SureSync 7 Version History

SureSync 7.2.19 Known Issues

Version 7.2.19 was posted on December 28, 2016

Version 7.2.19 is expected to be the last version of SureSync 7. We strongly encourage you to plan an upgrade to SureSync 8 to ensure access to all the latest fixes and enhancements.

Return to top of page

SureSync 7.2.18 Known Issues

Version 7.2.18 was posted on October 26, 2016

  • When processing a delta copy, it was possible to receive a “value does not fall within the expected range” error when receiving an invalid value from Microsoft’s Remote Differential Compression engine. This has been fixed in 7.2.19.
  • In some instances, a delta copy could get attempted on a destination file that is too small to support deltas. This has been fixed in 7.2.19.
  • On Windows 2008 R2/Windows Vista and newer, the default certificate used for TLS encryption is now generated with SHA256. 2003/XP will continue to use SHA1 due to lack of support. This change will happen automatically the next time the Communications Agent refreshes the certificate or you can force the change by stopping the Communications Agent, deleting the SureSyncSSL certificate from the local computer’s personal certificate store and restarting the agent. This has been fixed in 7.2.19.
  • In rare situations, a deadlock could occur in SyncFiles related to accessing the SureSync database. This would stop the SyncFiles from processing additional files. This has been fixed in 7.2.19.
  • The Communications Agent was not returning the temporary folder when enumerating, which prevented purging old, abandoned temporary file copies. This has been fixed in 7.2.19.
  • In rare timing circumstances where users are opening the same file on multiple machines when a path drop is occurring, it was possible for a lock to not be released. This has been fixed in 7.2.19.
  • When running with File Locking, the internal lock list in memory could have unnecessary internal locks. This could have a negative performance impact. This has been fixed in 7.2.19.
  • If running File Locking on a root drive letter (for example E:), the driver could attempt to lock items such as \$Extend\$Quota which could cause “an internal application error has occurred. File locking support terminated due to a severe error communicating with the driver. System.ArgumentException” an item with the same key has already been added.” This has been fixed in 7.2.19.
  • If a Job/Schedule/Monitor started up and had many Advanced Open File Support (VSS) file copy requests pending, it was possible for SureSync to attempt to generate multiple VSS snapshots at once which could negatively impact performance. This has been fixed in 7.2.19.
  • With File Locking, it was possible for the Communications Agent to not release a lock when a Job requested it to. This would result in locks remaining when they shouldn’t. Enhancements have also been made to locking during renames to avoid abandoned locks. This has been fixed in 7.2.19.

Return to top of page

SureSync 7.2.17 Known Issues

Version 7.2.17 was posted on July 19, 2016.

Endpoint

  • When attempting to synchronize an Endpoint client where an Alert definition exists, the Endpoint client would fail to connect with a database error that includes the text “AlertDefnId.” This has been fixed in 7.2.18.

File Synchronization

  • If a full scan determined a folder needed to be deleted but the folder could not be deleted due to it not being empty (for example, the folder included files excluded from processing), the scan would not be marked as satisified. This could result in the scan of that folder being replayed whenever the job was restarted. This has been fixed in 7.2.18.
  • In rare conditions, it was possible to encounter a ERROR_BAD_NET_NAME message when resolving a path. This would trigger a “failed to find FRN in journal” error that caused a full rescan. This has been fixed in 7.2.18.
  • If a Schedule contained multiple Jobs and one of the Jobs included a reparse point, it was possible for a “System.NullReferenceException: Object reference not set to an instance of an object” message to occur. This has been fixed in 7.2.18.
  • In a Job performing a long running folder scan, it was possible for the Node cleanup routine that cleans up obsolete database records to produce a “System.Data.DbConcurrencyException: Concurrency violation” error that stopped the Schedule or Real-Time Monitor. This has been fixed in 7.2.18.
  • A “System.OverflowException: Arithmetic operation resulted in an overflow” error could occur that would cancel a Real-Time Monitor. This has been fixed in 7.2.18.
  • It was possible to receive a “System.InvalidOperationException: Collection was modified; enumeration operation might not execute” error in the Event Viewer that would lead to a Real-Time Monitor being cancelled. This has been fixed in 7.2.18.

User Interface

  • When displaying Detailed Status on an Endpoint Real-Time Monitor from the server, the statistics from the Endpoint machines were not being displayed. This has been fixed in 7.2.18.
  • On the Endpoint tab, when selecting to display Detailed Status it was possible to receive a “System.Collections.Generic.KeyNotFoundException: The given key was not present in the dictionary” error message. This has been fixed in 7.2.18.

Return to top of page

SureSync 7.2.16 Known Issues

Version 7.2.16 was posted on April 11, 2016.

Communications Agent

  • When processing a delta copy, it was possible to receive a “System.ArgumentException: source array was not long enough. Check srcIndex and length, and the array’s lower bounds” error message. This has been fixed in 7.2.17.
  • When running on a computer without the C++ Runtime for Visual Studio 2013 installed would produce a “System.IO.FileNotFoundException: Could not load file or assembly ‘SPIVssLib_2003x64.dll’ or one of its dependencies. The specified module could not be found.” error in the Event Viewer. The installer for SureSync installs the C++ Runtime for Visual Studio 2015 (14.0). The VSS components have been recompiled to use the distributed runtime. This has been fixed in 7.2.17.
  • When testing or attempting to establish a connection, a “System.FormatException: String was not recognized as a valid DateTime” error could occur. This is due to a WPF/SOAP incompatibility with time zones. This would occur with a Communications Agent located in a UTC + time zone is using encryption and attempts to generate a connection to another Communications Agent located westward from the machine generating the connection. All eastward connections and connections from UTC or UTC – time zones were always successful. This has been fixed in 7.2.17.
  • The Communications Agent could fail to release a File I/O class during folder enumerations where a folder was missing on one or more of the paths in the Job. This could result in an accumulation of these File I/O classes that could cause path losses and these losses did not always recover. This has been fixed in 7.2.17.

Database

  • If a SureSync database had a very large number of Jobs which executed very frequently, the SureSync Desktop could become unresponsive because of the excessive refreshing required to accurately display status. This could take hundreds or even thousands of Jobs to produce the symptoms. This has been fixed in 7.2.17.

Endpoint

  • When launching the SureSync Endpoint Client, it was possible to receive a “System.NullReferenceException: Object reference not set to an instance of an object” error. This has been fixed in 7.2.17.

File Locking

  • If a file lock was generated for a rename event and the path is lost, it was possible for a lock to remain after the path is restored. This has been fixed in 7.2.17.
  • Real-Time Monitor could be stopped with a “System.NullReferenceException: Object reference not set to an instance of an object” error. This would occur when the Job attempted to report lock status before it was fully initialized. This is a timing issue that affects jobs that are very slow to startup (most likely due to slow connections). In addition, the monitor would need to have outstanding locks on an agent due to an interruption. This has been fixed in 7.2.17.
  • Enhancements and corrections have been made to folder renames with file locking. This has been fixed in 7.2.17.
  • If a file lock was generated for a rename event and the path is lost, it was possible for a lock to remain after the path is restored. This has been fixed in 7.2.17.

File Synchronization

  • When running a Trial Run, the amount of data reported to be copied would be double the actual amount. The size of the source was incorrectly added into the total making the value twice the amount (asssuming a 2 path Job). This has been fixed in 7.2.17.
  • FTP controls upgraded to the latest version. This has been fixed in 7.2.17.
  • It was possible to receive a “System.ArgumentOutOfRangeException: Non-negative number required. Parameter name: count. This action will not be retried” message when attempting a file resume. This has been fixed in 7.2.17.
  • The allowed time for an active file copy to not report back copy status has been increased from 6 minutes to 15 minutes. This has been fixed in 7.2.17.
  • If a Monitor had a large number of root paths on the same agent and that agent went offline, the agent and the job could loop. This would result in the agent and the job needing to be restarted. This has been fixed in 7.2.17.
  • It was possible for the NodeID counter to have an incorrect, obsolete value. This could produce “an item with the same key has already been added” and “violation of primary key constraint ‘PK_Nodes_Rel_NodeID” errors to occur. This has been fixed in 7.2.17.
  • When processing a file containing multiple alternate data streams where one stream that needed to be copied followed one to be skipped, it was possible to receive a “Warning: System.ArgumentNullException: Value cannot be null. Parameter name: Waithandle array may not be empty.” message. This has been fixed in 7.2.17.
  • When using an UNC path as a destination in a real-time monitor, it was possible for WriteToFile Error=0 error to occur when a path was lost during a file transfer. This would cause the monitor to not drop the path and not resume processing. This has been fixed in 7.2.17.
  • When processing files with multiple data streams, where one or more streams were skipped, the file would not be copied and would go into the Retries Queued with a “hash mismatch” error. This has been fixed in 7.2.17.
  • When launching multiple Real-Time Monitors which contained a large number of Jobs, it was possible to receive a “System.InvalidOperationException: Collection was modified; enumeration operation might not execute” error that would occur in only specific timing circumstances. This has been fixed in 7.2.17.
  • In rare timing situations, with a multi-path Job running under a Real-Time Monitor, it was possible for scans processed while a path was offline to not be fully retried on the recovered path when it came back online. A full scan would be needed to complete the actions to the recovered path. This has been fixed in 7.2.17.

Scheduler

  • A “System.ArgumentNullReference: Value cannot be null” error could occur that would stop the Scheduler service. This would occur when alerts were processed. This has been fixed in 7.2.17.

SQL Protection

  • When running a SQL Protection job, a “System.MissingMethodException” could occur canceling the Job. This has been fixed in 7.2.17.

User Interface

  • In the Log Viewer, the “Event Time” field would display MM/DD/YYYY when the locale called for DD/MM/YYYY. This has been fixed in 7.2.17.
  • On the Statistics tab of a Job, the “Last Modified Date” statistic could display the date incorrectly. This has been fixed in 7.2.17.
  • When adding a path to a Job with a one-way Rule that selected destinations individually, adding a new path would not show in the destination path list unless you switched to the “(all available destination paths)” and back. This has been fixed in 7.2.17.
  • When attempting to delete a path within a Job, it was possible to receive the error “an unexpected error has occurred. System.Data.DeleteRowInaccessibleException.” This has been fixed in 7.2.17.
  • In the SureSync Desktop, attempts to add a new path to an existing Job under a Real-Time Monitor were incorrectly being detected as UNC paths when they were correct Communications Agent path references. This prevented adding the path. This has been fixed in 7.2.17.

Return to top of page

SureSync 7.2.15 Known Issues

Version 7.2.15 was posted on April 22, 2016.

File Locking

  • If a file lock was established at the same time on multiple machines, it was possible for a lock to not be released. This has been fixed in 7.2.16.
  • When running a multi-way Real-Time Monitor with file locking, it was possible to encounter errors where the Job could not delete a folder because our locking engine has a handle on a folder that should have been released. This has been fixed in 7.2.16.

File Synchronization

  • When starting a real-time monitor, it was possible to encounter an “unable to start change journal. System.NullReferenceException: Object reference not set to an instance of an object” error which would cause the path to drop. This has been fixed in 7.2.16.
  • It was possible to receive a “System.Threading.ThreadAbortException error in the Application event log. This would occur when attempting to start a Schedule or Real-Time Monitor that was already running. This has been fixed in 7.2.16.
  • It was possible to encounter “System.NullReferenceException: Object reference not set to an instance of an object.” This has been fixed in 7.2.16.
  • A correction has been made to the throttling code. If you defined severely limited bandwidth such as 1Mbps and 20% dynamic throttling, it was possible for the block size used to be shrunk to 4K and many small blocks sent in parallel incorrectly. This could overload the agent and cause high CPU usage. This has been fixed in 7.2.16.
  • If Change Journal entries are used on a Job and the last rule is a Skip rule, folder security changes may not be processed unless a full scan occurs. This has been fixed in 7.2.16.

