Change report 5.1.0 Apollon V2

From SEPsesam


Change Report for Apollon V2

Updated: May 23, 2024 | The purpose of this change report is to list the changes, enhancements and fixes in the SEP sesam software.

===5.1.0.17 Apollon V2===
Change report
  • BUGFIX: (#33465) ReaR backup on SEP sesam RDS failed due to missing java libraries: exlude java symlink from REAR image by default
  • BUGFIX: (#33565) SQL error in Sesam main log 'column stop_time of relation cal_sheets does not exist'. Use correct DB:cal_sheets.end_time column name. Issue has no influence on proper Sesam execution
  • BUGFIX: (#31474) Backup with many sources should not fail if one of the sources does not exist. Problem: In the case of a UNC source, the option '-o ignore_missing_source' was ignored.
  • BUGFIX: (#33548) Remove update statements with ::timestamp from sqlite sesam_insert_start.sql
  • BUGFIX: (33466) Selective Path restore requesting data from first tape failed: 553 RETR Failed. Data Synchronization lost - Restore terminating. Problem: SGM file without 'Total' line in version < 5.0.
  • BUGFIX: (#33533) Syntax error in Sesam Oracle backup script on UNIX
  • BUGFIX: (#33426) In case of several Exchange server databases, only the number of mailboxes in a single database is counted. Solution: Sum up all mailboxe, count DAG databases only once. Add detail information to license report.
  • BUGFIX: (#33524) Prevent sm_dedup_interface "disk_used" when SI3-NG is offline
  • CHANGE: (#33467) Check if 'sm_sdslog2.xml' exists before copy
  • BUGFIX: (#33467) Do not exit update routine if copy of sm_sdslog2.xml fails, since sm_sdslog2.xml is not present on client and therefore cannot be copied.
  • BUGFIX: (#33514) Replication uses wrong source drive because of a race condition between DB update and DB select. Add DB:migration_results.state='W' and set it to 'Q' during update to correct source drive
  • BIGFIX: (#31493) Suppress event with higher prio: Do not use default start time window of 3 days to set 'DB:cal_sheets.end_time_set', which is used for priority calculation, but set it to 'NULL'. So Sesam day will be used.
  • CHANGE: Do not allow update of Sesam client/RDS from version < 4.4.3.84 to 5.1.0.x
  • BUGFIX: (#33381) Reduce number of 'Empty Trash' entries in 'DB:media_results'. Do not write DB entry if no data has been purged
  • NEW: (#33499) Do not allow Sesam startup in 'sm_main' as user 'root' on SiS server
  • CHANGE: (#33497) Re-work update part in SiS creation script
  • BUGFIX: (#33497) SiS Si3 service doesn't start after first configuration.Initialize SiS Si3 after Si3 INI file has been created, so custom UUID will be used
  • BUGFIX: (#33464) Update Sesam REAR during Sesam server update too
  • BUGFIX: (#33467) Windows update does not update sm_sdslog2.xml. Execute copy from 'templates' to 'ini' directory during Sesam update on Windows too
  • CHANGE: (#33395) Do not use '+' for space in URL encoding as default. This fix Xen restore with relocation if VM name contains spaces and 'umlaut'
  • BUGFIX: (#33483) Removed logout from ISCSI target if no new ISCSI device is found after SAN volume snapshot
  • BUGFIX: (#33421) The sm_sms_watch log file of a remote device contains many log lines about non-existent processes33421 - The sm_sms_watch log file of a remote device contains many log lines about non-existent processes. Solution: Remember that the reading performance process/thread is already finished.
  • CHANGE: (#33381) Reduce number of 'Empty Trash' entries in 'DB:media_results'. Do not create DB entry, if less than 1 MiB has been removed
  • CHANGE: Allow Java 11 for 5.1 release branch again
  • BUGFIX: (#33438) Avoid unnecessary buffering of written data to DataStore on Linux. Compilation on Windows failed
  • NEW: (#33438) Avoid unnecessary buffering of written data to DataStore on Linux. Solution: Call posix_fadvise64(..., 0, %llu, POSIX_FADV_DONTNEED) after 1024 written blocks
  • BUGFIX: (#33418) CBT backup not possible for newer OVIRT/OLVM Releases: check supported versions by real major number instead of string compare
  • BUGFIX: (#33416) Increased length of MAX_LIS_ENTRY_PATH to 4094
  • BUGFIX: (#33417) Immediate deletion after successful migration of an already migrated saveset uses the wrong saveset. Problem: sm_remove_saveset purge is called with backup saveset. Solution: Use source saveset. Additional: Trace list of valid savesets based on result_lbls.backup_id
  • CHANGE: (#32118) Correct dependency for Java 17 on SLES 15: Add 'jre >= 17 or java-17-openjdk'
  • BUGFIX: (#33406) Exclude savesets from migration where migration job already tries to occupy destination drive. Add DB:migration_results.state='A'
  • BUGFIX: (#33229) Kopano backup ends with TypeError: a bytes-like object is required, not 'str'. Solution: Python3 requires to decode byte to string
  • BUGFIX: (#33458) DB:media_results entries for datastore 'readcheck' are created using same session id and shown as group entries in UI. Reset session id before new DB entry is created
  • BUGFIX: (#33462) Linux exclude with ending \.+ does not exclude content of directory
  • BUGFIX: (#33463) SBC Windows include file saved with unicode encoding reads BOM as part of pattern
  • NEW: (#33216) Write additional heap dump file in case of Si3 OOM error
  • NEW: (#33070) Replication jobs marked as failed due to access errors using 'sm_ssh'. Add retry for get savesets from target store function
  • BUGFIX: (#33469) sm_replicate ends with 'str' object has no attribute 'decode'
  • BUGFIX: (#33459) - Maximum capacity check at DataStore properties must be adjusted when using binary format output
  • BUGFIX: (#33434) sbc on Windows cores when using include filter. Solution: Support of multiple include expressions and correct handling of first character in regular expression.

5.1.0.15 Apollon V2

Change report
  • BUGFIX: opennebula snapshot timeout not applied: increment retry counter accordingly
  • BUGFIX: (#32971) Selective Windows path restore fails with 'Next item or ESET is expected' using LISv2.
  • BUGFIX: (#33259) Restore from tape must check the given label.
  • BUGFIX: (#33097) Temporary migration tasks are not deleted for multi saveset backups. Delete task by last sub task, which has finished
  • NEW: (#33148) Exclude '.vswp.<PID>' files from vSphere configuration backup
  • NEW: (#30728) Do not start data store housekeeping, if data store is not reachable
  • BUGFIX: (#33181) Cannot read from tape device. Reason: Input/output error: Failed to read 524288 byte block with 65536 byte transfer.
  • CHANGE: (#33136) Start Java based daemons without console output option. Use 'javaw.exe' instead of 'java.exe' on Windows
  • BUGFIX: (#33276) Migration fails if EOM handling for target drive needs longer than 900 sec. Increase timeout to 1 day
  • BUGFIX: (#33111) Fixed error handling for iscsiadm not installed on the data mover
  • BUGFIX: (#33126) Selective restore from imported backup fails. Set 'DB:result_lbls.saveset_count>=1', so restore will not check label read from LIS file
  • NEW: (#33123) Pending VMware SFR restore task has to execute 'umount', if canceled
  • CHANGE: (#33112) Selective restore from linux backup on disk fails with .... change: Always trace STPD message 'go to segment number ...' and SMS message 'restore_operation: requested segment number ...'
  • CHANGE: (#31260) Backup to tape using block size up to 1 MB. Increase from 512KB to 1MB
  • BUGFIX: (#33201) During installation the SEPsesam\bin folder in ProgramData was not created. Now the script directory is also created during installation, not just at update.
  • NEW: (#31079) Sesam installation fails on RHEL 8.5/9 and SLES12 because Java not completely configured. Add '/usr/lib64/jvm/jre-11-openjdk' to search path
  • NEW: (#33228) Support PostgreSQL 16 for Sesam DB. Use 'postgres' instead of 'postmaster' as executable name for Postgresql server
  • BUGFIX: (#33218) Avoid string truncation during restore for update of 'DB:cal_sheets'. (Error vsnprintf truncated 558 >= 512 during)
  • NEW: add ini entry for maximum session inactive time config
  • NEW: (#33284) Backup failed with not supported reparse tag REPARSE_TAG_DROPBOX_HSM (0x00000044L). Add support for Dropbox HSM reparse tag.
  • BUGFIX: (#33279) After update, SESAM_BACKUP fails with 'invalid backup source'. Change incorrect source part 'gv_ro:skel/templatessesam/' by 'gv_ro:skel/template'
  • NEW: (#33216) Abort Si3 server in case of Out-of-Memory errors and write JVM crash dump file into '..log/sms' directory. Adapt log file path on Windows
  • NEW: (#33328) Add 'sm_vfs' to Sesam Linux client kit
  • BUGFIX: (#33185) sm_break2 could not terminate active task group result entries. Do not check for active task group entries during Sesam shutdown
  • CHANGE: (#33332) Reduce heap memory allocated by Si3 server, because index file is read using memory mapped IO instead of loading it into heap space
  • NEW: Si3-NG propose serverconfig show for what RAM is reserved (-v 2)
  • NEW: (#33360) Sesam server installation on Ubuntu 24.04
  • BUGFIX: (#33350) Citrix COPY level backup sporadically fails with "Curl: OpenSSL SSL_read: error:0A000126:SSL routines::unexpected eof while reading, errno 0"
  • BUGFIX: (#33353) Umount of saveset fails when backup is migrated on DataStore of another RDS
  • BUGFIX: (#33366) Proxmox backup of VM on ZFS fails with "SBC logging without final state message"
  • BUGFIX: (#33383) Abort of Hyper-V backup does set reference point files for DIFF and INC backups

