GUI client/REST server: 8cf187e74e wip: (#23460) - VMware CBT snapshot backup starts with full after VM moved to a different node bcs. of vMotion change: Restart CBT calculation in case no change ID can be found for a virtual disk and a FULL backup has to be performed. wip: Update Intellij Project.xml file wip: Fix build errors new: Add the 'sm_sms_watch' log to the backup logs archive. CHANGE: (#31748) - RMI service fails to start when locations link to each other wip: Update "dfix3" launch configuration change: Update build scripts to new Jenkins server wip: (#32201) - Automatic UI server VM cache refresh may not work anymore after running into timeout error change: Print the correct operation when successfully completing the refresh of a service. wip: Add missing interface. CHANGE: (#23460) - VMware CBT snapshot backup starts with full after VM moved to a different node bcs. of vMotion change: Ignore also SSLHandshakeException with message 'Insufficient buffer remaining for AEAD cipher fragment (2). Needs to be more than tag size (16)' BUGFIX: (#33762) - Opening help links with internet access trigger a NullPointerException CHANGE: (#33755) - javax.net.ssl.SSLHandshakeException is not handled in REST server Netty handler context CHANGE: (#31921) - Manual resetCBT does not work due to caching problems Auto deploy Web UI NG2 from Jenkins job sep-webui-ng2-510 build 51 (d9d57c2c787238008397dec06761d49bf6425593) change: Show value of 'description' property in 'Description' column. Add the number of tasks at the end of the description if available. Auto deploy Web UI NG2 from Jenkins job sep-webui-ng2-510 build 50 (1f571c088feda5f63b18e7fa9cabe8a846da1742) wip: (#33658) - Diff backup fails after revert to snapshot CHANGE: (#33658) - Diff backup fails after revert to snapshot wip: (#33658) - Diff backup fails after revert to snapshot wip: (#32201) - Automatic UI server VM cache refresh may not work anymore after running into timeout error wip: (#33689) - UNC backup with backslash exclude is not possible for "File Pattern" wip: (#32201) - Automatic UI server VM cache refresh may not work anymore after running into timeout error wip: (#33689) - UNC backup with backslash exclude is not possible for "File Pattern" BUGFIX: (#33689) - UNC backup with backslash exclude is not possible for "File Pattern" wip: Fix build error BUGFIX: (#32930) - After change/add ACL all existing "allow" ACL are lost NEW: Added column 'Backup Source' to client backup report and fix resolving the external resources from the resource bundles. BUGFIX: (#32549) - GUI hides shown entries after some seconds when opening results of a VMware task group wip: (#32201) - Automatic UI server VM cache refresh may not work anymore after running into timeout error wip: Fix build errors after merge. BUGFIX: (#33637) - Filter by date range does not show correct results in Datastore->Savesets BUGFIX: (#33652) - 33652 - Automatic poll of RSS feed does not work anymore Web restore assistant: 402468f6 wip: (#33766) - Wrong URL is used, when "New Web Restore" is called from "Tasks by Clients" tab wip: (#33775) - Bottom object is hidden at SFR with Web RA CHANGE: (#33753) - Offer Proxmox target store connected to target node only wip: (#33609) - Restore: It should be possible to select a data mover for all backups where a data mover is used. wip: (#33664) - Select PVE Storage default before continue wip: (#33589) - Proxmox datastore selection is wrong when the same store names are used on different nodes Web client: d9d57c2c7 wip: (#33301) - No action button shown for SCE notification right after installation wip: (#31548) - Simple-, Advanced-Mode in WEB-UI Kernel: b3916f71e8 BUGFIX: (#33824) Set 'SM_VAR_INI' variable in '/etc/sesam2000.ini' correctly BUGFIX: (#33707) The rear update was not performed during SEP sesam Server update since it is not supported in the sm_db_update.ini to have two conditional expressions combined. Solution: Move the update of rear to a separate Update section. new: Increase SEP sesam version. Old version is 5.1.0.22, new version is 5.1.0.23. bugfix: (#33376) sm_dedup_interface sporadically cores during si3 