Scheduler

  • It was possible to encounter a “System.ArgumentNullException: Value cannot be null” error from the Scheduler service. This has been fixed in 7.2.16.
  • It was possible for the Scheduler service to attempt to launch the same Schedule multiple times at the same scheduled time. This has been fixed in 7.2.16.

SQL Protection

  • If a SQL Protection job was restoring to a standby SQL server using SQL authentication, this would fail with an “Error performing restore. System.ApplicationException: Invalid source string on SQL Job” error. This has been fixed in 7.2.16.

User Interface

  • If you changed a one-way rule to a multi-way rule, the multi-way security option drop-down on the Copy Security tab would not be populated until the Desktop was relaunched. This has been fixed in 7.2.16.
  • If a new Priority was defined in Options in the SureSync Desktop, this Priority would not appear in the drop-down for a Schedule or Monitor until the Desktop was relaunched. This has been fixed in 7.2.16.
  • When attempting to update existing Jobs with the template job functionality, it was possible to receive an “Unable to clone job. System.Data.ConstraintException: ‘Column, RootNumber’ is constrained to be unique.” error. This has been fixed in 7.2.16.


Return
 to top of page

SureSync 7.2.14 Known Issues

Version 7.2.14 was posted on March 15, 2016.

File Synchronization

  • Rename support has been enhanced when running a File Locking real-time monitor. If a user has a file open in one office and a user in another office attempts to rename the same folder, the rename will not be prevented until no one has files open. This mymics the behavior of Windows Explorer on a single server. This has been fixed in 7.2.15.
  • When running with multiple Schedulers, it was possible to encounter a “System.Data.DBConcurrencyException: Concurrency violation: The UpdateCommand affected 0 of the expected 1 records” error that would cancel the Schedule or Monitor. This has been fixed in 7.2.15.
  • When a root path in a Job was the drive letter (volume root), it was possible to encounter an error “the parameter is incorrect” caused by a check to test for a recursive reparse point failing. This has been fixed in 7.2.15.
  • If a Change Journal was less than 1 day old, on each path recovery the Job would increase the size of the journal. This has been fixed in 7.2.15.
  • When testing a connection of a UNC path, it was possible to receive a “This type of CollectionView does not support changes to its SourceCollection from a thread different from the Dispatcher thread error. This has been fixed in 7.2.15.
  • When reading from an FTP site, only 4 bytes were being read at a time which would negatively impact Job performance. This has been fixed in 7.2.15.

Real-Time Monitors

  • A Real-Time Monitor could be canceled after encountering a “System.NullReferenceException: Object reference not set to an instance of an object” error. This has been fixed in 7.2.15.

User Interface

  • When adding a root path that is a drive letter with no folder, internally the path was being associated with the local SureSync agent until a Job ran that corrected the record. This could cause license counts to be wrong and other issues. This has been fixed in 7.2.15.
  • A “System.InvalidCastException: Conversion from type DBNull to type ‘Short’ is not valid error could occur when changing Rule types with Backup Root Paths defined. When changing Rule type, you will now be required to select the source if the new Rule method requires one. This has been fixed in 7.2.15.
  • When using “Replace Machine References” to change an agent name, the change would be made but did not display in the Desktop until the Desktop was closed and launched again. This has been fixed in 7.2.15.
  • If a Job was created without a Rule and placed under a Real-Time Monitor with other Jobs, you would be unable to delete the problem Job. This would produce an error “This Job is the only Job in the Real-Time Monitor. This has been fixed in 7.2.15.
  • On machines with UAC enabled, the Desktop was not properly determining group membership. This could result in a user who should have access to items in the SureSync database being unable to access them. This has been fixed in 7.2.15.

Return to top of page

SureSync 7.2.13 Known Issues

Version 7.2.13 was posted on March 4, 2016.

File Synchronization

  • In rare timing situations, an unnecessary loop occurred when more than 40 actions were queued and a retry was queued in the database with a retry time that had passed. The retry was not processed because there were too many actions already in the queue. This could result in high CPU usage for the SyncFiles.exe process. This has been fixed in 7.2.14.
  • When running an SFTP job, it was possible to have a “unable to access root path” error occur incorrectly. This was caused by some types of FTP servers not allowing the SureSync FTP code to change folders to the root. This has been fixed in 7.2.14.
  • In rare timing situations, when processing folder deletes, it was possible for a Job to stop processing. This occurred when a folder delete was delayed because another delete was attempting to be queued at the same time. This has been fixed in 7.2.14.
  • Fixes have been made to support for using a Windows File Server cluster as a root path. This has been fixed in 7.2.14.

User Interface

  • Test buttons would return an “unable to connect to the SureSync Journal service on AGENTNAME” error incorrectly. This has been fixed in 7.2.14.

Return to top of page

SureSync 7.2.12 Known Issues

Version 7.2.12 was posted on February 5, 2016.

File Synchronization

  • In certain timing situations where path losses were rapid, it was possible for a Job to get stuck on a “waiting for active copies to complete” state and not proceed forward. This has been fixed in 7.2.13.
  • Corrections have been made to the processing of encrypted files. This has been fixed in 7.2.13.

Licensing

  • When a trial with an expire date was added to permanent licenses, it was possible in rare circumstances for the trial to incorrectly be assigned the 14 day grace period for activation instead of the trial expire date. This has been fixed in 7.2.13.

Preview

  • A preview could hang after receiving some messages from the Job. The Job would eventually terminate and the Preview would not get any indication that the Job is gone. This has been fixed in 7.2.13.

SyncLockStatus

  • SyncLockStatus has been updated to retrieve a list of user accounts used to access mapped drives. Previously, SyncLockStatus filtered out events for all users other than the user launching the application. When running SyncLockStatus from a machine not a member of a domain, this would result in pop-up notifications not being displayed because the share was accessed as an account different than the user launching the program. This has been fixed in 7.2.13.

Return to top of pag

SureSync 7.2.11 Known Issues

Version 7.2.11 was posted on December 7, 2015.

File Synchronization

  • Corrections have been made to connecting to multiple UNC paths on the same machine when username and password are required. These issues could result in messages such as “unable to connect to UNC share name because the credentials different from other uses” and the Job not running. This has been fixed in 7.2.12.
  • When running a compare Rule, it was possible to receive “incorrect function” errors. This has been fixed in 7.2.12.
  • When performing file copies with tracing on, it was possible to have a copy hang for 5 minutes if the file was a 0 byte file. The copy would be queued for a retry and complete. This error would not occur if tracing was disabled. This has been fixed in 7.2.12.

User Interface

  • Setting the “Multi-Way Security” option on the Copy Security tab of a multi-way Rule would not always save correctly. This has been fixed in 7.2.12.
  • In the SureSync Desktop, if you defined destination paths explicitly instead of using the “(All available destination paths)” option and switched from a one-way rule to a multi-way rule and back it was possible to have no source path defined. This would result in the Job not performing any operations until the source was selected. This has been fixed in 7.2.12.
  • When configuring a new Job/Rule with the wizard in the SureSync Desktop it was possible to have the destination paths not defined in a one-way rule. This occurred if you defined only one path in the wizard and then clicked back to add more paths. This would result in the Job failing stating it did not have enough root paths to execute. This has been fixed in 7.2.12.
  • In the SureSync Desktop, if you altered a mirror rule to a multi-mirror rule and then set it back to a mirror you would get a message “please select a source root path for this rule” even though one was defined on the Roots tab of the rule. This has been fixed in 7.2.12.
  • When deleting a Root Path from a Job, the SureSync Desktop was not always correctly deleting old file history. This has been fixed in 7.2.12.

Return to top of page

SureSync 7.2.10 Known Issues

Version 7.2.10 was posted on October 22, 2015.

Communications Agent

  • Please note that the version of the Communications Agent that will be displayed when testing a connection in this release is 8.0.01.11. This is expected and correct.
  • An “UnobservedTaskException caught – System.AggregateException” error could be logged in the Application event viewer. This was a harmless error in disposing objects. This has been fixed in 7.2.11.

Endpoint

  • The /XMLPath option for Endpoint would check for the new configuration file each time the client launched. If the client had no Internet connection, an error would be presented about the Endpoint not being configured. This support has been enhanced to maintain the configuration and update when the /XMLPath configuration file has changed. This has been fixed in 7.2.11.
  • Enhancements were made to the Endpoint Client for handling the machine being put to sleep. Previously, the status panel would not load for a running job after waking from sleep. This has been fixed in 7.2.11.6.

File Synchronization

  • If a Rule contained a =#*.*, any file that started with a number would be excluded from journal processing but would be processed on a full scan. A change has been made to ignore the # mask and treat it literally. This has been fixed in 7.2.11.
    2] Enhancements were made to the Endpoint Client for handling the client machine being put to sleep. Previously, the status panel would not load for a running job after waking from sleep. This has been fixed in 7.2.11.
  • Performance enhancements have been made to the full scan process when copying security has been defined on the job. In some cases, a Job could get slowed down processing journal entries generated by the propagation of security from the SureSync Job. This has been fixed in 7.2.11.
  • A correction to licensing was made where an expired activation could be counted twice. This has been fixed in 7.2.11.
  • A SqlDateTime overflow error could cancel a Real-Time Monitor. This has been fixed in 7.2.11.
  • A long running Job that computed a large number of journal entries to be processed instead of executing an optimization scan could produce a “System.OverflowException: Arithmetic operation resulted in an overflow” error. This has been fixed in 7.2.11.
  • When synchronizing via UNC path with a username and password defined for the path, if you had multiple Jobs referencing the same UNC path only one would work. The others would report that the root path could not be found. This has been fixed in 7.2.11.
  • Tweaks to real-time path recovery have been made. This has been fixed in 7.2.11.

SQL Protection

  • With SQL Protection, it was possible to receive the error “this differential backup could not be restored because the database has not been restored to the correct earlier state” when doing restores to a remote standby server. This has been fixed in 7.2.11.

User Interface & Status

  • In rare cases, the status service could timeout after 6 minutes and would not reconnect correctly to Jobs/Schedules/Monitors. This could result in these items running but the user not being able to display status or cancel the item. This has been fixed in 7.2.11.
  • In the SureSync Desktop, corrections were made to selecting a folder in the browse panel that was in the root of the display. For example, a folder on the Desktop would show in the root of the path tree and clicking on it would result in the Root Folder of \ instead of whole correct path. This has been fixed in 7.2.11.
  • SPIBrowseLib4.dll was added to the installer. This component needs to be included for browse to work correctly for components such as SQL Protection. This has been fixed in 7.2.11.
  • When using “Replace Machine References” to update a Communications Agent name to another name, testing to the new agent would fail because the new connection information was not completely built. This has been fixed in 7.2.11.

Return to top of page

SureSync 7.2.9 Known Issues

Version 7.2.9 was posted on July 20, 2015.

Communications Agent

  • A number of fixes and improvements have been made to communications between agents that should address path loss situations. This has been fixed in 7.2.10.
  • In rare circumstances with many paths, it was possible for the wrong encrypted credential to be sent to an Communications Agent. Instead of a credential error, the incorrect encryption would be reported as a cryptographic error and the path would be dropped. This has been fixed in 7.2.10.

Endpoint

  • For Endpoint, another variable has been added as an option for building Server Root Paths. The %ClientUser% variable will create a folder named after the user. If you define a server path like C:\Data\%ClientComputer%\%ClientUser% the data will be synchronized into a folder for the machine and the user. For example, C:\Data\Laptop1\User1. This allows for multiple users to use the same Endpoint machine. This has been fixed in 7.2.10.

File Locking

  • With a specific set of lock/unlock events, likely triggered by renames, it was possible for a lock to not be released. This issue is very timing specific. This has been fixed in 7.2.10.
  • The file locking driver could return an unknown error and be unloaded. The error was not getting passed to the Job which could result in files not being locked. The error will now be passed and the path will be dropped. The Job must be restarted to trigger a reload of the driver. This has been fixed in 7.2.10.