5.1.0.14 Apollon V2 SP2

Change report
  • BUGFIX: VFS log was flooded and VHD mount hanged on Windows when log level was set to 2
  • BUGFIX: (#33143) Mismatch in UNIT license count between 'sm_info' and CE report. Problem: VIRTUAL clients may be counted twice. Solution: Do not count clients with access_mode 'VIRTUAL'.
  • BUGFIX: Reparse point is saved as directory by Windows path backup without 'VSS'. Bugfix: Avoid that after delayed directory processing a reparse point (e.g. Microsoft Windows Data Deduplication IO_REPARSE_TAG_DEDUP) could be saved as directory.
  • BUGFIX: Allow setting of 'pgdump'/'mysqldump' options by 'DB:tasks.backup_options'
  • BUGFIX: Proxmox backup fails with error Unhandled exception occurred [too many values to unpack (expected 2)]
  • NEW: Support of Java 21 for Sesam on Linux - support for SLES15 SP6
  • BUGFIX: MaxDB backup fails with 'Invalid backup source [2>&1]', because started without 'cmd.exe'. Remove redirect, because 'sm_Popen_buf()' already returns STDOUT and STDERR
  • BUGFIX: Follow up jobs are not being started anymore when cancel option of newday is set. Solution: Call 'sm_qm set option follow_up 0' only for caller 'maintenance' or 'shutdown'

vSphere:

  • BUGFIX: Incorrect disc size when calculating the changed blocks

GUI/WebUI:

  • BUGFIX: Creating a schedule with execution time every Friday will be executed every day from Friday in the future
  • Solution: Limit the maximum end time to 23:59h for weekly schedules.
  • CHANGE: Change message, if the VSS option for a Windows path backup is switched off
  • BUGFIX: correct link to custom calendar description
  • BUGFIX: CLI writes 'null' to data base when invoking "add replicationtask" with -B parameter
  • CHANGE: Update generic report and adjust report header in all reports.
  • BUGFIX: WebRA wrongly sets -a resourcepool for a vApp instead of -a vapp

5.1.0.14 Apollon V2 SP1

Change report

Kernel/Backup Modules:

  • BUGFIX: After update to 5.1.0.14 the access to XEN hypervisor doesn't work sometimes
  • Hyper-V
  • BUGFIX: Backup of dynamic disks with DIFF/INCR backup level
  • BUGFIX: Single File Restore (SFR) of Hyper-V INC/DIFF backup from dynamic disks
  • BUGFIX: OpenNebula - Restore with skip_net option failed with "Error initiating VM creation"
  • BUGFIX: BSR Windows
  • BSR Restore selection: The selection of the save set and the disk to be restored was very slow
  • BSR Recovery did not work due to an internal error
  • BUGFIX: Single item restore using VM mount on Windows does not work, except in case of backup type vSphere.
  • BUGFIX: Si3 replication: Massive logging during replication could lead to poor performance and even shutdown of Si3 service
  • vSphere
  • BUGFIX: After update to Apollon V2 (5.1.0.14), the first INCR vSphere virtual machine backup fails

GUI/WebUI:

  • BUGFIX: Server seems to require a client certificate from the browser for SSL handshake
  • BUGFIX: Splash Screen is badly ordered
  • BUGFIX: Exclude of VMDK in WebUI during vSphere restore does not work
  • CHANGE: WebUI must offer either a folder or a vApp as destination at vSphere restore, not both - Reject next step if both are set.
  • BUGFIX: SFR target path must accept UNC notation - Do not remove leading '//'.

5.1.0.14 Apollon V2

Change report

GUI Bugfixes: 5c41c64d90

Web-UI Bugfixes: ce72b537

Web-UI NG Bugfixes: b8cad2285

Kernel Bugfixes:

  • 4fb94869b4
  • BUGFIX: (#33174) Selective restore from CPIO backup on further tape fails with 'Skipping tapemarks on tape failed.'.
  • BUGFIX: (#33132) Info meta file backup to SMS when tape is idle and after migration
  • BUGFIX: (#32069) vSphere restore of backup made with 'single saveset' option fails
  • BUGFIX: (#33155) New LISv2 for most backup types except NSS (SIDF) or encrypted saveset (BF64 or AES)
  • BUGFIX: (#33104) DB2 Backup failed. Problem
  • BUGFIX: (#33139) Selective restore of a replicated backup fails. Solution
  • BUGFIX: (#33110) SAP ASE backup failed with Error
  • BUGFIX: License report for SEP Unit shows FrontSide Storage with B instead of GB
  • BUGFIX: Si3
  • BUGFIX: (#33069) do not check the page that still writing on backup
  • BUGFIX: (#31712) optimize incr PCCK iterator by using pages index
  • BUGFIX: extend page deep check condition using max allowed uploading time parameter
  • BUGFIX: (#33069) flush page body file after each chunk append
  • BUGFIX: (#33081) Successful status set to warning by restore 'POST' interface
  • BUGFIX: (#33062) Windows backup with Reparse Point to folder under root fails with 'Error while processing standard attributes', e.g. 'C
  • BUGFIX: (#33068) Update 'sm_rescan.sh' to recent version to avoid crash of VM during attach
  • BUGFIX: (#32920) vSphere mount/attach using savesets made with Sesam version < 5.0
  • BUGFIX: (#32970) Mount of vSphere saveset using URL encoded path names doesn't work. Do not URL encode '[' and ']' in vSphere data store names for LISv1
  • BUGFIX: (#32603) return status with error if sbc_proxy cant import libguestfs-tools
  • BUGFIX: Replace non UTF-8 characters in Dovecot mail subject lines
  • BUGFIX: (#33038) Do not report directories as processed during 'plain' restore on Linux
  • BUGFIX: (#33056) Many 'Curl
  • BUGFIX: (#32710) Print Sesam backup modules included in 'sbc_proxy' in one line and with own sbc message number
  • BUGFIX: (#32882) sbc_rhev
  • BUGFIX: (#33052) fixed Citrix Restore on Windows. Restore target returned from XEN sbc_pre was not double qouted. Check for double quotes in sbc_proxy and add them to target returned from a module if not already double quoted
  • BUGFIX: (#31255) Write a 'tar' archive with saveset META of complete media after backup has been done.
  • BUGFIX: (#33044) OpenNebula restore to different datastore. Add new datastore to relocation parameter of 'sbc_proxy'
  • BUGFIX: Si3
  • BUGFIX: sbc_rhev
  • BUGFIX: (#33039) Extended Linux file attributes not saved. Correct cast for getting EA buffer size
  • BUGFIX: (#32181) Complete LIS v2 for MTF savesets. Problem
  • BUGFIX: (#32971) Selective Windows path restore fails with 'Next item or ESET is expected' using LISv2. Problem
  • BUGFIX: (#32971) Selective Windows path restore fails with 'Next item or ESET is expected' using LISv2. Problem
  • BUGFIX: (#33023) Hyper-V RCT/vSphere CBT mount of FI and FDI backup fails. Problem
  • BUGFIX: (#25968) Check also for empty variable when checking checkbox for hide servicepack dialog.
  • BUGFIX: (#32950) Cancel of Hyper-V backup must remove VM snapshot
  • BUGFIX: (#32971) Selective Windows path restore fails with 'Next item or ESET is expected' using LISv2. Problem
  • BUGFIX: (#32603) Print traceback in case of exception in sm_mount arguments
  • BUGFIX: (#32761) add exponential retry on s3 requests; do not add page to corruption store if PCCK fails because of s3 connection issues
  • BUGFIX: (#33023) Hyper-V RCT/ vSphere CBT mount of FI and FDI backups fail sporadically. Add sm_sort to SEP sesam binaries on Linux
  • BUGFIX: (#33023) Hyper-V RCT/ vSphere CBT mount of FI and FDI backups fail sporadically. Problem
  • BUGFIX: (#33025) Fixed volume prefix in Pure storage dir
  • BUGFIX: (#32181) Complete LIS v2 for MTF savesets. Problem
  • BUGFIX: (#32181) Complete LIS v2 for MTF savesets. Problem
  • BUGFIX: (#33020) Selective restore of CPIO saveset including ACL and EA in case of LisV2. Solution
  • BUGFIX: Ignore encoding errors by read of LIS file during saveset mount
  • BUGFIX: (#32983) Remove protocol prefix like 'http
  • BUGFIX: (#33000) Use correct BAT entries when mounting Microsoft Hyper-V backup.
  • BUGFIX: (#25968) the checkbox responsible for hiding dialog asking the user if servicepack installation should be executed was not hidden in the next update.
  • BUGFIX: (#32960) Call 'sm_mount umount -s ALL' during shutdown of Sesam server only
  • BUGFIX: (#33010) sbc pipe communication blocks itself by semaphore if VDDK is missing on system. Problem
  • BUGFIX: (#33004) If existing VFS mount exists, skip adding the same info again to vfs_mounts.json
  • BUGFIX: (#32761) add setRetentionPeriod requests to queue; extend s3 retry strategy; regulate logs level in PageFilesTracker
  • BUGFIX: (#25176) Enlarge size of 'sm_sbc_pre' command string to hold handle backup source
  • BUGFIX: (#32905) MSSQL restore with lisv2 is successful, but does not restore any data. Solution
  • BUGFIX: (#32985) If disks cannot be unmounted, write warning log and continue with VFS stop
  • BUGFIX: Start 'guestmount' in own process group to avoid killing it by 'sm_sshd' if 'sbc_proxy ' terminates
  • BUGFIX: (#32985) bat script to detach virtual disks from Windows is kept after mount and is deleted during unmount. diskpart_umount is called during unmount so that all attached drives are detached
  • BUGFIX: (#32984) Fixed XEN Copy restore.
  • BUGFIX: (#32620) Hyper-V RCT snapshot name not unique. Problem
  • BUGFIX: (#32982) Fixed VHDX path for mount. Always set location for VHDX to MAPPER/merge_result
  • BUGFIX: (#22150) sm_break doesn't work with a task-group. Problem
  • BUGFIX: (#29404) Changed semaphore to global for SAN backups. Changed rescan back to scsiadm -m node to rescan only relevant target for new devices. Lowered rescan iterations to max 2
  • BUGFIX:(#30917) Fixed removal of Huawei snapshots. Improved code in sbc_oceanstor by implementing new function validate_response() to check for errors in response after API calls to Oceanstor
  • BUGFIX: (#32721) Do not write invalid time string into DB
  • BUGFIX: Remove segment file on RDS instead of Sesam server after migration has finished
  • BUGFIX: (#31936) Use default ISO date format for 'DB
  • BUGFIX: (#29404) Changed iscsiadm command for rescan target to what is more correct according to documentation
  • BUGFIX: (#29404) Fixed check for backup type VSPHERE CONFIG
  • BUGFIX: (#32950) Cancel of Hyper-V backup must remove VM snapshot. Problem
  • BUGFIX: (#32956) Adjust Hyper-V reference point files for DIFF and INC backups. Problem
  • BUGFIX: (#32970) Mount of vSphere saveset using URL encoded path names doesn't work. Do not URL encode '[' and ']' in vSphere data store names
  • BUGFIX: (#29404) Fixed unhandled exception if ISCSI client cannot be started
  • BUGFIX: (#32956) Adjust Hyper-V reference point files for DIFF and INC backups. Problem
  • BUGFIX: (#32970) Mount of vSphere saveset using URL encoded path names doesn't work
  • BUGFIX: (#32925) sm_restore calls sm_mount with wrong -c parameter. Set saveset parameter correctly again
  • BUGFIX: (#32954) Hyper-V RCT must be called with max defined virtual disk size. Problem
  • BUGFIX: (#32948) Create snapshot before reading RCT information of a Hyper-V VM. Problem
  • BUGFIX: (#32711) do not delete object if it currently checked by OCCK
  • BUGFIX: (#29404) Lock semaphore during snapshot creation and scan for new devices and during snapshot deletion. Restore and backup of config files was executed 2 times for VSphere due to copy paste error in sbc_pre function of sbc_proxy module VADP
  • BUGFIX: (#32945) Hyper-V DIFF backup cores due to extra ordinary RCT block range. Problem
  • BUGFIX: (#31877) All aborted restore tasks should have a corresponding message. Bugfix
  • BUGFIX: (#31877) All aborted restore tasks should have a corresponding message. If sub restore task could not be started then set state to 'X' and msg to reason
  • BUGFIX: (#29404) Fixed VMware backup using hardware snapshots for backup type vsphere+config
  • BUGFIX: (#32688) Fixed Incremental NDMP backup for Netapp 7. Incremental level was set higher than maximum level 9 for Netapp 7. Some instances of Netapp 7 have empty string for host_os_vers but this one has value set, so condition had to be fixed.
  • BUGFIX: (#4787) Better analyze of errors during backup. Bugfix
  • BUGFIX: (#32925) sm_restore calls sm_mount with wrong -c parameter
  • BUGFIX: (#32924) Backup of encrypted files does not backup ACL. Add NACL stream to backup
  • BUGFIX: (#32833) Make DB
  • BUGFIX: (#31866) do not start incr fsck from scratch in case of previous corruption state
  • BUGFIX: (#31866) change fsck check date counting
  • BUGFIX: (#32903) vSphere restore with high ascii characters fails with
  • BUGFIX: (#32906) vSphere backup fails, if the VM name contains string 'IMAP'
  • BUGFIX: (#31877) All aborted restore tasks should have a corresponding message. If drive_handler info/getlabel was aborted then state was set to 'X'. Now use TapeState = ST_INTERRUPTED to set correct state 'c'
  • BUGFIX: (#32694) Port SAP modules using Python3. Replace multiprocessing.Queue with queue.Queue
  • BUGFIX: (#32880) Unit license
  • BUGFIX: (#17310) New LIS file format to store extended file information. Add major minor os_specific info to Shared hardlink data file.
  • BUGFIX: (#25420) Purge info from vfs_mount.json if VFS has been already stopped or saveset ID not found in DB
  • BUGFIX: (#32866) Do not set '-a <recover>' parameter, if 'restore as path' option is set
  • BUGFIX: (#31463) Restore of a single file overwrites the parent directory. Windows
  • BUGFIX: (#32843) Use 'DB