version check bugfix: (#33789) Fixed XenServer dir. VM record "metrics" had a null rereference for a VM. Fixed by checking for null reference and avoiding exception BUGFIX: (#33786) Core dump in 'sm_sms_backup', because of wrong format specifier in trace message Fix compiler warnings change: 'di server status' should allow connection running up to 5 min and 'di server stop' up to 15 min BUGFIX: (#33774) - Fixed Xen CBT backup with alternative excludes used. Removed exclude argument -x from SBC command for backup of XEN config files bugfix: (#33515) Add relocation feature for backup on Windows. Problem: Recent merge missed changes in makevolb.c bugfix: (#33515) Add relocation feature for backup on Windows. Change: Implement new function iRelocateItem() to perform relocation for Volume, directory and file. If relocation is not performed then item is backed up with origin path. Add summary message: 'sbc-3325: Info: ... path relocations performed'. Note Volume is only relocated if pattern matches exactly, e.g. -R .*/localhost/C\$==::==/file_srv1/home bugfix: (#33515) Add relocation feature for backup on Windows. Problem: Since commit c3cacc2 compilation of sbc_vmware failed with error C1311 due to new element in structure _MTF_GLOBAL (int iVolumeLengthMod) which must be correctly initialized bugfix: (#33515) Add relocation feature for backup on Windows. UNC path backup with relocation new: (#33515) Add relocation feature for backup on Windows. Relocation for UNC paths new: (#33515) Add relocation feature for backup on Windows. Remove extra slash new: (#33515) Add relocation feature for backup on Windows BUGFIX: (#33757) Adjust counting of Unit license if higher tiers are available. Problem: Since commit 492731e (#23477) the sort order of Unit T1/T2/T3 calculation and output was reversed and broke calculation of remaining higher licenses. Solution: Improve calculation of remaining UNIT_T3/T2 units and use them for lower units bugfix: (#33706) Since update SESAM_BACKUP failed with: Backup source could not be found. Exiting.. Additional: Remove saveset directory after unmount. change: (#27612) Merged all changes from branch develop for Proxmox BUGFIX: (#33706) Since update SESAM_BACKUP failed with: Backup source could not be found. Exiting.. Solution: Since commit 7952d37 sbc_proxy doesn't take care about changes to o.opts therefor o.sbcargs must be modified NEW: (#27612) Proxmox backup: Update DB:results.sub_type and source with finally used values bugfix: (#27612) Fixed ZFS and LVM incremental restore. Exited restore loop after first saveset from the chain (full) was restored change: (#27612) Incremental backup for Proxmox: Use virtual disk name to select correct saveset duing restore change: (#27612) Use restore id send by 'sbc_com' to check active main restore task new: Increase SEP sesam version. Old version is 5.1.0.21, new version is 5.1.0.22 afer Proxmox LVM beta release BUGFIX: (#33756): OVirt: Restore of VM with ISCSI attached Volumes fails: add initial_size parameter during disk creation bugfix: (#33690) Proxmox LVM backup using BDA saves additional empty item. Initialize char arrays bugfix: (#33747) resolve azure dependencies issue; add content to tmp validation file on Azure connect to handle issue with null range exception change: (#33638) Changed name of restored VM volumes to include restored VM ID and removed restore ID from the name bugfix: (#33702) Removed code that tried delete snapshot during restore. ZFS function get_volumes() thrown an exception if executed for a non existing pool. Changed to write log and return empty list. fixed: (#33638) Complete storage dictionary wad added to restore target instead of just VG name change: (#27612) added log 3500 with VM ID to be restored change: (#33638) attach restored LVM volumes to VM based on the real target data store instead hardcoded default "local-lvm" change: (#33638) restore LVM volumes to LVM VG returned by API based on target data store instead of default hardcoded Proxmox VG "pve" change: (#27612) Fixed message 3122 for containers. VM name is empty, use VM ID new: Increase SEP sesam version. Old version is 5.1.0.20, new