File Synchronization

  • When under a heavy load, file copies may be deliberately delayed. In this case, it was possible for SureSync to fail to “ping” the connection to keep it active. This could result in a timeout on the file I/O connection that would drop a path. The path would be immediately recovered. This has been fixed in 7.2.10.
  • A change was made to not leave a wrap trace active when connections are restored after being dropped. This has been fixed in 7.2.10.
  • Optimizations have been made to the process for purging obsolete history from the database when folders are deleted in a multi-directional rule. This has been fixed in 7.2.10.
  • Settings have been optimized for named pipe communication between SureSync components on the same machine. This has been fixed in 7.2.10.
  • When using multiple Schedulers in an environment with many rapidly executing Schedules, it was possible to have some Schedules be rescheduled incorrectly to their next scheduled start time instead of being executed. This has been fixed in 7.2.10.
  • When using the Scripts tab to execute a script on a remote machine, it was possible to receive a “Warning: System.EntryPointNotFoundException: Unable to find an entry point named ‘GetExitCodeProcess’ in DLL ‘advapi32.dll'” message. This has been fixed in 7.2.10.
  • In very rare cases, when synchronizing files via UNC path, it was possible for a write request to take an exceptionally long time to return back to SureSync (in some cases as long as 50 minutes). SureSync would return a network path not found after a 1 minute timeout. If the old I/O write request returned it could reference an old thread that was released and crash the Communications Agent with an unhandled exception. This has been fixed in 7.2.10.
  • When using Advanced Open File Support, it was possible for SureSync to not get a snapshot expire time. This could cause the file to be retried quickly and result in it being copied many times in a row. This has been fixed in 7.2.10.
  • In SQL Server, there is no collation sequence that supports both the German character ß and ss in the same folder (or two folders in the same directory). For example FileßTest.doc and FilessTest.doc will compare the same in SQL. This could produce the error “an item with the same key has already been added” which would cancel the job. Code has been added to provide a descriptive error about this conflict. To resolve, one of the folders or files must be renamed. This has been fixed in 7.2.10.
  • When processing very large files via UNC path over a slow connection using open file support, it was possible for CRC calculations to timeout. This would result in the file being retransmitted. This has been fixed in 7.2.10.
  • When using Advanced Open File Support, it was possible for the same file to be queued for retry multiple times resulting in extra copying. This has been fixed in 7.2.10.
  • When copying small (8K or less) file with multiple steams, it was possible to receive a HashMismatch which would result in the file being queued for a retry and copied again. This has been fixed in 7.2.10.
  • With a Rule using a rule subfolder (to include only specific folders instead of the default \), security objects were copied only to the folders contents and not the folder itself. This has been fixed in 7.2.10.
  • A “System.Data.DeletedRowInaccessibleException” error could occur that would cancel a Job. This has been fixed in 7.2.10.

Licensing

  • When using a server license and synchronizing with a workstation via UNC path, if the validation of that workstation failed it would be flagged a server incorrectly. This has been fixed in 7.2.10.
  • If a workstation license required activation and the grace period days for activation expired, it was possible for it to be incorrectly set a server. This has been fixed in 7.2.10.

SQL Protection

  • When running a backup in SQL Protection as a user configured as a language other than US English, it was possible to receive a “Invalid value specified for EXPIREDATA parameters” error message. This has been fixed in 7.2.10.

User Interface

  • A correction was made to the licensing panel. When using server licenses and synchronizing with workstations covered by the server licenses, the panel would incorrectly ask for activation. You would then be unable to activate the machine because an activation was not really needed. This would not stop Jobs from running and was purely cosmetic. This has been fixed in 7.2.10.
  • A correction has been made to the activations panel. When running a workstation version, it was possible for the activation process to gray out the activate button for the workstation thinking it was covered by a server license. This has been fixed in 7.2.10.
  • Checkboxes have been added to the Activations panel allowing you to select “Show Activated,” “Show Unactivated” and “Show Deactivated.” You can now select any combination of those options and the display will be customized. This setting will be remembered each time the panel is launched. This has been fixed in 7.2.10.
  • The export functionality for logs has been enhanced. Large logs will be split into multiple files when exporting to avoid problems with row limits in programs such as Excel. Clicking on Export in the Log Viewer ribbon bar will launch the new panel. This has been fixed in 7.2.10.

Return to top of page

SureSync 7.2.8 Known Issues

Version 7.2.8 was posted on June 9, 2015.

Communications Agent

  • The error “an entry with the same key already exists” could occur in the application event viewer from the SPIComService/SPIComService251. This has been fixed in 7.2.9.
  • A connection drop was not being properly handled by a Communications Agent during delta processing. This could leave an active file copy that made no progress because it was waiting for writes to a destination to complete. This could result in an error about file copy progress not being detected for 6 minutes. This has been fixed in 7.2.9.
  • Optimizations have been made to delta processing to increase performance. This has been fixed in 7.2.9.
  • File block offset invalid errors could occur when processing a file with deltas. This has been fixed in 7.2.9.

File Locking

  • With File Locking, locks were not released or re-tested if a path was lost after file close events were already sent to the job. This could result in some locks not being released. This has been fixed in 7.2.9.

File Synchronization

  • A rename of a folder could fail because the new name already existed and SureSync was failing to generate an error and retry. An error will be reported and the operation retried. This will result in the old and new folder name existing on all paths instead of one of the folders missing on the other paths. This has been fixed in 7.2.9.
  • When a Schedule running multiple Jobs encountered a Job that was already busy (running manually or in a Monitor), the Schedule would incorrectly cancel the entire Schedule instead of moving to the next Job. This has been fixed in 7.2.9.
  • Changes have been made to better detect file copies that were not making progress. This has been fixed in 7.2.9.
  • When synchronizing files via a UNC path that contains symbolic links, a second validation of time/date stamps is performed to ensure accurate file size and attributes information is received. In some environments, this can cause a performance problem. An option been added to the Options tab of the Root Path which allows this additional check to be disabled. Unchecking “validate symbolic links” will suppress the 2nd check. This has been fixed in 7.2.9.
  • A large number of file lock events could cause a full scan to be broken into sub scans. This in turn could block processing temporary folder deletes, which blocked processing new requests. This could cause file processing to slow down significantly. This has been fixed in 7.2.9.
  • If a subfolder scan that was generated from a long running scan is interrupted by an offline path, a re-scan of the folder may not be triggered when the path comes back online. The scan would be generated on the restart of the job but that should not be needed. This has been fixed in 7.2.9.

User Interface

  • If a system clock was set backward in time, it could cause a Schedule or Real-Time Monitor to drop its subscription to the status service. This would result in the Schedule or Real-Time Monitor successfully synchronizing files but being unable to display status in a Detailed Status panel. This could happen with virtual machines where clocks drift or if a machine was synchronized to an incorrect time source. This has been fixed in 7.2.9.
  • When setting an invalid schedule start time (a time in the past) on a Run Once Schedule that was in a held state, the desktop was not prompting about the invalid time. The Schedule would simply be put back into held state. This has been fixed in 7.2.9.

Return to top of page

SureSync 7.2.7 Known Issues

Version 7.2.7 was posted on April 21, 2015.

ArchiveSync

  • When an archive rule was not defined for an Archive Job, the Job would produce a number of “object reference not set to an instance of an object” and “error while thinning archive”: messages. This has been fixed in 7.2.8.

Communications Agent

  • It was possible to receive a “the process was terminated due to an unhandled exception” error from the Communications Agent service on a machine in certain instances when a Job was shutting down. This has been fixed in 7.2.8.
  • When using deltas and compression, it was possible to have an error generated that would result in a path loss. This has been fixed in 7.2.8.
  • When defining a timeout value in the Local Communications Agent Config Utility, it was possible to have the value defined in the SureSync database override that value. This could result in a shorter timeout being used than what was defined in the utility. This has been fixed in 7.2.8.
  • A Communications Agent service could fail to start with a “System.Data.ConstraintException: Failed to enable constraints. One or more rows contain values violating non-null, unique and foreign-key constraints” error in the Application Event log. This error is caused by a configuration file for the Communications Agent missing a required record. Changes have been made to prevent this from occurring. This has been fixed in 7.2.8.

Endpoint

  • When Endpoint shuts down, it will now check the database to determine if any Jobs are running that are not reporting status. This has been fixed in 7.2.8.
  • In Endpoint, the Endpoint Client could produce an “object reference not set to an instance of an object” when losing and reestablishing a connection. This has been fixed in 7.2.8.
  • In Endpoint, when a new database was created the tray defined one database GUID and the Jobs a different one. This would result in the Jobs not executing properly until the tray was closed and launched again. This has been fixed in 7.2.8.
  • In the event viewer on a 64-bit Endpoint client machine it was possible to receive a “Could not load file or assembly ‘SPIVssLib_2003.dll’ message when the Agent service started. This has been fixed in 7.2.8.
  • When an Endpoint job started, a “thread was being aborted error” could be produced. This has been fixed in 7.2.8.
  • If a user launched the Endpoint Client and terminated it while the Jobs were still starting it was possible to have those Jobs remain in Task Manager. This would result in duplicate sets of SyncFiles running. A change has been made for the client to wait for pending jobs to cancel completely before shutting down. This has been fixed in 7.2.8.

File Locking

  • In non-English versions of Windows, it was possible to receive an “unable to login as network service to access the UNC share list needed for file locking error. This has been fixed in 7.2.8.
  • When a file was encountered that had a reparse point and no primary data stream, the file copy could hang. This has been fixed in 7.2.8.

File Synchronization

  • When running a Job manually, the desktop could produce a “System.NullReferenceException: Object reference not set to an instance of an object” message. This has been fixed in 7.2.8.
  • Performance enhancements related to file transfers. This has been fixed in 7.2.8.
  • It was possible to receive a “no direct connection is available from AGENTNAME to AGENTNAME. Root path cannot be accessed or does not exist.” error when a Job was idle for long periods of time. This has been fixed in 7.2.8.
  • During a file resume, it was possible to receive a “System.ArgumentNullException: Value cannot be null. Parameter name: second. This action will not be retried” error message. This has been fixed in 7.2.8.
  • The timeout setting on the sending of a block has been increased from one minute to 5 in order to deal with very slow and/or high latency networks. In these environments, better performance can sometime be achieved by lowering the overall number of copy threads. This has been fixed in 7.2.8.
  • When an invalid change journal checkpoint was attempted to be used by a Real-Time Monitor, a “the parameter is incorrect” error could occur resulting in a path
    drop. A change has been made to automatically detect this error and trigger a full scan which is the necessary step to resolve. This has been fixed in 7.2.8.
  • In some instances, a path recovery could be logged as a trace level log entry. This would result in the recovery not showing on the default log viewer filter level. This has been fixed in 7.2.8.

SQL Protection

  • A SQL Protection backup could timeout if the backup took longer than 10 minutes to complete. This has been fixed in 7.2.8.
  • A SQL Protection restore could timeout if the restore took longer than 5 minutes to complete. This has been fixed in 7.2.8.
  • In SQL Protection, the ability to define compression for the transmission of files has been added into the user interface. This option can be found on the SQL Destination tab. This has been fixed in 7.2.8.
  • In SQL Protection, changes have been made to restore log file backups in order if a connection loss occurred during an execution of a schedule. Previously, only the most recent log backup would restore resulting in an error until the next full was performed. This has been fixed in 7.2.8.