  • BUGFIX: (#32705) Follow up event and saveset migration for external backups. Do not start migration by 'sm_copy' a 2nd time
  • BUGFIX: (#32858) Mount of Exchange DAG rpex fails with STATUS=ERROR MSG=name 'sm_Popen2' is not defined. Problem
  • BUGFIX: (#32842) Restore verify failed for cpio saveset
  • BUGFIX: Restore of MySQL 'all' backup
  • BUGFIX: Restore of MySQL backup into dump file
  • BUGFIX: (#31866) do not start incr fsck from scratch in case of previous corruption state
  • BUGFIX: (#31866) change fsck check date counting
  • BUGFIX: (#32903) vSphere restore with high ascii characters fails with
  • BUGFIX: (#32906) vSphere backup fails, if the VM name contains string 'IMAP'
  • BUGFIX: (#32694) Adapt 'SAP DB' backint for Python 3. Set loglevel as 'int'
  • BUGFIX: (#32860) restrict count of GC, FSCK workers for cloud connectors; extend FSCK description in Fsck.md documentation; print count of workers on PCCK and OCCK start; minor s3 log changes
  • BUGFIX: (#31877) All aborted restore tasks should have a corresponding message. If drive_handler info/getlabel was aborted then state was set to 'X'. Now use TapeState = ST_INTERRUPTED to set correct state 'c'
  • BUGFIX: (#32694) Adapt 'SAP DB' backint for Python 3. Shorten 'ssid', because PY3 'time.time()' function return longer string than with PY2
  • BUGFIX: (#17310) New LIS file format to store extended file information. The MTF SFMB, SSET, ESET, TAPE, VOLB, PNAM and DIRB do not contain data so set size as offset_to_data and data=0
  • BUGFIX: sm_Popen functions expect trace function as param but not sbctrace struct
  • BUGFIX: (#30917) Fixed API call to list snapshots. Removed wrong argument from GET request
  • BUGFIX: sm_Popen functions expect trace function as param but not sbctrace struct
  • BUGFIX: python2 modules have to copy libpython2.7.so.1.0
  • BUGFIX: copy sbc_find_imap sbc_imap_subject to bin dir
  • BUGFIX: python SCC modules rebuild each time
  • BUGFIX: SLX modules have to copy libpython2.7.so.1.0 to bin dir
  • BUGFIX: (#32694) Port SAP modules using Python3. Replace multiprocessing.Queue with queue.Queue
  • BUGFIX: (#32880) Unit license
  • BUGFIX: fix variable name
  • BUGFIX: (#17310) New LIS file format to store extended file information. Add major minor os_specific info to Shared hardlink data file.
  • BUGFIX: (#32860) restrict count of GC, FSCK workers for cloud connectors; extend FSCK description in Fsck.md documentation; print count of workers on PCCK and OCCK start; minor s3 log changes
  • BUGFIX: (#17310) New LIS file format to store extended file information. Add Windows major minor numbers for hardlinked files. Bugfix
  • BUGFIX: (#25420) Purge info from vfs_mouns.json if VFS has been alread stopped or saveset ID not found in DB
  • BUGFIX: SQL syntax errors in sesam_db_pg.sql
  • BUGFIX: (#32866) Do not set '-a <recover>' parameter, if 'restore as path' option is set
  • BUGFIX: (#31463) Restore of a single file overwrites the parent directory. Windows
  • BUGFIX: (#32843) Use 'DB
  • BUGFIX: (#32844) STPD processes are not terminated when read finished before end of data. Solution
  • BUGFIX: (#32841) FTP log file during backup contains misleading message according checksum calculation
  • BUGFIX: (#32847) Return OK when mounting to an alread existing mount
  • BUGFIX: (#32831) Update of SEP sesam on Windows could stuck in endless loop when Windows Event Log is opened during update. Solution
  • BUGFIX: (#32839) During remote update also move python311.dll to sesam_update folder to make sure that Windows Installer does not have files in use left in bin\sesam folder.
  • BUGFIX: (#32835) Only adjust access rights for installation type server and only adjust access rights of subfolder var\db_pg neither Program Files\SEPsesam\bin nor ProgramData\SEPsesam\var.
  • BUGFIX: (#32838) Fixed mount error if mount path set with option -M ends with '\'. Removed ending '\' during meta preparation process
  • BUGFIX: (#32381) Sesam using PostgreSQL DB on Windows doesn't start anymore, if Sesam service account has been changed. Problem
  • BUGFIX: (#32370) Search for 'sm_ui.jar' in process list instead of old 'sm_gui.jar'
  • BUGFIX: (#32819) Update to Apollon V2 failed with 'sm_setup.exe update -o dailog=... sm_setup.exe returned -1'. Solution
  • BUGFIX: (#29591) Common abort handling. Problem
  • BUGFIX: (#32370) Set timeout for hard kill by 'systemd' for Sesam service to 720 sec, because internal timout in 'sm_break' is 600 sec
  • BUGFIX: (#32563) SIR doesn't work because of dangling symbolic link. Problem
  • BUGFIX: (#29591) Common abort handling. Problem
  • BUGFIX: (#32807) sm_sms_watch process for drive 0 just checks free space when watch process is started. Problem
  • BUGFIX: (#32805) Role backup cannot be added to sm_java.policy via sm_setup. Added new permission de.sep.sesam.gui.server.BackupPermission in sm_java.policy
  • BUGFIX: Some localized messages with tags for sesam daily log did not start at position 14.
  • BUGFIX: (#32467) Windows backup fails with 'Not supported Microsoft reparse tag 0x00000036' (IO_REPARSE_TAG_CITRIX_PM). Add logic to GetStandardAttr4MTF().
  • BUGFIX: (#32758) Windows SBC skips first directory if nosbc file is in backup source
  • BUGFIX: (#32713) Migration of FDI chain with savesets on different mediapools and not sufficient EOL will repeat and never delete saveset on origin side. Solution
  • BUGFIX: (#32793) No incremental MS-SQL backup possible, if umlaut in DB name. Solution
  • BUGFIX: (#32781) Do not unmount saveset if guestmount returns error
  • BUGFIX: (#32568) Windows
  • BUGFIX: (#32776) First media initialization for a deduplication data store fails, if not performed on the first drive. Use first drive of data store for 'sm_dedup_interface init' call
  • BUGFIX: (#32690) Hyper-V DIFF/INC restore may not restore all data in case of dynamic disks.
  • BUGFIX: sbc_vmware core on nvram upload if error happen and get in retry loop
  • BUGFIX: Do not try to delete migration tasks, if still referenced by DB
  • BUGFIX: (#32688) NDMP backup switched to level 0 after every 2 due to missing {} in condition
  • BUGFIX: (#32495) sm_java forward SIGINT to java process in case of Si3-NG. Change Si3/Si3-NG detectionto original one
  • BUGFIX: (#31908) fix pages-index loading of zero page files
  • BUGFIX: (#32495) sm_java forward SIGINT to java process in case of Si3-NG
  • BUGFIX: (#32559) If VHD fails to be unmounted on Windows, continue with stop VFS because it causes no problems on Windows. After VFS is stopped, attached VHD drives automatically disappears from Windows list of disks
  • BUGFIX: (#32736) "sm_client drop" remove lis/<date> dir newer then 7 days
  • BUGFIX: (#31977) set GC threshold using start command; stop GC if no space available
  • BUGFIX: (#32495) add event listener on Windows for graceful server termination
  • BUGFIX: (#32220) Added signal handler to sbc_proxy RHEV module
  • BUGFIX: (#32730) Insert a retry of sm_main status after the complete update is done to make sure that sm_main status did not happen during stpd or sms restart. Also replace all & with && and all | with || if they are logical operators.
  • BUGFIX: (#32728) RHEV CBT backup ends with unhandled exception
  • BUGFIX: (#32702) Fixed mount of IMG vSphere backup. The issue was caused by fix for mount of a VM on DS with spaces in name
  • BUGFIX: (#32717) Wrong error message is displayed, if Sesam server is not allowed to connect to client
  • BUGFIX: (#26506) EOL adoption for external backups. In case of 'SAP ASE' the 'INCR' specifies an incremental backup since the last backup of any kind
  • BUGFIX: (#32706) Recover of LIS file from saveset fails because Sesam date based LIS directory doesn't exist. Create directory before calling 'COPY_REFORMAT_LIS()'
  • BUGFIX: (#32702) vSphere mount of a VM on DS with spaces in name fixed for IMG backup. Replaced spaces with underscores in LSL and MAP file.
  • BUGFIX: (#32702) vSphere mount of a VM on DS with spaces in name fixed for CBT backup. Replaced spaces with underscores in LSL and MAP file.
  • BUGFIX: (#30890) add missing license files.
  • BUGFIX: (#32701 Restore with Exchange Recovery Pro (mount) does not work for Exchange Server DAG
  • BUGFIX: (#32703) Removed option n from sm_mount help, not used
  • BUGFIX: (#32668) Error during VMware verify with follow up for Windows MTF saveset. Solution
  • BUGFIX: (#32685) Sesam server installation takes longer because SSH keys are tried to be downloaded, even if STPD isn't started yet. Solution
  • BUGFIX: (#32596) SAP IQ backup and restore via libsepsybase.so library. Problem
  • BUGFIX: (#32596) SAP IQ backup and restore via libsepsybase.so library. Problem
  • BUGFIX: (#32596) SAP IQ backup and restore via libsepsybase.so library. Problem
  • BUGFIX: (#16513) Archive adjustment with media introduction over barcode only. Problem
  • BUGFIX: (#31704) fix null pointer exception on PCCK when page meta is missing
  • BUGFIX: (#27672) Mount of BSR Pro image backups automatically as raw saveset
  • BUGFIX: (#32603) sm_mount returns successfully mounted drives in the returned status MSG in case of guest_mount returns with warning
  • BUGFIX: (#32644) - Citrix TLS CERT verification for sbc_xen build is set to CERT_REQUIRED, check_hostname and chippers are not set to ALL when there is a custom certificate set with "-a cert". Default certificate available in Citrix server does not meet security requirements of newer SSL library used in with Python 3.11 and change in NBD client code was made to disable certain requirements in order for backup to function with the new library. with this change, custom user certificate still must meet higher security standards