version is 5.1.0.21 for Proxmox LVM beta release change: (#27612) Write log message 3122 for restored VM also for VZDUMP restore and for restore of containers. Set VM name with API call for restored VM via VZDUMP rstore change: (#27612) Set name of the restored VM to r--- change: (#27612) Set name of the restored VM to include backup ID and restore ID bugfix: (#33702) Fixed ZFS restore, only one volume was restored new: (#27612) Add 'COM_INQUIRE' for Proxmox bugfix: (#27612) Fixed message codes for logs during Proxmox switch backup sub type new: (#33702) Merged changes for Proxmox ZFS restore to release branch change: (#27612) Changed log code to warning 2989 when backup sub type is not set by the user and Sesam sets it automatically CHANGE: (#33743) create CHUNK logger for the specific log so it will be NOT included in backup log file; delete INFO_APPENDER bugfix: (#27612) Fixed parsing of LVM size during Proxmox restore, did not work for all cases bugfix: (#27612) Fixed backup of a restored thin lvm VM. During restore, restored volumes were not created as thin volumes, so thin backup of such VM failed, snapshot could not be created change: Merged changes for Proxmox module from develop branch bugfix: (#33724) File restore from a tape ends with 'no valid MTF Stream ID was found'. Problem: Since recent commit the Linux ACL and ExtendedAttributes restore failed (CI: test24BackupACLEALinux). Solution: Add padding to file len if blk_size is set BUGFIX: (#33724) File restore from a tape ends with 'no valid MTF Stream ID was found'. Solution: If MTF stream block ends with '4-byte Stream Alignment' then no further SPAD stream follows NEW: (#27612) Incremental backup for Proxmox using thin LVM volumes wip: (#33690) Proxmox LVM backup using BDA saves additional empty item. Skip empty item with no Warn new: (#27612) Set backup sub type for Proxmox restore NEW: (#27612) Proxmox backup: Call TRIM() for DB:results.sub_type BUGFIX: (#32907) sm_sms_watch log shows 'ERROR: qm_lib_close_answer() returned with -1 != 0, state from STATUS=... 0'. Solution: Asynchron ReadFile() may return unhandled ERROR_IO_PENDING (997): Overlapped I/O operation is in progress. bugfix: (#33662) GUI filter may show same sesam day twice at media actions because media init writes DB:media_results.sesam_date' without trailing '00:00:00'. Extend size of date variable BUGFIX: (#29530) Exchange Server Backup - Avoid possible risk of accidental Log truncation. Problem: An issue could be overlooked when copying metafiles from RDS. BUGFIX: (#33683) Error in the daily log when executing a remote command. Solution: If processing is interrupted due to search pattern (here PID) then do not write to daily log. Write result to daily log after final processing. BUGFIX: (#33636) Update regular expression in 'DB:task.exclude' from BRE to ERE Update 'cm_db_access' to current develop branch NEW: (#33638) Incremental backup for Proxmox using LVM storage backend NEW: (#33618) Recreatation of LIS file from saveset using LISv2. Use LISv1 if 'int=sidf' is found in saveset info NEW: (#33618) Recreatation of LIS file from saveset uses LISv2 for default bugfix: (#30544) SEP sesam BSR Pro must be terminated with a distinct message if it is not installed on the client. Bugfix: type on registry key BUGFIX: (#30544) SEP sesam BSR Pro must be terminated with a distinct message if it is not installed on the client. End with 'SEP sesam BSR Pro for Windows is not installed on the client.'. Additional: Remove obsolete code lines. Restart service always. Keep log files until next backup. BUGFIX: (#33632) Start of saved MySQL restore task doesn't set recover options bugfix: (#32987) QM starts job in queue with weight '0'. Compilation on Linux failed. BUGFIX: (#32987) QM starts job in queue with weight '0'. Solution: Only jobs that have been requested for this resource queue are started in this case. change: (#33528) Add protocol to 'http://localhost:11000' CHANGE: (#33528) Use 'http://localhost:11000' STPD http port for source side deduplication. Set it for external backups CHANGE: (#33528) Use 'localhost:11000' STPD http port for source side