User Interface

  • When permissions errors were encountered in the rule subfolder(s) browse, the error was not presented correctly to the user. This has been fixed in 7.2.8.
  • If an Archive Rule was defined with no intervals, the Desktop would allow the Rule to be saved and defined in a Job. An Archive Rule with no intervals is invalid. The Rule defined for the Job would disappear. Changes have been made to the Desktop to prompt for intervals to be defined so a valid Archive Rule is created. This has been fixed in 7.2.8.
  • The agent configuration utility has been updated to allow a timeout period of up to 10 minutes. Previously, the maximum value was 5 minutes. This has been fixed in 7.2.8.
  • When configuring a Schedule with “Run Schedule on Selected Days,” the “Number of times to retry a failed Schedule” option was not present. This has been fixed in 7.2.8.
  • Detailed Status could appear to stop updating when using the “display all locks” or “display only blocked users” options. This was due to status message buffers being sent larger than expected. This has been fixed in 7.2.8.
  • When adding a path to an existing Job, it was possible to receive a “System.NullReferenceException: Object reference not set to an instance of an object.” error message. This has been fixed in 7.2.8.


Return
 to top of page

SureSync 7.2.6 Known Issues

Version 7.2.6 was posted on February 24, 2015.

Communications Agent

  • When using deltas and compression, it was possible for a file copy to produce a “file block offset invalid” error. This has been fixed in 7.2.7.
  • When using delta processing where a Communications Agent needed to route through the SureSync machine to reach another Communications Agent, it was possible to receive “file block sequence” errors. All agents must be updated to receive the fixes. This has been fixed in 7.2.7.
  • Block out of sequence type messages could occur during delta processing. To receive the fix, all agents in the environment must be updated. If a new agent encounters an old agent, deltas will be suppressed. This has been fixed in 7.2.7.

Endpoint

  • With Endpoint, when configuring e-mail alerts the clients would fail to send the alert correctly with an error “SMTP Profile not found for [Name of Profile]. This has been fixed in 7.2.7.
  • Endpoint has been enhanced to allow Jobs on a client to start quicker than the 10 second to 5 minute random interval if server load is low. This allows client jobs to start faster when resources are available. Both the server and clients need to be updated to benefit from this change. This has been fixed in 7.2.7.

File Synchronization

  • Performance optimizations have been made when running a large amount of file copies in parallel either via multiple Jobs or many threads in a single Job. The maximum active copies is now set to 40 with the remaining requests going into a “delay queue” to be processed as a spot in the 40 opens up. A new status message will now appear in Detailed Status when file copies are delayed in this queue. The bytes copied column will report “(delayed)” in this situation. This cap prevents performance problems caused by overloading the machine with requests. This has been fixed in 7.2.7.
  • When processing renames in a multi-directional Rule with AOFS, history could get deleted too early resulting in a file being added back. This has been fixed in 7.2.7.
  • When encountering a path drop, it was possible to receive a secondary “NullReferenceException” error. This has been fixed in 7.2.7.
  • When processing a large full folder scan that was broken into sub-scans, it was possible for a sub-scan request to not be saved because the a parent folder scan was already queued. This has been fixed in 7.2.7.
  • If a Schedule or Real-Time Monitor was defined to use the default alert level, the default could end up not being used and alerts could be sent on each detail event. This has been fixed in 7.2.7.
  • Corrections were made to the compare rule. This has been fixed in 7.2.7.
  • If a file was created and quickly moved to another folder location, the file could incorrectly end up in both the old and new location. This has been fixed in 7.2.7.
  • When using “Allow user access to source file being copied” it was possible to receive a file in use error when doing renames of source files. This has been fixed in 7.2.7.
  • Reporting on file resumes has been enhanced. In the Log Viewer, when a resume is used you will see a “File resume used starting at location message. Copy status has also been enhanced to reduce the size listed of the file copy by the amount already processed. This has been fixed in 7.2.7.
  • Corrections were made in timing situations where blocks could be received out of order. This has been fixed in 7.2.7.
  • Minor licensing fixes related to expired trials and exceeding licenses. This has been fixed in 7.2.7.
  • When processing rapid folder renames, it was possible to get unexpected results. A fix was made to delay processing of folder renames if an active copy was being performed in a subfolder. This has been fixed in 7.2.7.

Installer

  • The /XMLPath installer switch for SureSync Endpoint and SyncLockStatus used to retrieve a configuration file from a network location was not being processed correctly by the tray applications. This has been fixed in 7.2.7.

SQL Protection

  • In SQL Protection, if a Schedule was configured to perform only Full and Log backups, the Log backups were not correctly scheduled. This has been fixed in 7.2.7.
  • With SQL Protection, a schedule could get stuck on “waiting for tasks to complete” if an error or exception was encountered. This has been fixed in 7.2.7.

SyncLockStatus

  • Corrections were made to relaying SyncLockStatus clients through a Communications Agent to the main SureSync machine with the Scheduler. This has been fixed in 7.2.7.

User Interface

  • In the SureSync Desktop, renaming a Schedule would not re-sort the Schedules in the tree view. This has been fixed in 7.2.7.
  • In the SureSync Desktop, when cloning Schedules with a copy/paste it was possible for the Schedule to not have a run time rescheduled. This has been fixed in 7.2.7.

Return to top of page

SureSync 7.2.5 Known Issues

Version 7.2.5 was posted on February 13, 2015.

Endpoint

  • With Endpoint, in rare timing situations it was possible for blocks to be processed out of sequence when retrieving a database from the server. This could potentially result in jobs stopping/starting and SyncFiles.exe processes being left in Task Manager. This has been fixed in 7.2.6.
  • In Enterprise Status, disabling an Endpoint monitor would not be accurately reflected on the SureSync desktop. This has been fixed in 7.2.6.

File Synchronization

  • If a job started when it was already busy (not a normal condition), the job correctly aborted. However, the instance of SyncFiles was left in memory. This was most likely to occur within Endpoint. This has been fixed in 7.2.6.

SQL Protection

  • In SQL Protection, it was possible for the Scheduler to reschedule a newly configured Schedule to run a differential backup before a full. This would result in errors on execution about not having a full backup. This happened mainly with “day of week” filters where the day the Schedule was taken off “held” status was defined to run a differential and no full had previously been performed. This has been fixed in 7.2.6.
  • In a rare situation, where a SureSync job had previously referenced a machine now used only with SQL Protection it was possible for the licensing to still request a regular SureSync server license. This license was not needed because no SureSync job was being run involving the machine being protected by SQL Protection. This has been fixed in 7.2.6.

Return to top of page

SureSync 7.2.4 Known Issues

Version 7.2.4 was posted on January 16, 2015.

Communications Agent

  • Various performance optimizations have been made to delta processing. This has been fixed in 7.2.5.

Endpoint

  • With Endpoint, if a client was running multiple Jobs and only one of those Jobs lost connection to the server it was possible for the failed job to not be restarted. This has been fixed in 7.2.5.
  • When an Endpoint Client was launched and a Scheduler was not running on the server, the client would wait 10 minutes to attempt to reconnect. This time has been shortened to between 3 and 5 minutes to allow quicker recovery of the Jobs on the client. This has been fixed in 7.2.5.
  • When an Endpoint Client was launched and a Scheduler was not running on the server, the client would not fire the timer to attempt a reconnect until the form was displayed. This could result in jobs not starting until the next time the client was launched or until the client user interface was loaded. This has been fixed in 7.2.5.
  • In Endpoint, if you activated a client that was previously deactivated the newly processed activation would not be detected by the client until the Scheduler service on the server had been restarted. This has been fixed in 7.2.5.

File Locking

  • In certain timing conditions with file locking, a lock could incorrectly not be released. This involved a file being closed and quickly opened again. This has been fixed in 7.2.5.

File Synchronization

  • When a request to an API took longer than 10 minutes to complete, it was possible to receive an error about a timeout occurring to the local agent. This is most likely to occur with the APIs to copy and set security. The request would complete on the worker thread performing the action but the error would still be logged. This has been fixed in 7.2.5.
  • A “the transaction associated with the current connection has completed but has not been disposed” error could occur on a SQL transaction that timed out due to an unusually high number of updates to commit to the database. This would cancel the job that encountered the error. This has been fixed in 7.2.5.
  • When performing a directory enumeration that takes longer than 2 minutes to complete, SureSync will send the completed results buffer for processing. This would only occur on machines having performance issues on reads, mostly with security copying enabled. When this occured, SureSync would not resume the enumeration of that directory which could result in files not being copied. This has been fixed in 7.2.5.
  • The compare rule was not functioning correctly. This has been fixed in 7.2.5.
  • Open file retries with a Schedule were getting reported incorrectly as exceptions even though the files were being copied. This has been fixed in 7.2.5.
  • If more than 180 file copies were active to or from a particular Communications Agent, an array used in status could overflow a buffer. This would result in copies hanging and jobs being canceled. This primarily impacts Endpoint in large environments with hundreds of clients sending to the same server agent at the same time. This has been fixed in 7.2.5.
  • If a Real-Time Monitor was run that contained both folder and file reparse points, messages would only be presented for the folders. This has been fixed in 7.2.5.
  • The message presented in Real-Time Monitors encountering a file reparse point were logged incorrectly as a warning when they should have been an exception. This has been fixed in 7.2.5.

User Interface

  • In the licensing panels, the activations panel was incorrectly adding the “activated” and “deactivated” counter together into the “activated” counter. The main licenses panel had the correct count. This has been fixed in 7.2.5.

Return to top of page

SureSync 7.2.3 Known Issues

Version 7.2.3 was posted on December 26, 2014.

System Requirements

  • The .NET Framework 4.5.2 is now required and will be installed by the installer if it is not found on the system. It is possible the Microsoft prerequisite installer launched by the SureSync installer will prompt for a reboot if .NET code is in use on the system. This would be the same as installing the .NET update with Windows Update or from Microsoft’s web site. This has been fixed in 7.2.4.

Endpoint

  • For SureSync Endpoint, optimizations were made to the handling of hundreds of incoming requests to avoid the .NET Framework from dropping connections incorrectly. This has been fixed in 7.2.4.

File Synchronization

  • In a rare timing condition, a NullReferenceException could occur in get_ServiceConfigsRow that would result in a file copy being canceled. This has been fixed in 7.2.4.
  • Corrections were made to tracing when requesting a driver trace. If requested, both the driver and journal trace were not being generated correctly. The journal trace would generate if the driver trace was off. This has been fixed in 7.2.4.
  • A correction was made to recognize a specific path loss condition where the path between two agents was lost but not the path from the Job. This has been fixed in 7.2.4.
  • A warning “Unexpected error processing file. The data is invalid.” could occur when a file changed on the destination side before SureSync was done with a file copy from the source. This has been fixed in 7.2.4.
  • Changes were made in the multi-threading of the File I/O service to allow it to handle a larger number of active connections without issue. Previously, when hundreds of connections were activat a time, File I/O timeouts could occur. This has been fixed in 7.2.4.
  • A harmless “System.NullReferenceException: Object reference not set to an instance of an object.” error could be logged in the Application Event Viewer when losing a path. This would be from SPIFileJrnlLib.Job.WorkerThread(). The error had no impact on the Job but could clutter an event viewer. This has been fixed in 7.2.4.
  • When the “allow user access to source file being copied” option enabled with a one-way Rule, you could delete a source file while that file was actively copying and the file would not be cleaned up until the next full scan. This has been fixed in 7.2.4.
  • In certain cases, an Advanced Open File Support (AOFS) copy would fail to schedule a retry when copying out of an existing VSS snapshot. This would result in the last update to the file not being copied to the destination until the next change event occured. The retry will now occur at the expiration of the snapshot ensuring that all files copied out of the snapshot are completely processed and current. This has been fixed in 7.2.4.

Return to top of page


 

SureSync 7.2.2 Known Issues

Version 7.2.2 was posted on December 5, 2014.

Communications Agent

  • Corrections were made to recognize a specific path loss condition where the path between two agents is lost but the connection from the machine running the Job to the Communications Agents in question is online. This has been fixed in 7.2.3.

File Locking

  • In specific timing situations, it was possible for a file lock to fail to be recognized by an agent. This would only happen when a Monitor was just starting up. This has been fixed in 7.2.3.
  • If multiple Monitors with locking overlapped a path, it was possible for an “unexpected error in LockListService. System.NullReferenceException: Object reference not set to an instance of an object.” error to occur which would cancel the monitor. This has been fixed in 7.2.3.