  • BUGFIX: (#32326) Correct handling of abort flag '-d -2' set by QM in 'sm_remote'
  • BUGFIX: (#31704) fix race condition when chunk was linked on swept page; fix bug when deep PCCK check for local pages was skipped
  • BUGFIX: (#31250) Current backup tape isn't used, if another one without tape trailer is loaded into drive. Do not call new tape initialization, but try to load requested tape
  • BUGFIX: (#26506) EOL adoption for external backups. Starting with backup 'SAP ASE' the level Full/Diff/Incr are now treated as backup chain. Diff will look for previous Full, Incr for previous Incr with same based_on_full
  • BUGFIX: (#32381) Sesam using PostgreSQL DB on Windows doesn't start anymore, if Sesam service account has been changed. Also during a new installation Sesam service does not start using a given user. Solution
  • BUGFIX: (#32672) Fixed mount of existing NDMP mount. Do not check for mounted partitions via sbc_proxy guest mount
  • BUGFIX: (#32667) NSS volumes are not select able in UI during restore using VMDK attach. Return "/mnt" for NSS volumes. Execute sbc_vadp_nss_post.sh during detach. Removed cleanup on non existing meta files to avoid unnecessary error log
  • BUGFIX: (#32605) Hyper-V RCT backup accidentally destroys all user snapshots.
  • BUGFIX: (#32256) Show all MS-SQL databases on one server by one 'dir' command. Problem
  • BUGFIX: (#31781) SEP sesam Unit License. Problem
  • BUGFIX: (#32652) Aborted tasks have no DB
  • BUGFIX: (#32524) Handling of the follow up events in case of an abort. Problem
  • BUGFIX: (#32657) Option to avoid starting follow up event by QM. New command 'sm_qm set option follow_up' to display follow up setting
  • BUGFIX: (#30358) RHEV backup cannot be cancelled. Terminate VHD data backup first
  • BUGFIX: (#32648) MSP license must not add BSR licenses to Level 1
  • BUGFIX: (#32654) Do not search for restore_id in vfs_mounts.json during mount or attach. Log restore_id only if it was found in json during umount or detach
  • BUGFIX: (#29199) To ensure that a pending reboot of Windows does not affect the SEP sesam update the function CheckRebootRequired() will be executed before the actual update starts informing the user as early as possible about a pending reboot.
  • BUGFIX: (#32626) SAP IQ
  • BUGFIX: (#32645) Excute "readlink -f" during guest umount on Linux. Replace mount dir with its symlink during check for mounted partition, if symlink exists
  • BUGFIX: (#32601) A large Log LSN number (>64Bit) leads to a wrong result when comparing the Log first LSN with the Full last LSN
  • BUGFIX: (#32381) Sesam using PostgreSQL DB on Windows doesn't start anymore, if Sesam service account has been changed. Solution
  • BUGFIX: (#32615) enable RCT backup with configuration version 10
  • BUGFIX: (#26809) Validate of ORACLE multichannel backup. Linux
  • BUGFIX: (#32629) vSphere SFR via cross mount fails with server on Windows and RDS on Linux. Fixed path separator'
  • BUGFIX: Submit 'sm_sshd' by sm_main with start time 'now'
  • BUGFIX: (#32619) Nutanix snap_del variable was checked too early, before it was set by the main sbc_proxy function
  • BUGFIX: (#32617) Proxmox restore fails, because 'sm_restore' tries to restore virtual disk saveset. Add PROXMOX_VE and OpenNebula to list of backup types, where no FDI chain will be fetched
  • BUGFIX: (#32565) Fixed mount from Linux to Windows client. Mount path send to sbc_proxy must be formatted with separator "/"
  • BUGFIX: (#25420) Check if disk partitions is mounted during umount, if it is not mounted, continue with stopping VFS
  • BUGFIX: (#32609) sbc_rhev
  • BUGFIX: (#25420) Fixed setting of -K binary in sm_qm. Binary path is determined by reading of gv_ro_bin value on the mount client if client is set in QM
  • BUGFIX: (#25420) Do not set stop executable during QM submit, if no end time is given
  • BUGFIX: (#32596) SAP IQ backup and restore via libsepsybase.so library. Solution
  • BUGFIX: (#32619) Nutanix snap_del variable was checked too early, before it was set by the main sbc_proxy function
  • BUGFIX: (#32270) Hyper-V RCT INC/DIFF backup is much slower than FULL. Problem
  • BUGFIX: (#32596) SAP IQ backup and restore via libsepsybase.so library. Problem
  • BUGFIX: (#32609) sbc_rhev
  • BUGFIX: (#25420) Fixed check for VFS mount client to determine if sm_mount or sbc_proxy is to be set as sm_qm -K argument
  • BUGFIX: (#30977) Set DB
  • BUGFIX: (#32542) All replication jobs are set to 'failed', if one saveset cannot be found on source media pool
  • BUGFIX: (#32591) added "-n node.startup -v manual" to iscsi Linux client login command
  • BUGFIX: (#32597) Migration does not work for savesets with DB
  • BUGFIX: (#32589) Update 'DB
  • BUGFIX: (#32592) Forward error message to Sesam server for external backups using XBSA. Solution
  • BUGFIX: (#31869) Break during SEP sesam BSR Pro (BSR) Windows backup. Check for STPD abort message '520 Operation aborted by Sesam Server'
  • BUGFIX: (#32553) Abort of external backups. If STPD message '520 Operation aborted by Sesam Server' appear in LIS then set state to '3' aborted
  • BUGFIX: (#32456) Canceled Microsoft Exchange backup before data transfer started could fail with 'Subprocess sm_sms_backup undefined broken'. Solution
  • BUGFIX: (#32522) gc stuck if unexpected exception throws during page content loading
  • BUGFIX: (#8554) Execute hard termination not during first loop
  • BUGFIX: (#30613) Restore does not work, because sesam does not find the volume/directory in the lis file. Problem
  • BUGFIX: (#32209) Race condition reading 'sm.ini' on Windows. Problem
  • BUGFIX: (#32432) Restore of big sparse file ~ 3GB will fail. Code revised. Even in case MTF_NO_SPAD the PDS_I_PROCESSED_FILE must be signalized.
  • BUGFIX: (#32408) Backup of open sparse files without vss will not process all data. Open with GENERIC_READ, FILE_SHARE_READ/WRITE to allow accessing a file which is opened by another process. Additional
  • BUGFIX: (#30241) The original error message is overwritten by 'No segment file {...}.dat found'. Solution
  • BUGFIX: (#32528) Update date in 'sm.ini' not set correctly during Linux Sesam server update. Fix parsing of token in parser
  • BUGFIX: (#32577) Do not remove DB entry and meta files of saveset, if delete of saveset file fails. Scan for 'SUCCESS' message instead of 'ERROR'
  • BUGFIX: (#32348) Verify saveset with VMware backup does not work. Solution
  • BUGFIX: (#32573) Abort of restores. Problem
  • BUGFIX: (#32565) Fixed setting of mount path when -M is not set and gv_rw_mnt is not set. Mount path in that case is set at gv_rw_tmp/mnt
  • BUGFIX: (#32578) Sesam service doesn't start, if STPD INI files couldn't be read by Python config parser. Ignore exception and continue with next file
  • BUGFIX: (#32565) Added processing of -M argument for sbc_proxy mount/umount actions. sm_mount sets -M for sbc_proxy. if -M is not set during umount, sm_mount reads it from mounts.json file. arg.mntdir is set only in ARG at one place. removed function get_mnt_dir() from sm_vfs.py, it was redundant functionality
  • BUGFIX: (#32557) Fix setting of image transfer timeout via parameter
  • BUGFIX: (#32557) OLVM/RHEV backup fails due to imageservice timeout
  • BUGFIX: (#32557) VM is locked in backup state if transfer ticket expired
  • BUGFIX: (#32570) sbc cores during restore to a different Nutanix cluster. Init variable
  • BUGFIX: (#32545) sm_alarm interface couldn't be executed, if backup has been aborted.
  • BUGFIX: (#32568) sm_alarm interface could not be executed if the attached file has a path delimiter '\'
  • BUGFIX: (#31872) Set abort status when the corresponding restore task is completely finished. Solution
  • BUGFIX: (#32559) Check return from umount command in sbc_proxy diskpart for Windows and do not stop VFS process in sm_mount if sbc_proxy returns error
  • BUGFIX: (#31856) Selective generation restore ends with warnings due to existing directories. Problem
  • BUGFIX: (#32557) VM is locked in backup state if transfer ticket expired
  • BUGFIX: (#32557) Fix setting of image transfer timeout via parameter
  • BUGFIX: (#32557) OLVM/RHEV backup fails due to imageservice timeout
  • BUGFIX: (#32325) Disconnect VDDK connections correctly during termination of vSphere backup. Problem
  • BUGFIX: (#32262) Break of hanging backups using 'sbc_smdr' fails. Extend search pattern to find 'sbc_smdr' processes too
  • BUGFIX: (#32543) Upload of vSphere '.nvram' fails, if datacenter contains UTF-8 character. URL encode datacenter name set by UI too
  • BUGFIX: (#31636) Abort of vSphere restores doesn't find processes to send signal to them. Use DB
  • BUGFIX: (#30696) Break of Nutanix backup/restore doesn't kill STPD/SBC of VHD saveset
  • BUGFIX: (#29404) open/close pvim connection for every virtual disk to fix consolidation issue
  • BUGFIX: (#32512) Restored VMs with new name have the same GUID as the original VM
  • BUGFIX: (#32141) Abort of SAP Hana backup does not work as expected. Warning resolved
  • BUGFIX: (#32488) Alarm interface fails with NPE up from 5.1 when sending mail also in case the start time frame expired since no attachment is available. Solution
  • BUGFIX: (#32141) Abort of SAP Hana backup does not work as expected. Remove of orphaned 'sm_sbc_com_ext' job
  • BUGFIX: (#25229) sbc must not print "operation was OK" when error occurs during data transfer
  • BUGFIX: (#32491) If arguments are no valid, exit with error "Invalid arguments". If backup_type is not valid, exit with error "%s is not a valid backup type"
  • BUGFIX: (#32534) Wrong relocation parameter for VMDK restore set. Problem
  • CHANGE: (#17310) Use new LIS file format (LISv2) as default