deduplication on clients with Sesam version >= 5.1.0.18 NEW: (#31976) Adjust the installer to also accept the new O&O package which has a new major number 19. NEW: (#24717) Enable deduplication in 'sm.ini' during drive configuration all the time BUGFIX: (#33675) Tape initialization during EOM continues, even if label from following tape couldn't be read BUGFIX: (#33603) Execute programs stored in 'gv_rw_bin:' before from 'gv_ro_bin' BUGFIX: (#31504) fix paths to logback related files on Windows bugfix: (#33425) fix LoggerContext usage so change logback.xml in runtime will not switch off logging to /events/current.log and si3 server will not hang. Add "neverBlock"=true for AsyncAppender. Add log to see logback configuration file is changed on startup; bugfix: (#33425) remove redundant thread blocking synchronization for TrieMap in PagesFilesTracker bugfix: (#33369) fix log the invoke null pointer exception bugfix: (#33369) fix tmp pages repair for cloud connectors; fix local deep page check on repair; cleanup logs bugfix: (#32501) fix FSCK Failed case to not finishing twice; improve logging for FSCK and GC sweep bugfix: extend logs on FSCK; add extra-condition on PCCK fast check bugfix: check if file exists after FIles.copy operation as this operation is not atomic bugfix: (#33369) do not iterate previously corrupted pages on repair; mark original page with corruption mark when move it to trash during repair; provide check action type in post modification test for smart task managing bugfix: (#33369) fix postModificationTest check condition bugfix: (#33369) use valid tmp files for repair if no other file available; run inrc PCCK after repair if server state is NOT OKl fix logs bugfix: (#33584) remove map.synchronized for thread-safe TrieMap for operations in ObjectStore bugfix: fix PCCK fast check condition bugfix: (#33544) persist fsck run state even if there were no items to check bugfix: (#33544) fix object lock on OCCK; fix OCCK and corruption store logs bugfix: enable pages local check for azure; enable async GC for Azure; clean up logs; bugfix: fix gc log when for corrupted page negative count of chunks to keep was logged; remove backup log about total appended links count bugfix: fix heap space calculation in dedup_purpose_jvm bugfix: (#33335) add check of object data links on backup finish; add corresponding entry in ini file ti initiate object links check bugfix: fix pages update retention method; add total repaired pages count in sm_dedup_cli output; fix logs bugfix: bypass retention period for pages deletion on GC bugfix: (#33374) fix started/finished workers calculation during task stop bugfix: (#29261) fix logic with handling ports on startup bugfix: fix repair function to skip unpaired page body files on pages recovery bugfix: (#31846) fix "amountOfFreedData" calculation for gc status output bugfix: add logs on repair bugfix: (#33335) cache object links during backup bugfix: (#30360) move essential log line from debug level to info bugfix: fix issue when retention period was not updated for uploaded object; fix object deletion from s3 for versioned stores bugfix: fix NullPointerException during pages index iterating; increase awaiting timeout for s3 chunk reading, add retry in case of timeout exception bugfix: fix replication ssl issue bugfix: (#32761) fix s3 purge command; improve s3 logging bugfix: (#33202) update dir trackers on purge bugfix: (#33202) optimize s3 purge commands bugfix: (#33202) s3 versioning changes bugfix: (#33202) optimize GC chunk reading and deep PCCK bugfix: (#33202) wrap store versioning check bugfix: (#33202) add content length info to s3 put object request bugfix: (#32761) fix logs; extend delete object feature with general delete method bugfix: (#32761) improve s3 error handling; optimize retention period setting; add version based delete; optimize resource usage on server start; log error during version cleaning; disable page fix 1 as redundant bugfix: sweep all pages on gc, not only from page-index BUGFIX: fix race condition during PCCK and GC when non thread safe library for sha1 hash calculation was used in async code bugfix: (#32571) increase