File Synchronization

  • From various components including the Scheduler, it was possible to receive a “System.OverflowException: Arithmetic operation resulted in an overflow” error. This has been fixed in 7.2.3.
  • If you have multiple file copies active from the same source location and the file copy takes more than 5 minutes, the file copy on a path may be terminated indicating it is unresponsive, thinking there is no progress in the file copy. This does not happen if there is a single destination for a file copy. This has been fixed in 7.2.3.

Licensing and Activations

  • Activations for SyncLockStatus and SureSync Endpoint Clients were not being correctly enforced, allowing these clients to run past the 14 grace period without requiring the license be activated. After upgrading to 7.2.3, if using these products please go to the Licenses tab and process any required SyncLockStatus and Endpoint activations. This has been fixed in 7.2.3.

User Interface

  • In the SureSync Desktop on the Attributes tab of a Rule, checking “If Rule Applies Force On” for a particular attribute was not being saved correctly. After running a Job, the value would reset to unchecked. This has been fixed in 7.2.3.
  • In the SureSync Desktop on the Attributes tab when unchecking the “Force attributes on and off only on files that are copied” option a full scan will be scheduled. This is required for the files that have already been processed by a Change Journal enabled job to have the “force on” for the attribute applied. Otherwise, a full scan had to be manually specified after the change. This has been fixed in 7.2.3.
  • When older configuration data was merged into a configuration dataset in the Local Communications Agent Config Utility, it was possible to receive an error indicating a column cannot contain a null value. This has been fixed in 7.2.3.

Return to top of page

SureSync 7.2.1 Known Issues

Version 7.2.1 was posted on September 15, 2014.

ArchiveSync

  • When restoring folders from an archive using a Restore Job, the folder security values in the archive may not be restored. This has been fixed in 7.2.2.

Communications Agent

  • An error was present in the delta savings percentage statistic logged in the log viewer for jobs processing deltas. This would result in the percentage saved showing 0% even though deltas were used properly. This has been fixed in 7.2.2.
  • Changes were made to optimize ping processing and other communications processes within the Communications Agent when dealing with large numbers of paths (50+). The agent could get stuck processing pings and tests to lost connections that could cause other requests to timeout. This was most obvious in Endpoint with more than 50 clients where jobs could cancel and restart repeatedly on the client side due to connection issues. This has been fixed in 7.2.2.
  • If machines involved in a delta copy did not have a direct connection to each other, a path drop could occur while processing the delta. This has been fixed in 7.2.2.
  • In SureSync 7, Communications Agent names must be the NetBIOS name of the computer. In v6, fully qualified domain names were allowed in some situations. Code has been added to automatically convert the FQDNs to NetBIOS names when encountered to ease upgrades. The access name can still be a FQDN and will be left alone during an upgrade. This is just the agent name which shows in the drop-down in Communications Agent configuration panels and as the agent name for root paths. This has been fixed in 7.2.2.
  • When copying between two remote Communications Agents that do not have access to each other, the communication is supposed to automatically “fall back” to going through the SureSync machine to reach the agents. This was not always working properly which would result in file copy actions that started and stuck at 0%. Additional symptoms were possible. This has been fixed in 7.2.2.
  • It was possible to receive NullReference Exceptions from the I/O service for a Communications Agent. This has been fixed in 7.2.2.

Databases

  • Errors were reported when copy a SQL database to an Access database. This has been fixed in 7.2.2.
  • Improvements were made to speed recovery from a remote SQL server powering the SureSync database being offline. It is recommended to have the job lock folder be on a machine other than the remote SQL server for best results. This has been fixed in 7.2.2.12.

Endpoint

  • With Endpoint, the clients will now stager their launch time. Previously, if a server service was restarted or a configuration change was made all the clients would shut down at once and start back up at basically the same time. This could result in server side timeouts. Now each client will randomly wait a period of time between 1 second and 5 minutes before starting their Jobs. This will reduce load on the server and scale better. This has been fixed in 7.2.2.
  • Optimizations have been made to the Endpoint Client to address timeouts to the server which could result in jobs stopping. This has been fixed in 7.2.2.
  • In SureSync Endpoint, it was possible to encounter a “System.Data.RowNotInTableException” from the Scheduler service. This has been fixed in 7.2.2.
  • When modifying a Rule in SureSync Endpoint, you could be prompted to “please select a source root path for this Rule” when saving the change. The change would correctly save anyway and the source was already correctly defined. This has been fixed in 7.2.2.
  • In rare circumstances on the SureSync Endpoint tab, it was possible to receive a “you have unsaved changes” prompt when clicking between monitors when there was no change. This has been fixed in 7.2.2.
  • Filters have been added to the Log Viewer to allow displaying log entries from only a specific Endpoint client. This has been fixed in 7.2.2.
  • When displaying an Endpoint Monitor, Detailed Status has been enhanced to provide the ability to show only the statistics and actions for a specific client. This is accessed by a drop-down menu in Detailed Status. This has been fixed in 7.2.2.
  • When using an Access database on the server side, it was possible for an Endpoint Job to not correctly create all required database records for a new client being added. This would result in a “the given key was not present in the dictionary” error. This issue did not happen with a SQL database. This has been fixed in 7.2.2.
  • Endpoint Monitors remained in the Enterprise Status display after being deleted until a Communications Agent service restart occurred. This has been fixed in 7.2.2.
  • The Endpoint client now has additional status information to notify the end user when a job is stopped or when a job is pending start. This has been fixed in 7.2.2.

File Synchronization

  • When processing files through a reparse point, the directory entry would report that a file was 0KB and later the test would show the real size of the file. This would result in a “source file changed during processing of an open file” error which would fail again keeping the file from processing. This has been fixed in 7.2.2.
  • If you had Jobs configured using UNC paths and added a Job that synchronized files directly between two remote machines using the Communications Agent, it was possible for the program to incorrectly ask for a Communications Agent license on the main server where it was not being used. This has been fixed in 7.2.2.
  • With a Schedule, a “source file changed during processing of an open file. The file or folder was not processed and will be retried on the next scheduled run” error could happen continually on files being frequently processed with VSS. This has been fixed in 7.2.2.
  • Changes were made to prevent incorrect path drops when running UNC path based schedules that involved 10 or more paths. The number of paths combined with slow response times from some machines on the network could make SureSync drop the local agent connection. This has been fixed in 7.2.2.
  • In some cases, custom connection configuration information may not be used. For example, if a Job is running on a remote machine and you request Detailed Status this request could fail in environments with custom connection configurations. This has been fixed in 7.2.2.
  • When enumerating a folder with hundreds of files on a machine with degraded file I/O performance (for example, when a drive is 100% busy) and the SureSync job was collecting security values it was possible for the collection of directory information to take too long to complete. This would result in a timeout that would drop the path. This has been fixed in 7.2.2.
  • If a job copies a zero byte file followed by files that contained named data streams, it was possible in rare instances for the named data streams to not be copied. Named data streams are rare, as well as zero byte files. This has been fixed in 7.2.2.
  • Changes have been made to advanced open file support processing to address issues encountered with Excel files. Excel changes dates on files when they are opened and then sets them back to the old date stamp if nothing has changed. This could result in an open file copy not being processed. This has been fixed in 7.2.2.
  • An “incorrect function” error could occur in an FTP job interacting with specific types of FTP servers. This has been fixed in 7.2.2.
  • Additional changes have been made to handling of reparse points. A reparse point on a file can cause “source file changed during processing or an open file” warnings. Also, code has been added to produce warnings if reparse points are in the folder structure being processed by a Change Journal enabled Job. Change Journal events happen in the actual location of the file, not the reparse point so actions on those reparse points are only seen during a full scan. If using reparse points, a Schedule with Change Journal support disabled is recommended. This has been fixed in 7.2.2.
  • Optimizations were made to avoid timeouts when processing Jobs with UNC paths. These timeouts could result in “[name of local agent] failed to receive a response from ‘[name of local agent]’ after x minutes” messages. In some environments with slow system response times when accessing via UNC, creating a custom priority with smaller block sizes might also help. This has been fixed in 7.2.2.
  • Changes were made to prevent a job from hanging on “waiting for tasks to complete.” This could occur when the delayed requests queued internally were already at maximum before a path went offline. This has been fixed in 7.2.2.
  • System.IO.IOException: Incorrect function errors were possible during a synchronization job. This has been fixed in 7.2.2.

Preview

  • Performing a Preview of a Schedule would switch the status of the Schedule to completed and not perform a reschedule action if the Preview was canceled. This would result in a Schedule that did not execute until the user placed it on hold and took it off hold to trigger the recalculation of the next execution time. This has been fixed in 7.2.2.

Scheduler

  • When defining a failover partner for SQL mirroring, the connection string used by the SureSync Scheduler was not properly having the failover partner defined in the connection string resulting in problems when the primary database went offline. This has been fixed in 7.2.2.
  • Fixes were made to multiple Scheduler service environments. When access to the database was disrupted (for example a connection failure on the SureSync machine running the Job), running Jobs did not terminate in a timely manner which caused incorrect behavior such as “job is already running” type messages when the other machine attempted to start the same Job. This has been fixed in 7.2.2.

SQL Protection

  • A SQL Protection Job could be incorrectly canceled with a licensing error when appropriate licensing was present. This has been fixed in 7.2.2.
  • In the SQL Protection section of the SureSync Desktop, the right click menu under New incorrectly showed Schedule instead of Job. In SQL Protection, both are created in a single Job wizard. This has been fixed in 7.2.2.
  • With SQL Protection, in some instances it was possible for a full backup to encounter a “no destinations are available for this action” error at the end of the Job when doing restores. This has been fixed in 7.2.2.

User Interface

  • In the SureSync Desktop for Endpoint, if you deleted all paths on the “Client Root Paths” tab and clicked apply you would be presented with a warning that at least one path had to be defined. This warning would repeat each time you clicked apply even after adding a path. This has been fixed in 7.2.2.
  • When using the Replace Machine References option under Home | Communications Agents it was possible to receive a “cannot insert duplicate key” error message when trying to do a replace. This has been fixed in 7.2.2.
  • A “System.Data.OleDb.OleDbException: Syntax error (missing operator_ in query expression” error could occur when switching between tabs in the SureSync Desktop. This has been fixed in 7.2.2.
  • When browsing a remote agent, the browse panel could return a “System.ArgumentNullException: Value cannot be null” error. This has been fixed in 7.2.2.
  • Adding or changing a path in the Job Trace Folder or User Interface Trace Folder fields on the Logging tab of Options would not enable the “Save and Close” or “Apply” buttons. You would have to click onto another field to enable the buttons. This has been fixed in 7.2.2.
  • A splash screen has been added to Detailed Status. When receiving a slow response time from a status service, it was possible for it to not be obvious that the program was launching and waiting for information. This has been fixed in 7.2.2.
  • If you set a file size filter on a Monitor Interval and later went back and changed the Interval file size filter to “All Files,” the change would not save. This has been fixed in 7.2.2.
  • If you create a new Restore Job and select the “Run Now” option when the Restore Job Wizard is completed, the Job did not run. To execute the Job, the ribbon bar needed to be used. This has been fixed in 7.2.2.
  • In the SureSync Desktop, a Restore Rule incorrectly duplicated the Restore Options tab. This has been fixed in 7.2.2.
  • It was possible to receive a “please select a source root path for this rule” message on a Job. The Job would not process data because the source path was not defined. This has been fixed in 7.2.2.
  • When using the “Replace Machine References” function to substitute one Communications Agent name for another, a “violation of primary key constraint” error could occur. This has been fixed in 7.2.2.
  • A System.NullReferenceException: Object reference not set to an instance of an object error could occur when browsing a remote agent. This has been fixed in 7.2.2.12.
  • On the Activations panel, with a long list of machines it was possible for checking a box for “activate” or “deactivate” to incorrectly check multiple machines. This has been fixed in 7.2.2.
  • After processing an activation or deactivation, if you clicked on Home and loaded the Communications Agent Configuration Utility it was possible to receive an incorrect “you have unsaved changes” message. This has been fixed in 7.2.2.
  • Deleting a Rule from a Job under a Schedule would not always refresh the display properly. This has been fixed in 7.2.2.