  • CHANGE: (#30633) Implement logic for changing logs retention time exactly in places where server sanity status is changed (FSCK, page&obj corruptions stores); delete previous logic with invoking changing on "status" command
  • CHANGE: (#32482) Migrate a saveset to same media pool one time only as default
  • CHANGE: GLBV gv_conf_pattern_lis_v1/gv_conf_pattern_convert_lis_v2_to_v1
  • CHANGE: (#31800) Update backup source of Sesam backup after migration of Sesam DB from Sqlite to PostgreSQL has been done
  • CHANGE: 'sbc-3223' message for 'Filename changed to..'
  • CHANGE: sbc_rhev
  • CHANGE: Log information about saveset read in VFS with default log level
  • CHANGE: (#33034) Log rename of system libraries during Linux file restore to log loglevel 2
  • CHANGE: (#33021) trace logs with level 1 in VFS in case of errors and exceptions
  • CHANGE: (#32464) Migrate a saveset only once to same target pool
  • CHANGE: Allow chained Sesam events on Windows too. Fix quoting of command execution
  • CHANGE: (#33009) Added trace include and use in VFS RCT
  • CHANGE: Warning
  • CHANGE: (#32983) Changed snapshot names for HW snapshot backups
  • CHANGE: (#32999) Avoid NACL streams in Hyper-V RCT range files
  • CHANGE: Simpler building of SQL statement for creation of SESAM_BACKUP
  • CHANGE: (#31800) Improve Sesam backup task handling. Save 'gv_rw_ini
  • CHANGE: (#31800) Improve Sesam backup task handling. Save 'gv_rw_ini
  • CHANGE: (#31800) Improve Sesam backup task handling. Do not resolve Sesam 'glbv' in 'sm_backup', but forward them to 'sbc'
  • CHANGE: (#31800) Execute Sesam backup as 'DIFF' instead of 'INCR' to allow complete Sesam restore by 2 backups
  • CHANGE: (#32992) Add new 'gv_rw_bin' directory used store custom script files to Sesam backup source
  • CHANGE: (#31936) Use default ISO date format for DB
  • CHANGE: (#31936) Use default ISO date format for DB
  • CHANGE: (#17013) Set 'DB
  • CHANGE: (#32980) Use ENV variable 'SM_INI' instead of 'SESAM_INI' to change path of default '/etc/sesam2000.ini', because 'SM_INI' is used already in many Sesam modules for a longer time
  • CHANGE: (#32983) Changed snapshot names for Nutanix and fixed delete of snapshots on XEN
  • CHANGE: Write import of saveset message correctly to Sesam status file
  • CHANGE: Set 'fsck' status to OK for read only data stores
  • CHANGE: (#32983) Common naming convention for Sesam snapshots in XEN module
  • CHANGE: (#29404) changed log prefix for Pure and Nimble to sos
  • CHANGE: Log sbc 'since' time in ISO timestamp format
  • CHANGE: (#31936) Use default ISO date format for DB
  • CHANGE: Use 'psz_now_iso()' function to get current time
  • CHANGE: (#29404) Added log about lvid of the new device we search for after snapshot
  • CHANGE: (#31681) SIR doesn't work if 'guestmount' will be started by 'sm_sshd'. Solution
  • CHANGE: (#17013) Start vSphere backups with URL encoded source as default
  • CHANGE: (#32500) change documentation related to custom logback file and XML variables
  • CHANGE: (#32500) Disable colourized log output for Si3NG. Add user defined XML file so switch colorized logs again
  • CHANGE: (#32970) Added option -o uridecode=0|1 If set to 1, vmdk file names are uri decoded. Default value is uridecode=0
  • CHANGE: (#32970) Replace %20 in VMDK names with _
  • CHANGE: (#32970) Replace of VMDK names for LIS mapping done in one place, added new list vmdk_list to hold vmdk names with spaces and ()][/\ chars replaced.
  • CHANGE: (#29404) snapshot list prints all attributes of Nimble snapshots
  • CHANGE: Revise cm hal pipe logging (it should be possible to filter out read messages)
  • CHANGE: (#32875) Simplified sbc_proxy backup sbc_pre functions for all modules
  • CHANGE: Set 'DB
  • CHANGE: Put 'sm_sms_tapeutil' into 'bin/sesam' to avoid including 'python3' folder 2 times in Sesam kits
  • CHANGE: (#32694) Remove sbc_mysql and sbc_pgsql executables from Sesam build. Function has been moved into sbc_proxy
  • CHANGE: (#32694) Remove sbc_mysql and sbc_pgsql executables from Sesam build. Remove sbc_zarafa
  • CHANGE: Remove backup types 'Open-Exchange Server', 'OpenLDAP' and 'Scalix' from Sesam
  • CHANGE: (#32863) Build vSphere module using VDDK 8.0.2 as default
  • CHANGE: (#31950) Changed prefix for Oceanstor API logs
  • CHANGE: (#32823) Use DB
  • CHANGE: (#32807) sm_sms_watch process for drive 0 just checks free space when watch process is started. Change
  • CHANGE: Disable 'get_cpu_info()' function call, until core dump on Debian has been fixed
  • CHANGE: (#32807) sm_sms_watch process for drive 0 just checks free space when watch process is started. Change
  • CHANGE: (#32694) Remove Scalix, OpenXchange and LDAP modules
  • CHANGE: (#30917) Modified sbctrace to automatically skip 4000 logs from printing if loglevel is less than 4. Set loglevel and tested in sbc_proxy OCEANSTOR
  • CHANGE: (#32823) Use DB
  • CHANGE: Remove backup types 'Open-Exchange Server', 'OpenLDAP' and 'Scalix' from Sesam
  • CHANGE: remove outdated code
  • CHANGE: (#32863) Build vSphere module using VDDK 8.0.2 as default
  • CHANGE: use own functions to detect CPU info
  • CHANGE: (#31950) Changed prefix for Oceanstor API logs
  • CHANGE: Python 3.11.4 -> 3.11.6
  • CHANGE: remove api-ms-win-core-*.dll again from build. It is not needed anymore.
  • CHANGE: (#32823) Use DB
  • CHANGE: (#32807) sm_sms_watch process for drive 0 just checks free space when watch process is started. Change
  • CHANGE: Disable 'get_cpu_info()' function call, until core dump on Debian has been fixed
  • CHANGE: (#32876) Sesam 'dev' using OpenSSL 3. Build Linux PPC kits with SSL 111, because of OpenSSL 3 doesn't work correctly
  • CHANGE: Disable 'get_cpu_info()' function call, until core dump on Debian has been fixed
  • CHANGE: (#32807) sm_sms_watch process for drive 0 just checks free space when watch process is started. Change
  • CHANGE: (#32694) Migrate sbc_vadp modules to Python 3
  • CHANGE: (#32876) Build Sesam using OpenSSL 3
  • CHANGE: Use Sesam 'pathes' from 'sm_common3.py'
  • CHANGE: (#32694) Port SAP modules using Python3
  • CHANGE: build using OpenSSL 3.2.0 (/usr/local/sesam-ssl-320-develop)
  • CHANGE: set correct cx_freeze command for all sesamdev versions different then 102
  • CHANGE: Remove some unused code
  • CHANGE:check of target DB for all PostgreSQL backup during restore
  • CHANGE: (#30917) Modified sbctrace to automatically skip 4000 logs from printing if loglevel is less than 4. Set loglevel and tested in sbc_proxy OCEANSTOR
  • CHANGE: remove trailing whitespace
  • CHANGE: (#32823) Use DB
  • CHANGE: (#32840) Use LISv1 for savesets written with SIDF format
  • CHANGE: (#23784) Saveset migration without 'DB
  • CHANGE: (#31049) Enhance backup chain check for MSSQL log backup. Change
  • CHANGE: (#23477) New license for MSP. Change
  • CHANGE: (#32769) vSphere restore fails with new VM name and white space in VMDK name
  • CHANGE: (#23477) New license for MSP. Change
  • CHANGE: (#4787) Better analyze of errors during backup. Treat message 'sbc-2010 Unable to access item ...' as error message. Otherwise this message is overwritten by 'Error from sbc find proccess'.
  • CHANGE: (#32603) fixed guest mount so that search for new attached volumes is not repeated more times than needed
  • CHANGE: Start Si3-NG server by 'sm_java ... start sds2'
  • CHANGE: (#24200) sbc_proxy checks for correct format of Sesam gv_rw_tmp directory, it must end with "/var/tmp/" and if it is not the case, sbc_proxy exits with error
  • CHANGE: (#32603) Simplifed code to set VFS mount client. Use arg.mount_client in sm_vfs.py
  • CHANGE: (#32710) Support for 'arm64' CPU architecture
  • CHANGE: (#27781) Write number of removed saveset to DB
  • CHANGE: (#32727) Write main media initialization log file into 'gv_rw_prot_media
  • CHANGE: sbc_rhev
  • CHANGE: (#26506) EOL adoption for external backups. Starting with backup 'SAP ASE' the level Full/Diff/Incr are now treated as backup chain. Change
  • CHANGE: (#32495) Graceful shutdown of Si3-NG server. Minor changes in sm_java
  • CHANGE: (#30890) - Review and update Java part 3rd party licenses
  • CHANGE: (#32318) Save 'sbc_pre/post' scripts in 'gv_rw_bin' for clients with version >= 5.1.0.9. If script already exists in 'gv_ro_bin' it will stay in this directory
  • CHANGE: (#29346) Correct error message, if processiing of LIS or SGM files failes for external backups
  • CHANGE: (#32618) Backup fails with misleading error message if segment file missing. Use 'E019_BACKUP_ERROR_PARSE_SGM' in case of no SGM file found too
  • CHANGE: (#32618) Backup fails with misleading error message if segment file missing. Do not compare backup size and items processed between SBC and LIS, if copy of SGM or LIS file has already failed
  • CHANGE: (#17318) Set the checkbox for the http link for first steps active only during installation not during update.
  • CHANGE: (#32627) SAP IQ INCREMENTAL backup. SAP IQ uses command_type 3 (UNKNOWN) in case of BACKUP DATABASE INCREMENTAL
  • CHANGE: (#32524) Handling of the follow up events in case of an abort. Enable follow up execution again, if not called for complete Sesam shutdown
  • CHANGE: (#25420) Add info about sbc_proxy mount client as -C argument to sm_vfs
  • CHANGE: (#32657) Option to avoid starting follow up event by QM. Change CLI to 'sm_qm option follow_up <0|1>'
  • CHANGE: Added missing info to sm_mount help
  • CHANGE: (#17318) In the last MSI installer windows set the checkbox for the http link for first steps active. Until now the user had to check the checkbox himself.
  • CHANGE: (#32621) Make sure that leftover unneeded files are removed explicitely during update process