finished workers counter in case FSCK finish due to error; extend OCCK status to indicate objects that links to corrupted pages bugfix: (#32761) change await timeouts for several s3 operation from inf to limited, increase timeouts count; improve logging on s3 pages/object deletion; extend s3 version cleaner with retry logic bugfix: remove redundant logs on PCCK bugfix: (#31846) fix calculation for average time page processing bugfix: add pendingForDeletion list to track object in queue for deletion bugfix: do not reload object header into ObjectStore when object in queue for s3 deletion bugfix: (#33202) optimize GC chunk reading and deep PCCK bugfix: (#33202) wrap store versioning check bugfix: (#33202) add content length info to s3 put object request bugfix: (#32761) fix logs; extend delete object feature with general delete method bugfix: (#32761) improve s3 error handling; optimize retention period setting; add version based delete; optimize resource usage on server start; log error during version cleaning; disable page fix 1 as redundant bugfix: sweep all pages on gc, not only from page-index BUGFIX: fix race condition during PCCK and GC when non thread safe library for sha1 hash calculation was used in async code bugfix: (#32571) increase finished workers counter in case FSCK finish due to error; extend OCCK status to indicate objects that links to corrupted pages bugfix: (#32761) change await timeouts for several s3 operation from inf to limited, increase timeouts count; improve logging on s3 pages/object deletion; extend s3 version cleaner with retry logic bugfix: remove redundant logs on PCCK bugfix: (#31846) fix calculation for average time page processing bugfix: add pendingForDeletion list to track object in queue for deletion bugfix: do not reload object header into ObjectStore when object in queue for s3 deletion bugfix: fix page related logs; remove redundant page list ordering during PCCK bugfix: log details why deep page check activated bugfix: (#33138) do not return null to recover error on page meta data retrieving for Azure connector bugfix: fix local deep page check; improve pcck logging bugfix: add logs to postCreationTest bugfix: add prefix p to all logs where pageId present bugfix: (#32761) retry s3 operations on GC, FSCK when Completion exception occurs; increase pageStore cache size for cloud connectors; fix NullPointerException during replication stop when server shutting down bugfix: (#33109) bump netty-tcnative-boringssl-static version bugfix: (#31504) handle logback related paths based on OS bugfix: add exponential retry on setretention command; improve error handling on OCCK; extend logs; bump akka version; revert some older changes bugfix: fix tests bugfix: retrieve pages list from page index on server status call instead of using s3 pages list command 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: fix page deep checking condition bugfix: (#33069) flush page body file after each chunk append bugfix: (#32812) increase akka.http.server.parsing.max-method-length bugfix: (#33662) GUI filter may show same sesam day twice at media actions because media init writes DB:media_results.sesam_date' without trailing '00:00:00' CHANGE: (#32318) Set 'gv_rw_bin' directory to '/var/bin' instead of '/bin'. (/var/opt/sesam/var/bin) commit bdae332c1ee185887dac0549221207bf0c14e3db Author: meteorologix Date: Tue Jul 23 11:31:29 2024 +0200 CHANGE: (#32318) Set 'gv_rw_bin' directory to '/var/bin' instead of '/bin'. (/var/opt/sesam/var/bin) bugfix: (#31932) DB:restore_results are not purged during 'sm_newday' anymore. Problem: Since the fix with commit 40bdff3 the message 'W014-BASICS [...]: CLEANUP error catching date from DB:results' is printed into daily log NEW: (#31976) Add new BSR Pro version 19.1.147 to SEP sesam package new: Increase SEP sesam version. Old version is 5.1.0.19, new version is 5.1.0.20. new: Increase SEP sesam version. Old version is 5.1.0.18, new version is 5.1.0.19. BUGFIX: (#31504) fix paths to logback related files on Windows BUGFIX: (#33652) Automatic poll of RSS feed does not work. Set URL protocol to 'https://' during Sesam server update