Return to top of page

SureSync 7.1.13 Known Issues

Version 7.1.13 was posted on July 22, 2014.

ArchiveSync

  • If you use the ArchiveSync option to create an archive and use open file support, the same file could be copied many times if the file changes while the copy is in progress. This has been fixed in 7.2.1.

Communications Agent

  • If two remote Communications Agent machines were unable to directly communicate with each other, the Job would drop the path instead of correctly route the communication through the SureSync machine. This has been fixed in 7.2.1.
  • In rare circumstances, a Communications Agent could produce an “object reference not set to an instance of an object” error logged into the Application Event Viewer that could result in a path drop. This has been fixed in 7.2.1.
  • Occasionally, a certificate generated by a Communications Agent would be damaged and result in the service being unable to start. Another symptom of this problem is a System.Security.Cryptography.CryptographicException: Bad Data” error in the event viewer. The Communications Agent now performs tests on its certificate and will automatically generate a new one if needed. This has been fixed in 7.2.1.
  • When processing files with deltas, a “System.ArgumentException: Value does not fall within the expected range” error message that would cancel the job. This has been fixed in 7.2.1.

Endpoint

  • SureSync Endpoint, a powerful new solution in the SureSync Data Protection & Availability Suite has been released. This product can be used to protect, synchronize and share data with remote laptops and workstations where name resolution from the server side is not present.

File Synchronization

  • Some Schedules could hang at the end a Job if a scan was pending. This has been fixed in 7.2.1.
  • Truncate deletes were not properly being used in one-way Rules. This could result in folders with hidden files in them not being deleted completely. This has been fixed in 7.1.14.
  • For newly created Jobs, the default file masks has been changed to *.*. This tells SureSync to copy everything out of the box. Previously, some types of temporary files were excluded (-*.tmp | -~*.doc | -~$*.doc? | -~$*.xls? | -~$*.ppt?). This has been fixed in 7.2.1.
  • For newly created Jobs, the default on the Attributes tab have been changed to include Hidden and System files. Previously, these were excluded. Excluding these file types can result in issues like folders not deleting completely because they contain a hidden Thumbs.db. The excluded file makes the directory not empty and prevents the folder delete. This has been fixed in 7.2.1.
  • A “Open failed. FileIOGuid not returned for open for path” error message could be produced when copying a file. This has been fixed in 7.2.1.
  • No encryption or non-default encryption configurations could drop the source path when attempting to copy a file. This has been fixed in 7.2.1.
  • In some cases the number of actions queued could increase to hundreds or thousands. This should normally be limited to 50 since higher numbers do not improve performance and can increase paging rates. This has been fixed in 7.2.1.
  • In some cases with FTP Jobs, a Rebex.Net.FtpException (550) error on a folder could cause the Job to cancel. This has been fixed in 7.2.1.
  • With FTP Jobs, a “System.NullReferenceException” error could occurring during folder creations. This has been fixed in 7.2.1.
  • In some circumstances, a Schedule could execute Jobs out of the defined order. This has been fixed in 7.2.1.
  • If only copying ownership on the Copy Security tab and SureSync determined that ownership already matches, an “unable to copy security objects” error message could incorrectly be produced. This has been fixed in 7.2.1.
  • Optimizations have been made to the block sizes used for communications on different network conditions. This results in better performance. However, in Windows 2003 and XP memory usage for the SPIComService.exe process can appear higher when using a block size greater than 64KB. This occurs because of .NET 4.0’s memory manager. The memory is not actually in use and will be released to other processes if needed but the memory used in Task Manager can look larger. This is harmless but can raise questions. The memory increase can be avoided by using a priority with 64KB block sizes (or smaller). This has been fixed in 7.2.1.
  • FTP timeouts on block writes have been increased to allow more reliable synchronization with slower, older FTP servers. This has been fixed in 7.2.1.
  • When copying a file via FTP that was only a single buffer in size, an “incorrect function” error could be generated. This has been fixed in 7.2.1.
  • In rare timing related circumstances, a “Retry path. The file or folder was not processed and will be retried on the next scheduled run” exception could be logged. When this occured, the file would be added to retries queued and not be processed. The copy would be retried on the next execution. This has been fixed in 7.2.1.

Preview & Trial Run

  • An agent name column has been added to the Trial Run output to more clearly note paths in the output file. This has been fixed in 7.2.1.
  • If you entered a path in the “Trial Run Output Folder” in Options, save it and then clear it Trial Run would fail in future executions. This has been fixed in 7.2.1.
  • Trial Runs would not be generated for Jobs with \ / : * ? “ < > | characters in the file name. These characters are not allowed in file names. The Trial Run will now replace illegal characters with underscores in the file names. This has been fixed in 7.2.1.

System Requirements

  • .NET 4.5.1 is required when running Windows 2008 / Vista and newer. This version has better .NET memory management which results in smaller memory usage for the Communications Agent service. 2003/XP will use the old .NET 4.0 Framework for the Communications Agent service.

User Interface

  • The SureSync Desktop could produce “System.Collections.Generic.KeyNotFoundException” pop-up error messages after upgrading a SureSync database to v7. This has been fixed in 7.2.1.
  • “Unable to cast object of type” error messages in the SureSync Desktop were addressed. This has been fixed in 7.2.1.
  • When clicking on the “Remove All” button in a Local Communications Agent Config Utility could produce a “ForeignKeyConstraint” error. This has been fixed in 7.2.1.
  • In the Local Communications Agent Config Utility, changing the Encryption type would not save correctly preventing the use of non-default encryption types. This has been fixed in 7.2.1.
  • An “object reference not set to an instance of an object” error could occur in the Communications Agent Configuration panel in the SureSync Desktop. This has been fixed in 7.2.1.
  • The client configuration form, displayed in Endpoint, SyncLockStatus and Enterprise Status could fail to enable the “save” button to allow a save of changes after the dialog has been used once. This has been fixed in 7.2.1.
  • In the SureSync Desktop, switching between tabs on the top of the interface in a new database would fail to fully enable the newly created Group Folder until you switched to another tab and back. This has been fixed in 7.2.1.
  • Enterprise Status and the SureSync Desktop have been enhanced with new icons for SQL Protection and Endpoint. The different icons in Enterprise Status help differentiate between the types of Jobs being displayed. This has been fixed in 7.2.1.
  • When using the Remote Installation Utility, it was possible to receive a “ERROR_PATH_NOT_FOUND” error. This has been fixed in 7.2.1.

Return to top of page

SureSync 7.1.12 Known Issues

Version 7.1.12 was posted on July 2, 2014.

Communications Agent

  • When SureSync is installed on a machine controlling a synchronization between a source and a destination on machines remote to the SureSync machine, it was possible to have an incorrect path drop if the remote source could not directly connect to the remote destination. The path drop would occur only on the first run and then work correctly until the Communications Agents are restarted. At that time, it would drop the path one time and be fine again. The direct connect traffic was not being correctly automatically routed through the SureSync machine on the first attempt. This has been fixed in 7.1.13.
  • An internal System.NullReferenceException error could occur transferring data between Communications Agents. This problem could surface in a couple of ways including files in the Retries Queued counter in Enterprise Status with a reason “RetryPath” or symptoms such as frequent path drops. This has been fixed in 7.1.13.
  • An issue with delta processing to multiple locations with extremely large files was addressed where the processing could timeout and the file would not be processed. This has been fixed in 7.1.13.

File Synchronization

  • When running on a machine with FIPS compliant algorithms enforcement turned on, a “System.InvalidOperationException: This implementation is not part of the Windows Platform FIPS validated cryptographic algorithms.” error would be produced. Corrections have been made to some internal hashing algorithms to be fully FIPS compliant. This has been fixed in 7.1.13.
  • FTP Jobs could encounter a “error repositioning file” error. The error actually related to a permissions error with creating folders and had nothing to do with a file. Changes have been made to the path validation and creation FTP routines to correctly detect the error and report it. This has been fixed in 7.1.13.
  • On extremely fast hardware, SureSync could process folder deletes so quickly that the file system had failed to release all handles to the folders being deleted. This would result in a “temporarily unable to delete file or folder” error and the folder not being deleted. Internal retries have been added on this error to allow the handles to fully clear. The retries allow the folders to be deleted correctly the first time. This has been fixed in 7.1.13.
  • A change has been made to the default Attributes settings for new Jobs. Previously, files and folders with system and hidden attributes were excluded from processing. System and hidden folders and files will not be processed by default. These can be excluded by checking the checkbox for “Exclude Rule If” on the Attributes tab for these types. This change ensures all file types are included by default. This has been changed in 7.1.13.

User Interface

  • When launching the SureSync Desktop a “ForeignKeyContraint DesktopFoldersRelationRoots requires the child key value (2) to exist in the parent table” error message could be presented. This has been fixed in 7.1.13.
  • In the SureSync Desktop, the Scheduler Services panel (used to install a Scheduler Service) could terminate if you clicked the Install button and then the Cancel button without entering anything. This has been fixed in 7.1.13.
  • In the SureSync Desktop, when editing a Real-Time Monitor Interval that had a “Smaller than” setting for the file size filter you would be unable to change the setting to “equal to or larger than” and have that setting change save. This has been fixed in 7.1.13.
  • In some instances, changes made in the Local Communications Agent Config Utility would fail to save. This has been fixed in 7.1.13.

Return to top of page

SureSync 7.1.11 Known Issues

Version 7.1.11 was posted on June 10, 2014.

Database

  • During a database upgrade, it was possible to receive a “System.Data.ConstraintException: Failed to enable constraints” error that would halt the upgrade. This has been fixed in 7.1.12.
  • SureSync’s SQL database support has been enhanced to allow using SQL servers using only SQL authentication. This has been fixed in 7.1.12.

File Synchronization

  • In rare circumstances, an incorrect connection configuration could be used by a Communications Agent when a Job is providing connection values instead of a local configuration file. This could cause file copies to not be processed and then a path to drop. This has been fixed in 7.1.12.

User Interface

  • It was possible to receive an incorrect “You have unsaved changes” warning message when clicking between tabs on the Local Communications Agent Config Utility. This has been fixed in 7.1.12.
  • When using the import Communications Agent Configuration functionality, it was possible to receive a “Column name ‘Access Name’ was not recognized” error which would halt the import. This has been fixed in 7.1.12.

Return to top of page

SureSync 7.1.10 Known Issues

Version 7.1.10 was posted on May 30, 2014.

File Synchronization

  • One of the SureSync components involved in FTP processing was looking for an old version of the FTP engine dll files. This could cause path drops accessing FTP paths. This has been fixed in 7.1.11.
  • If a full scan was occurring while a path was offline, it was possible for the scan to hang. All paths need to be online for a full scan to complete successfully but the scan should never hang. This has been fixed in 7.1.11.
  • A “SetCredentials request failed. System.InvalidOperationException: DataTable Internal index is corrupted: ‘5’” error could occur. This has been fixed in 7.1.11.
  • A fix was made to SyncFiles to use a culture sensitive compare for non-English languages. The compares would be made using the current culture settings for the operating system. This has been fixed in 7.1.11.
  • When encountering an “out of space” error on a server, the path was not being identified which made determining which path was out of storage difficult in Jobs with many paths. The path with no space will now be properly logged. This has been fixed in 7.1.11.
  • Performance improvements have been made to the FTP engine. This has been fixed in 7.1.11.
  • When running a FTP job, a System.NullReferenceException: Object reference not set to an instance of an object” error could occur when a timeout was encountered. This has been fixed in 7.1.11.
  • When running a FTP job, a “command not understood” error could be encountered. This has been fixed in 7.1.11.
  • A scan running for more than one hour is split into sub-scans. The additional subfolder scans were incorrectly being scheduled for the next execution of the schedule. With a UNC path based Schedule, this would result in the full scan starting over again and potentially the broken up subfolder scans not being processed. This has been fixed in 7.1.11.