  • CHANGE: (#32642) mount path for all modules VM disk partitions and for PATH is set to <VFS mount path>/fs/<saveset ID>
  • CHANGE: (#32589) Abort of restores using saveset mount. Add saveset id in pipe name used to copy files over STPD instead of restore task, so 'sm_kill_stpd' will find the correct process
  • CHANGE: (#32573) Set DB
  • CHANGE: Use 'DB
  • CHANGE: Allow DB
  • CHANGE: (#32573) Abort of restores. Set status to '3' instead of 'c', if sbc has already been started before abort action
  • CHANGE: (#28353) Continue selective restore when ESET is missing. Restore should be marked as "successful with warnings"
  • CHANGE: (#32537) Change Java dependency for Sesam server on SLES12 from 'jre >= 11' to 'java-11-openjdk', because recent Java RPMs from SuSE doesn't provide 'jre >= 11' tag any more
  • CHANGE: (#32348) Verify saveset with VMware backup does not work. Solution
  • CHANGE: (#29591) Changed log in sbc_proxy when termination signal is received to sbc-1028
  • CHANGE: (#28009) Adjust sm.ini including logical block size if a GUI with a client is installed. Only for a pure GUI installation these adjustement are skipped.
  • CHANGE: (#32370) Improve logic of sm_break2 when active jobs are found. Change
  • CHANGE: (#32188) Do not use TR() function in signal handler. Use "safe" i_cm_signal function by default where "action" is wrapped with thread.
  • CHANGE: (#25420) Added quotes to all optional arguments in QM submit and to VFS arguments in sm_mount
  • CHANGE: (#25420) Addopted sm_qm.py and submit sm_vfs with kill_wait_time=600, stop_proc="sm_stop"
  • CHANGE: (#29840) added help output to main sbc_proxy
  • CHANGE: (#32370) Improve logic of sm_break2 when active jobs are found. Change
  • CHANGE: (#32370) Improve logic of sm_break2 when active jobs are found. Change
  • NEW: Update default VDDK version 7.0.0 to 8.0.2
  • NEW: Switch to OpenSSL 3.2.0
  • NEW: Sesam release 5.1.0.14
  • NEW: (#32562) Do not allow use of SEP SiS Immutable Storage data stores with SEP Sesam essential license
  • NEW: (#33053) Support Windows Server 2025. Add oper_systems and syntaxes entries for 'Windows Server 2025'
  • NEW: add sbc-3277 with sbc submodule versions used in case of -o shortlog specified
  • NEW:(#33043) opennebula
  • NEW: (#17310) New LIS file format LIS 2.0 to store extended file information. New
  • NEW: (#30633) Increase server logs retention time when server sanity status becomes NOT OK; set back (with delay in 24 hours) original retention time when server sanity status becomes OK again.
  • NEW: (#28448) run sm_rexec directly over smssh if available without calling to sm_rexec. Use direct ssh calls only if DS configured with options 'use_direct_ssh=yes|y|1'
  • NEW: Do not purge logfiles, DB entries and Si3 trash, if 'gv_conf_no_purge_all=1' is set
  • NEW: (#33022) Show processing of ACL/EA items during restore
  • NEW: (#28448) run sm_rexec directly over smssh if available without calling to sm_rexec.
  • NEW: Add PPID to 'sm_client' log
  • NEW: (#31933) Set custom pattern for detecting Sesam backup task by 'gv_conf_sesam_backup_pattern=<regex>'
  • NEW: (#21538) Check if PostgreSQL DB runs in archive mode before starting incr. backup
  • NEW: Support backup/restore of files with \n and/or \r in path (-o encode_source=0|1)
  • NEW: (#25176) Adapt backup source of Sesam backup during update. Try to update full path names used in older Sesam installations too
  • NEW: (#25176) Adapt backup source of Sesam backup during update. Detect source from older Sesam installations
  • NEW: (#25176) Adapt backup source of Sesam backup during update
  • NEW: (#32980) Installation of Sesam client as non root user
  • NEW: (#28448) run sm_rexec directly over smssh if available without calling to sm_rexec. Use direct sm_ssh calls for DI
  • NEW: (#32944) Write output from PRE/POST scripts into backup or restore log directly
  • NEW: (#27274) Set of source side deduplication drive for 'sbc -b -o com'. Add example in documentation
  • NEW: (#19569) Set backup data mover by task group (DB
  • NEW: (#30267) Set datamover in DB
  • NEW: (#32500) add implementation for logback XML variables, and it's merging with custom file
  • NEW: (#24208) Mount BTRFS root volume on mount point <mnt>/disk<disk num>/root and all subvolumes on <mnt>/disk<disk num>/subvolumes. Moved code to execute guest_mount to a separate function
  • NEW: (#31504) Do not overwrite (changed) sm_sdslog.xml / sm_sdslog2.xml during sm_config_drives. Problem
  • NEW: (#32639) Source side deduplication for HPE Catalyst
  • NEW: (#20346) Source side de-duplication for external backups
  • NEW: Use operating system user 'postgres' as default for 'PostgreSQL' backup, because most servers are installed using it. Can be overwritten by '-a osuser=<user>' option
  • NEW: (#32705) Follow up and migration event for Oracle backup
  • NEW: (#26482) Support of 'SESAM_TASKEVENT' parameter in Oracle backup module
  • NEW: insert restore pre and post files for powershell.
  • NEW: (#32863) Add VDDK VMware-vix-disklib-8.0.2.x86_64.zip to installer skel directory.
  • NEW: Detect running 'MariaDB' server, which is started as 'mariadbd' instead of 'mysqld'
  • NEW: (#21538) PostgreSQL FULL/DIFF/INCR backup. Restore backups as 'Path'
  • NEW: (#31950) Implemented VMWare backup using Huawei OceanStor hardware snapshots
  • NEW: (#30917) after snapshot is created, move it to the same LUN group as the source LUN is in. If the client has configured access for that LUN group, snapshot becomes automatically visible as a new ISCSI disk
  • NEW: (#32833) Make DB
  • NEW: (#21538) PostgreSQL
  • NEW: (#30917) Implemented deleting of Oceanstor snapshots and some code optimizations
  • NEW: (#32863) Use correct sbc for vSphere backup using VDDK >= 8.0.1
  • NEW: (#30917) fixed dir snapshots to print snapshots of LUNs instead of FS snapshots
  • NEW: (#30917) Added function to create Oceanstor snapshots and fixed dir snapshots to print snapshots of LUNs instead of FS snapshots
  • NEW: (#30917) Implemented dir function to print Oceanstor FS snapshots
  • NEW: (#27064) Clarify information which block size is used during write/read to tape in sms loggings. Now with read operation trace
  • NEW: (#32852) Remove SQLite '\var\db' directory after migration of Sesam DB to PostgreSQL
  • NEW: add site.conf template for BSR Linux on multipath systems
  • NEW: (#30917) implemnted DIR function to print Oceanstor file systems
  • NEW: (#31463) Restore of a single file overwrites the parent directory. Since SEPsesam version 5.1.0.10 the SBC option '-o over=file' is used to avoid overwriting directories.
  • NEW: (#32718) Use Python 'clicks' library as replacement for 'getopt/args'
  • NEW: (#32705) Follow up event and saveset migration for external backups
  • NEW: (#21538) PostgreSQL FULL/DIFF/INCR backup. Restore backups as 'Path'
  • NEW: (#32833) Make DB
  • NEW: (#31950) Implemented VMWare backup using Huawei OceanStor hardware snapshots
  • NEW: (#30917) after snapshot is created, move it to the same LUN group as the source LUN is in. If the client has configured access for that LUN group, snapshot becomes automatically visible as a new ISCSI disk
  • NEW: (#32833) Make DB
  • NEW: port SAP modules from python2 to python3. Add new dll needed to VS2019 section. api_ms_win_core_console_l1_2_0.dll.
  • NEW: port SAP modules from python2 to python3. Add new dll needed. api_ms_win_core_console_l1_2_0.dll.
  • NEW: (#30917) Implemented deleting of Oceanstor snapshots and some code optimizations
  • NEW: (#32863) Use correct sbc for vSphere backup using VDDK >= 8.0.1
  • NEW: (#30917) fixed dir snapshots to print snapshots of LUNs instead of FS snapshots
  • NEW: (#30917) Added function to create Oceanstor snapshots and fixed dir snapshots to print snapshots of LUNs instead of FS snapshots
  • NEW: (#30917) Implemented dir function to print Oceanstor FS snapshots
  • NEW: (#27064) Clarify information which block size is used during write/read to tape in sms loggings. Now with read operation trace: 'serv_get_tape_label:: label = %s, blocksize %lld, date %s, info [%s]' Change check of target DB for all PostgreSQL backup during restore
  • NEW: (#32852) Remove SQLite '\var\db' directory after migration of Sesam DB to PostgreSQL
  • NEW: add site.conf template for BSR Linux on multipath systems
  • NEW: (#32833) Make DB
  • NEW: (#30917) implemnted DIR function to print Oceanstor file systems
  • NEW: (#31463) Restore of a single file overwrites the parent directory. Since SEPsesam version 5.1.0.10 the SBC option '-o over=file' is used to avoid overwriting directories.
  • NEW: Increase SEP sesam version. Old version is 5.1.0.11, new version is 5.1.0.12.
  • NEW: (#30917) sbc_proxy login to HUAWEI REST API, gets system info and logout
  • NEW: (#31025) sbc_proxy guest mount on Windows mounts volumes of mounted virtual disks to sub folder instead of adding new drive letters to the system. Mount directory has format
  • NEW: Use https for Desktop links to WebUI and Restore Assistant created by installation.
  • NEW: (#31950) Implemented login to OceanStor REST API
  • NEW: (#32796) Support for Ubuntu 23.10
  • NEW: Reduce number of SQL statements executed by 'sm_pur_status' to remove DB
  • NEW: (#32782) Use common message format definition for all modules writing to backup log. New function 'trkv(msgnr,key,value)' to print out aligned message
  • NEW: (#13406) Enhance backup speed of DIFF/INC backup by checking modification time by 'smfind'
  • NEW: (#32779) Call 'sm_alarm/sm_notify' after execution of command event. Added into templates for Windows sm_alarm.cmd/ps1
  • NEW: (#32779) Call 'sm_alarm/sm_notify' after execution of command event. Added into templates for Windows sm_notify.cmd/ps1
