versions bugfix: (#21101) Restore failed: Error while unpacking data. GetStreams: ... no valid MTF Stream ID was found. Problem: SPAR stream has unique "Sparse Frame Header", which length should be used for Offset. Solution: Increase offset in "Sparse Frame Header" field length BUGFIX: (#21224) "CVssBootManager::IsEFI(): Failed to get UEFI configuration!" error on none UEFI system(Problem: bcdedit tool wors unstable when several instances are invoked at one time. Solution: Use mutex to synchronize bcdedit access) BUGFIX: (#20803) Restore at a customer machine doesn't work - NetApp02 -> Backup in DS ok -> Migration to Tape -> ok -> Backup deleted -> Restore from Tape fails. Problem: not all of file attributes read if error occurred, so the next file is read with incorrect offset; Solution: read all of the file attributes, so offset is correct) bugfix: (#21121) emulate writing to merged files; content written to memory bugfix: (#21121) fake writing for all file types but ft_pureVirtual BUGFIX: (#21221) Migration of INC/DIFF backups ends with warning if referenced migrated save set is also replicated. Problem: The matching media pool priority is overwritten by the priority of the replicated saveset (here: 'DATA_STORE'). Solution: Reset priority only if new priority is higher - 'REQUESTED_POOL' is the maximal priority! BUGFIX: (#21005) Si3 replication increases retention time of original saveset to EOL of media pool used for replication. Problem: Only the first update command used SQL clause 'and label like '{pool}''. Solution: Use SQL clause for further savesets in chain in case of migration and replication BUGFIX: (#20803) Restore at a customer machine doesn't work - NetApp02 -> Backup in DS ok -> Migration to Tape -> ok -> Backup deleted -> Restore from Tape fails(Problem: incorrect ACL logic; Solution: Handle extended attributes independently).