Scheduler

  • In certain situations, a rescan of the scheduler queue would attempt to start the same job multiple times. Since the same Job can only be active once, this would produce errors in the extra executions of the Schedule. This has been fixed in 7.1.11.

User Interface

  • On new SureSync installations, it was possible to receive the message “An internal application error has occurred. Error number 1026. The configuration database has not been loaded. This has been fixed in 7.1.11.
  • When clicking on the “…” button to display a browse dialog for “this rule applies to the following subfolders” you could receive a “System.ArgumentOutOfRangeException: Index was out of range” which would prevent the browse from displaying. This has been fixed in 7.1.11.
  • When clicking on the “…” button on the “Set Rule Type” Job wizard panel to display a browse dialog for “this rule applies to the following subfolders, expanding a machine would result in the dialog spinning and never returning the contents. This has been fixed in 7.1.11.
  • When clicking on the “Alert Users” button on the Alerts tab of a Schedule or Monitor, a “System.ApplicationException: InitServiceConfig was not called” error message would be presented and the form would fail to load. This has been fixed in 7.1.11.
  • When entering information into an SMTP Profile in Alerts Configuration, some of the error handling was not correct. For example, if you entered an e-mail address in an incorrect format the form would tell you that and then close without saving instead of allow you to make a correction. This has been fixed in 7.1.11.

Return to top of page

SureSync 7.1.9 Known Issues

Version 7.1.9 was posted on May 8, 2014.

Communications Agent

  • In environments with multiple SureSync machines referencing the same Communications Agent, a Communication Agent could select configuration values incorrectly from a remote machine when the local configuration values should have been used. This would result in frequent path drops. This has been fixed in 7.1.10.

File Locking

  • In rare instances, a Microsoft bug could impact the functionality of the locking driver. Please see http://support.microsoft.com/kb/975759 for more information (and hotfix details). This bug caused the Communications Agent service to crash. Changes have been made to avoid the crash and cancel the job if this Microsoft error is encountered. This has been fixed in 7.1.10.

File Synchronization

  • A truncate delete operation that took longer than 5 minutes to complete was not properly rescheduled to resume. This could cause a folder structure to only be partially deleted. This has been fixed in 7.1.10.
  • The deleted statistic at the end of a log entry or in the status panels was incorrect. The number would be higher than the actual number of deleted files due to some files without actions incorrectly being counted as deletes. This is a harmless display error. This has been fixed in 7.1.10.
  • A change was made to SyncFiles to attempt to prevent “Violation of PRIMARY KEY constraint ‘PK_FileActions_Rel_NodeID_FileName'” errors when synchronizing files with certain character sets. This has been fixed in 7.1.10.
  • When using an FTP server as a source, it was possible to receive the message “(1 retry) Abort. The file or folder was not processed and will be retried on the next scheduled run.” This would prevent the files from being synchronized. This has been fixed in 7.1.10.
  • It was possible to receive a “System.OverflowException: Arithmetic operation resulted in an overflow error” when processing file resumes where the file changed a second time on the source before the resume had completed processing. This has been fixed in 7.1.10.

User Interface

  • If you altered the subfolder list with the SureSync Desktop and saved the changes, the display of the subfolders was not updated in the left-hand tree. This has been fixed in 7.1.10.
  • Adding a server to the Scheduler list in Enterprise Status could produce a “collection was modified” error. This has been fixed in 7.1.10.
  • When adding a user to the security tab of a Group Folder or other SureSync Desktop object, you would receive an “object reference not set to an instance of an object” error. This has been fixed in 7.1.10.

Return to top of page

SureSync 7.1.8 Known Issues

Version 7.1.8 was posted on April 30, 2014.

Communications Agent

  • When attempting to use a non-default port, non-default encryption method or both it was possible for the defaults to still be used. This would create path drops and prevent data from being synchronized. This has been fixed in 7.1.9.

File Synchronization

  • In certain circumstances, when running only workstation licenses, a workstation license would not properly be allocated to a workstation machine when one was available. This would generate an error about a machine being unlicensed. This has been fixed in 7.1.9.
  • When copying security via a Job using UNC paths, it was possible to receive a “Error reading security values for a file. Security values will not be compared or copied.” error. This has been fixed in 7.1.9.
  • When running a UNC path Job with a large number of UNC root paths, it was possible for the process of enabling privileges and logging into the remote shares to delay some requests. This would cause additional requests to exceed the timeout values allocated and create unnecessary path drops. This has been fixed in 7.1.9.
  • Highly nested folders would fail to delete with truncate deletes enabled when the path was longer than 240 characters. This has been fixed in 7.1.9.
  • Corrections were made to folder delete processing to address the message “Temporarily unable to delete file or folder. The file or folder was not processed and will be retried on the next scheduled run” messages. This has been fixed in 7.1.9.

User Interface

  • The “name” and “description” fields in the new Group Folder wizard were grayed out incorrectly. This forced you to create the new Group Folder with the default generated name and then change it with the SureSync Desktop. This has been fixed in 7.1.9.
  • With certain databases, you could receive a “you have unsaved changes” prompt when clicking from object to object in the left hand tree view of the SureSync Desktop even when no changes were actually made. This has been fixed in 7.1.9.
  • In the Local Communications Agent Config Utility, if you attempted to define the listen port with a 0 for the default of 9033 (you can enter 9033 as well) an error would be generated saying that a port must be defined forcing you to enter the full 9033. This has been fixed in 7.1.9.
  • If you made a configuration change in the Local Communications Agent Config Utility that required a restart of the service on a machine running a Communications Agent only install an error would be produced about being unable to start the Scheduler service. The SureSync Scheduler service does not exist on a Communications Agent only install and therefore should not have been attempted to be started. This has been fixed in 7.1.9.

Return to top of page

SureSync 7.1.7 Known Issues

Version 7.1.7 was posted on April 24, 2014.

File Locking

  • If you opened a file for write access, saved a change, a path dropped in the SureSync locking enabled Real-Time Monitor, and then a few seconds later the file was reopened it was possible for the second file open to not be locked. This is extremely timing specific with a small window for the problem to occur. Changes were made to avoid locking on renames and to reject lock releases if the file has been re-opened for write access in the time window. This has been fixed in 7.1.8.

File Synchronization

  • If a folder rename is being processed at the same moment as a path failure, the retry of the folder processing could omit a re-scan of the old folder contents. This can leave folders unsynchronized until a full scan is performed. This has been fixed in 7.1.8.
  • Corrections were made to the processing of folder renames during periods when paths were offline. This has been fixed in 7.1.8.
  • Changes were made to make canceled Jobs, Schedules and Real-Time Monitors shut down faster. This has been fixed in 7.1.8.

User Interface

  • In the Licenses panel, the “Licenses Activated” counter was adding both activated and deactivated license counters together incorrectly. This has been fixed in 7.1.8.
  • A display issue was corrected in the Licenses panel when using only a single workstation license to synchronize with other workstations via UNC path. The Job itself had the licensing correct (the job would run with no error) but the Licenses panel was requesting an unneeded Communications Agent license and showing each workstation as needed a workstation license. This was a harmless display issue. This has been fixed in 7.1.8.
  • In certain circumstances, when executing a Job with a one-way Rule, the interface would clear the source path after successful execution. Additional executions would fail until the source path was specified again. This has been fixed in 7.1.8.
  • When using the Local Communications Agent Config Utility to make a change that requires a restart of the service on a UAC enabled machine, the prompted service restart would fail. The panel will now ask for the proper UAC elevation. This has been fixed in 7.1.8.

Return to top of page

SureSync 7.1.6 Known Issues

Version 7.1.6 was posted on April 5, 2014.

Communications Agent

  • If you changed the default port, encryption, or other values for a remote agent, file copy operations could end up using the wrong connection information. A default is created by an agent the first time a reference is made to a machine. If the Job specifies an overriding connection definition that new definition may be ignored. This would result in communication failing when using non-default settings. This has been fixed in 7.1.7.
  • When processing delta copies, it was possible to encounter a situation where the delta calculations would cause a path to drop repeatedly. This has been fixed in 7.1.7.

Database

  • An unexpected error #1023 could occur during a database upgrade. Old version databases subject to this problem contain a number of orphaned records that caused referential integrity errors on the database copy or conversion. This has been fixed in 7.1.7.

File Synchronization

  • The sending of an e-mail alert could fail with a “System.NullReferenceException: Object reference not set to an instance of an object” error. This has been fixed in 7.1.7.
  • When running a UNC path Job or Schedule, it was possible for the local agent to be dropped incorrect which would cancel the synchronization. This has been fixed in 7.1.7.
  • Corrections were made to passing path level credentials for UNC paths, in some environments logon errors were occurring incorrectly. This has been fixed in 7.1.7.
  • In certain situations, a SyncFiles.exe could stop processing files and the Actions Queued counter will continue increasing without processing files. This has been fixed in 7.1.7.

Real-Time Monitors

  • Intervals that applied to a specific day or set of days were not being saved correctly. This would result in the Intervals not behaving as expected. Intervals that applied to (All Days) were not affected. Any Intervals impacted by this problem need to be edited, changed, saved and changed back to save the Interval record. This has been fixed in 7.1.7.
  • If a folder was renamed in a Real-Time Monitor at the same time one of the paths dropped, it was possible for the replay of journal events from the path drop event to be merged incorrectly preventing one of the folders from being created. This has been fixed in 7.1.7.

User Interface

  • When adding a UNC path via the wizard that required a credential on the path, a “\\uncpath does not exist at this time. If it does not exist when this Job is run your Job will fail. Would you like to proceed anyway?” message would be incorrectly displayed. The test to validate the path was not passing the defined credential properly. This has been fixed in 7.1.7.
  • Corrections were made to Group Folder security in the SureSync Desktop. This has been fixed in 7.1.7.
  • The SureSync Desktop was not always displaying correct status for currently running Jobs, Schedules and Real-Time Monitors. This was caused by testing a connection in the Communications Agent Configuration panel which would drop status connections incorrectly. This has been fixed in 7.1.7.

Return to top of page

SureSync 7.1.5 Known Issues

Version 7.1.5 was posted on March 21, 2014.

Archiving

  • An ArchiveSync Job could be canceled with an “arithmetic operation resulted in an overflow” error. This error occurred because non-archive files were processed in an archive folder. The error was generated trying to read our archive security stream which was not present. Non-archive files are now ignored. These files could also be removed from the archive folder to avoid the problem. This has been fixed in 7.1.6.

Communications Agent

  • When making a modification in the Local Communications Agent Configuration Utility which requires the service to be restarted, an error would be generated about the SureSync Scheduler not being present on Communications Agent only machines. It should not have been attempting to restart the Scheduler service in those cases. This has been fixed in 7.1.6.
  • Agent connections were being dropped when there was no file I/O activity for an extended period of time. This would also prevent the status applications from getting status from an active job which would make the jobs appear to be hung. This has been fixed in 7.1.6.
  • A “System.Threading.ThreadAbortException” message could occur which could generate a large amount of tracing to be generated. This has been fixed in 7.1.6.
  • If an access name is defined (on the Connections tab for a particular Communications Agent) as an IP address, this should now be used for both incoming and outgoing transmissions allowing you to limit SureSync traffic to a single IP address. This has been fixed in 7.1.6.