  • NEW: (#32779) Call 'sm_alarm/sm_notify' after execution of command event
  • NEW: Restart Si3 server, if it exit with exit code '1'
  • NEW: (#32688) If FDI chain file is not found during NDMP incremental backup, switch to FULL backup
  • NEW: (#32688) Set NDMP backup max level to 31 for Netapp 9 or higher, 10 for EMC unity and to 9 for all others
  • NEW: Add maximum size liit during creation of Si3 deduplication of 40 TiB
  • NEW: (#32688) Check for sbc-2988 message, if NDMP backup has been switched from INCR to FULL by sbc
  • NEW: (#32688) NDMP backup switched to level FULL if there are 8 incremental backups already in the chain. Added new warning message 2988 to sbc_msg.c
  • NEW: (#32745) No throughput values are written to DB after replication to SIS appliance has finished. Read value by 'sm_dedup_interface' instead of log file
  • NEW: (#32736) Replace GNU findutils in sbc UNIX
  • NEW: (#32755) add additional storage information during browse of rhev storage volumes
  • NEW: (#32710) sbc_proxy prints CPU details and starts every network log with a key word Network
  • NEW: (#32720) Copy LIS and STPD/SMS/SI3 logfiles from RDS by STPD, "sbc -h INFO
  • NEW: (#27781) Remove old log files from 'prot/media'. Add new action 'sm_cross_check cleanup_media_log'
  • NEW: (#32710) sbc_proxy print all network info. Fixed formatting of memory info
  • NEW: (#32710) Added sbc_proxy print info for memory, CPU and network
  • NEW: (#32721) Remove media log files for deleted data stores
  • NEW: (#31950) HTTP client session for OceanStor connect to server
  • NEW: (#32710) Added AV info on Windows
  • NEW: (#32586) Set number of parallel processes used during data store purge
  • NEW: Support '-u <username>' option in 'sm_remove_saveset'
  • NEW: (#32318) Store custom Sesam scripts in extra directory
  • NEW: (#32586) Set number of parallel processes used during data store purge
  • NEW: (#32318) Add 'gv_rw_bin' to PATH variable in Sesam profile
  • NEW: (#30821) Support of LTO9 first time media initialization. Write to sm_slu_{day}.log if SM_SLU_LOGFILE is set in the process environment.
  • NEW: (#29307) Prevent expansion of GroupWise objects in client browser
  • NEW: (#32318) Allow execution of command events in path 'gv_rw_bin' too
  • NEW: (#32318) Search scripts like 'sm_notify, sbc_pre...' in 'gv_rw_bin' before 'gv_ro_bin'
  • NEW: Exit 'sm_break2' with 'STATUS= MSG=' line
  • NEW: (#32680) Exit Huawei NDMP backup with error for level COPY. Added logs with info about transferred backup data size
  • NEW: (#32618) New localized error message for problems during LIS file parsing. Do not use backup status '2' (no LIS), because even FULL restore might not work correctly
  • NEW: (#32680) Add OS 'Huawei OceanStor' with platform 'NDMP'
  • NEW: (#30890) Add MIT license authors and years for Kernel modules.
  • NEW: (#30890) Add WebUI specific third party licenses to git and to file _list_third_party_licenses.txt.
  • NEW: (#32746) Check validity of Sesam server certificate by client. sm_cmd using Url
  • NEW: (#32688) NDMP backup writes incremental level to a file in times for every backup task. For every full backup, value is set to 0 and for 1st INC backup it is set to 2. For every next INC backup, value is increased by 1 until max value of 9 is reached. If value is 9 and INC backup is started, it exits with error. The same value is set to NDMP ENV variable level, for INC backup
  • NEW: (#26678) Add information about media and media pools on a data store to a META file in JSON formt
  • NEW: (#32667) Added -o show_parent_mnt to sm_mount help
  • NEW: Allow use of '-c' option for action specific arguments in 'sm_event'
  • NEW: (#32348) Execute 'sbc' for verify saveset on data mover for clients with access mode 'PROXY'
  • NEW: (#25420) when PATH mount on RDS and qm_stop action executes sbc_proxy, _mounts file dooes not exist and sbc_proxy executes stop_vfs. Information about this is added to a file log/lgc/sm_mount-<ssid>.log
  • NEW: (#32524) Handling of the follow up events in case of an abort. Enable follow up execution again, if not called for complete Sesam shutdown
  • NEW: (#32524) Handling of the follow up events in case of an abort. Do not execute any follow up event, in case of Sesam server shutdown
  • NEW: (#32657) Option to avoid starting follow up event by QM. New command 'sm_qm set option follow_up=0/1' allows to switch off/on the follow up processing
  • NEW: (#27376) Purge old SMS/STPD logfiles recursively from remote device server
  • NEW: (#27376) Put STPD logs into date based subdirectories. Problem
  • NEW: (#31560) Add all missing third party licenses used by Sesam to 'skel' folder and adjust some of the exiting ones.
  • NEW: (#32631) sbctrace prints time with milliseconds if "gv_log_milliseconds=YES" is set in sm.ini
  • NEW: (#26680) Enhance message when active/queued job is canceled for any reasons
  • NEW: (#32585) Added option -o no_lisv1=1 to sm_mount to stop mounting if no LISv2 file found
  • NEW: (#32611) If more than one VMs with the same name exist during backup, backup exits with error. If more than one proxy VMs with the same name exist during hotadd backup or restore, operation exists with error
  • NEW: (#32573) Set default wait timeout for hard kill to 10 min
  • NEW: (#15302) Execute abort action in 'sm_break_task' directly again and do not call 'sm_stop' for it. By calling sm_stop control is lost by caller
  • NEW: (#28862) Cleanup savesets on data store, which have zero data size in DB
  • NEW: (#25420) Added functionality to execute detach during sm_mount QM stop. mounts.json file is searched for the command based on given SSID and action is determined from the command found there
  • NEW: (#32611) If more than one VMs with the same name exist during backup, backup exits with error. If more than one proxy VMs with the same name exist during hotadd backup or restore, operation exists with error
  • NEW: (#32573) Set default wait timeout for hard kill to 10 min
  • NEW: (#15302) Execute abort action in 'sm_break_task' directly again and do not call 'sm_stop' for it. By calling sm_stop control is lost by caller
  • NEW: (#31832) Break during SAP Hana backup. Set status '3' instead of 'X' if SAP Hana backup is aborted
  • NEW: (#31832) Break during SAP Hana restore. Set status '3' instead of 'X' if SAP Hana restore is aborted
  • NEW: (#32599) Add Sesam server name to VDI interface name
  • NEW: (#29591) Write warning message to Sesam main log, if graceful abort fails
  • NEW: (#32593) Added function -A cleanup. For Nutanix module, it executes delete_sesam_volume_groups if -a delete_vg is set
  • NEW: (#32591) implemented "-a delete_vg" for Nutanix to delete Sesam volume groups older than days set by the argument
  • NEW: Log final DB
  • NEW: (#29591) Use same basic abort message for backups and restores
  • NEW: (#32488) Set correct backup log path (NOT log) for backups, which terminate in pending queue
  • NEW: (#27166) Fast and graceful job termination by 'sm_break'. Enable core termination after long timeout (15 min for VM based backups) again
  • NEW: (#32573) Abort of restore
  • NEW: (#24714) Set original migration task and follow up event, if single failed backup is restarted
  • NEW: (#32132) Write data store properties as JSON file into DS root directory
  • NEW: (#31928) pass arguments to powershell parameters, escape path to executed script
  • NEW: Add DB
  • NEW: (#31928) call powershell scripts with -file parameter instead of -c
  • NEW: (#25420) For client mount, sm_qm -K sbc_proxy is set for starting sm_vfs job. sbc_proxy parses for -p <number> and to identify qm_stop action and executes umount for a saveset ID parsed from -j <job name>
  • NEW: (#29737) Create new process group on Linux if 'iRCMD()' is called for local command execution
  • NEW: (#25420) sm_mount parses for argument -p and if set, it parses for ssid from -j argument. It then executes "sm_mount umount <ssid>"
  • NEW: (#32488) Set correct backup log path (NOT log) for backups, which terminate in pending queue
  • NEW: (#32548) Add 'exec_state' flag to DB result tables
  • NEW: (#31928) sbc get backup source from ps1 script (sbc -b -s @NUL -f @sbc_gen.ps1 -o shortlog c
  • NEW: (#8554) Print 'sbc-1028 Error
  • NEW: sbc kill option to wait process finish before any activity performed (-w [<preaction_wait>
  • NEW: (#32141) Abort of external backup does not work as expected. Do not remove 'sm_sbc_com_ext' job from queue before caller can execute regular 'sbc_com closejob' function, which will do cleanup correctly
  • NEW: (#28169) Wait for for complete Sesam startup. Add desription for 'gv_conf_drive_config_timeout=<sec.>'
  • NEW: (#32498) sbc_proxy -A umount works with "ALL
  • NEW: (#32120) Added "prepare" to the list of actions to execute "sm_mount umount -s ALL"
  • WIP: (#31463) Restore of a single file overwrites the parent directory. Windows
  • WIP: (#31463) Restore of a single file overwrites the parent directory. Linux
  • WIP: (#30821) Support of LTO9 first time media initialization. In case of library MSL2024 (along with the 1/8 G2) loading a not-calibrated LTO-9 media is not possible. SLU now shows these media with 'tape=2' to indicate the non accesibility. SM_ROBOT writes message 'I038-MEDIA Media ... in slot ## must be calibrated before first use.'
Copyright © SEP AG 1999-2024. All rights reserved.
Any form of reproduction of the contents or parts of this manual is allowed only with the express written permission from SEP AG. When compiling and designing user documentation SEP AG uses great diligence and attempts to deliver accurate and correct information. However, SEP AG cannot issue a guarantee for the contents of this manual.