File Synchronization

  • In rare circumstances, a UNC path Job could attempt to use the user login information from the local SureSync agent instead of the information defined for the path. This could result in failures accessing the path. This has been fixed in 7.1.6.
  • A loop could occur in SyncFiles.exe related to attempting to throttle disk I/O activity. This has been fixed in 7.1.6.
  • Following a period of heavy file copying if the system went idle for 10 minutes a temporary path drop could occur. This has been fixed in 7.1.6.

User Interface

  • Attempting to edit a Rule Subfolder(s) definition by using the text field in the browse accessed by clicking the “…” button on the General tab of a Rule would result in the save not processing and a potential “object reference not set” error. This has been fixed in 7.1.6.
  • If you attempted to cancel an active Real-Time Monitor by clicking “Cancel” in the Ribbon Bar when you have the Rule under the Job for the Monitor selected, an error stating “you do not have permission to cancel .” would be generated. This has been fixed in 7.1.6.
  • IndexOutOfRange errors in the SureSync Desktop were addressed. This has been fixed in 7.1.6.

Return to top of page

SureSync 7.1.4 Known Issues

Version 7.1.4 was posted on March 14, 2014.

File Locking

  • File locking with many active locks (thousands) could result in a situation where the testing of locks took longer than the timeout period, causing path drops. This has been fixed in 7.1.5.
  • A Monitor with file locking could produce an “arithmetic operation resulted in an overflow” error message when handling very large active lock lists. This has been fixed in 7.1.5.
  • The file locking driver could cause a system to become unresponsive for a 2-6 minute period in very timing specific situations under heavy amounts of locking load. This has been fixed in 7.1.5.

File Synchronization

  • A “System.Data.RowNotInTableException: This row has been removed from a table and does not have any data” error could occur in multi-directional rules when processing a rename. This has been fixed in 7.1.5.

User Interface

  • The paste menu option was not available for a Rule or Job after doing a “copy” of a Rule in a different job. This has been fixed in 7.1.5.
  • After adding a new Rule to an existing Job, the Rule is created but the tree was not updated to display the new Rule. To see the new Rule, the Desktop had to be closed and opened again. This has been fixed in 7.1.5.
  • In the SureSync Desktop, it was possible to receive a “System.Data.ConstraintException: Column ‘ScheduleID, JobStep’ is constrained to be unique error. This has been fixed in 7.1.5.

Return to top of page

SureSync 7.1.3 Known Issues

Version 7.1.3 was posted on February 24, 2014.

Communications Agent

  • It was possible to receive a “System.ArgumentException: An item with the same key has already been added” error message when attempting to start a Change Journal which would result in a path drop. This has been fixed in 7.1.4.
  • In some instances, a path could drop due to a communication error and not recover. The job cleanup routine dealing with the communication failure would end up blocking the reconnection attempt in these cases. This has been fixed in 7.1.4.
  • A “NullReferenceException” could occur in SPILog.InsertMsg if a message for a Job was generated at the same time the Job session is being terminated. This could cause an agent to abort. This has been fixed in 7.1.4.

Detailed Status

  • Messages in Detailed Status related to rescans could occasionally display the wrong rescan reason. For example, you would see “first use of job” when it should have said “job was altered.” This is a harmless error because the scan being generated is the same either way. However, changes have been made to show the right message. This has been fixed in 7.1.4.

File Locking

  • With File Locking, in rare circumstances where 2,000 or more files were locked at a time, the driver could encounter a timeout which would prevent further opens on the machine for a 2 minute period. This has been fixed in 7.1.4.

File Synchronization

  • Changes have been made to make the retry queue more understandable. For one way rules, the path was displaying the source of the retry which can be confusing. It will now display the failed path on retries with a one-way Rule. With a multi-way Rule this change is not possible. This has been fixed in 7.1.4.
  • If a path that is local to the job location is dropped, the connection between the job and the status service could be dropped incorrectly. This would prevent you from canceling the job or getting status until the job re-established its connection. This has been fixed in 7.1.4.
  • If a job was run for the first time and the user made any change to the root path record in the user interface while the job was running, a “concurrency violation: the update command affected 0 of the expected 1 records” error could be generated. This has been fixed in 7.1.4.
  • If deleting folders in a destination of a mirror it was possible that you could experience a situation where all file copies stop for several minutes. This occurs because the destination thinks it has too many active file copies due to the active copies from the source to replace the files. This causes the job to attempt to throttle back. This has been fixed in 7.1.4.
  • If you have a root path that is a drive letter or a share name root, SureSync di not properly detect that the drive or share was missing. Attempts to access files would generate errors, but the path would not go offline as it should. Paths lost due to communications failures were properly recognized. This has been fixed in 7.1.4.
  • In a Schedule running a Job with UNC paths, if one of the UNC paths was offline at the start of the execution of the Schedule it could hang on “queuing folders scans.” This has been fixed in 7.1.4.
  • If a path was offline, it was possible for pending folder scans to not be processed to the online paths. These would be processed once all paths were online. This has been fixed in 7.1.4.
  • Folder scans which were started when a path was offline and still active when the path was restored were not being properly rescheduled. This has been fixed in 7.1.4.
  • A “System.Data.DeletedRowInaccessibleException: Deleted row information cannot be accessed in the row error could occur in multi-directional Rules with history. This has been fixed in 7.1.4.
  • With a Schedule running a Job with UNC paths, one of the paths being offline could result in the Schedule sitting on “waiting for file changes” and not completing. This has been fixed in 7.1.4.
  • Folder changes were incorrectly reporting “multiple changes were resolved by copying the latest.” This has been fixed in 7.1.4.

Real-Time Monitors

  • A Real-Time Monitor could enter into a state where Detailed Status showed “waiting for task to complete” but nothing was processed. This has been fixed in 7.1.4.

User Interface

  • When creating a new Group Folder, the Create section of the Ribbon bar was incorrectly disabled until a Desktop restart. This has been fixed in 7.1.4.
  • When copying a Job using the SureSync Desktop, it was possible to receive a “System.NullReferenceException: Object reference not set to an instance of an object” error. This has been fixed in 7.1.4.

Return to top of page

SureSync 7.1.2 Known Issues

Version 7.1.2 was posted on February 10, 2014.

ArchiveSync

  • A “the system cannot find the file specified” error could occur when using ArchiveSync and you had selected to copy security into the archive. This has been fixed in 7.1.3.

Communications Agent

  • When processing files with delta copies enabled, if the computing of the delta information took longer than 5 minutes a path drop would incorrectly occur. This has been fixed in 7.1.3.
  • When running a Job which only has Communications Agent paths remote from the Job location, it was possible for Communications Agent licenses to not be properly marked. This would result in the Job complaining of a machine not having a Communications Agent license. This has been fixed in 7.1.3.
  • In the file synchronization throttle routine, statistics are collected by tasks but these were not being released correctly. The bloat of the statistics list would result in processing slowing down to the point that communications paths would fault resulting in unnecessary path drops. This has been fixed in 7.1.3.

File Synchronization

  • If you have a folder scan that takes more than hour, it should be possible to interrupt the scan and resume later. Changes were made to address issues with resuming those scans. This has been fixed in 7.1.3.
  • There could be a multi-second delay when a Job is terminating if there was a significant amount of activity just prior to the termination. This has been fixed in 7.1.3.
  • If you changed volumes, such as an external hard drive, for a root path that was an archive, the necessary full folder scan was not properly triggered. For other cases, scans were triggered because the journal changed. The change journal is not used for archive destinations, which must rely on detecting volume changes. This has been fixed in 7.1.3.
  • If you add folders that are referenced by a delete action and the truncate delete option is turned off, empty folders may be left behind. This has been fixed in 7.1.3.

Preview

  • If you ran a Preview after modifying a source path in a one-way Rule, it was possible for empty folders to be left behind on the destination. These folders were folders that no longer existed in the new source. The contents of those folders would be deleted but not the now empty folder. This has been fixed in 7.1.3.
  • Running a Job from a Preview could result in no log being generated for that execution. This has been fixed in 7.1.3.

User Interface

    • A “no value given for one or more required parameters” error message could be produces when using filters in the Log Viewer while running an Access database. This has been fixed in 7.1.3.
    • An “object reference not set to an instance of an object” error could occur in Preview if attempting to cancel a Preview that encountered an error. This has been fixed in 7.1.3.
    • In the SureSync Desktop, an “Index was out of range. Must be non-negative and less than the size of the collection” could occur. This has been fixed in 7.1.3.
    • When selecting options such as “Rescan Files” from the SureSync Desktop’s right-click menu while a Rule was selected you could receive a “System.InvalidCastException: Unable to cast object of type” error message. This has been fixed in 7.1.3.
    • When making modifications to the Priorities tab of Options, it was possible to receive a “System.ArgumentException: This row already belongs to this table” error message. This has been fixed in 7.1.3.
    • When editing a root path with the SureSync Desktop, the apply button was not always being enabled correctly. When this occurred, you would be unable to save your change until you clicked on a different node in the Desktop, switched back and made the change again. This has been fixed in 7.1.3.
    • Numerous enhancements and corrections to exception handling with in the Job, Schedule, Restore and Monitor Wizards were made. This has been fixed in 7.1.3.
    • If you started or stopped a trace from the Detailed Status panel after a Job, Schedule or Monitor was already running that item could hang. This has been fixed in 7.1.3.
    • On a Job, if you attempted to uncheck the option “Allow Folder Truncation Delete” and click apply, the option would get turned back on when saved. This has been fixed in 7.1.3.

The Rule wizard was incorrectly showing all rule types when only one root path was defined. The only Rule types that should be available with one path defined are Delete and Skip. This has been fixed in 7.1.3.


Return
 to top of page

SureSync 7.1.1 Known Issues

Version 7.1.1 was posted on February 4, 2014.

Communications Agent

  • An “object reference not set to an instance of an object” error could occur within the status service of the Communications Agent. This could cause the Communications Agent service to be terminated. This has been fixed in 7.1.2.

File Synchronization

  • When a job was terminated due to a path loss detected only by the agent, the connection to the journal reading was not terminated and properly initialized for the next job in the Schedule. This has been fixed in 7.1.2.
  • In rare circumstances, a connection drops could occur when a block was received out of sequence. A dropped block could cause a job to temporarily “hang.” The dropped paths would be recovered and the failed file copy would be retried successfully but the path drop was unnecessary. This has been fixed in 7.1.2.
  • When a job is terminated, its cleanup could terminate the tracing of another running job. This has been fixed in 7.1.2.
  • Corrections were made to folder delete processing that could result in a folder deleted on one side of a multi-mirror being added back if the delete happened during a scan. This has been fixed in 7.1.2.

SyncLockStatus

  • In SyncLockStatus, it was possible for a “file available” pop-up to be displayed when the file had not yet been unlocked. This has been fixed in 7.1.2.
  • In SyncLockStatus, rapid pop-up notifications could result in part of the messages not being displayed. A tray application can only show 255 characters in a pop-up notification. Changes have been made to avoid displaying multiple messages at once that would exceed that limit. This has been fixed in 7.1.2.

User Interface

  • When saving trace options on the Logging tab of a Real-Time Monitor or a Schedule, it was possible to receive a “System.NullReferenceException: Object reference not set to an instance of an object.” This has been fixed in 7.1.2.
  • When using the browse on a Windows 2003 machine to define rule subfolder exclusions, you could receive a “System.ArgumentException: Value does not fall within expected range” message. This has been fixed in 7.1.2.

Return to top of page

SureSync 7.0.34 Known Issues

Version 7.0.34 was posted on January 7, 2014.

Important Note:
7.1 is a minor upgrade to 7.0. It can be installed like any other 7.0 update.

File Locking

  • When using File Locking, the file filter driver responsible for locking had an internal limit of 64 shares on the machine. If a server had more than 64 shares, the driver could encounter an error that would lead to a bluescreen. Changes have been made to remove the limit. This has been fixed in 7.1.1.

File Synchronization

  • If a file size filter is specified for a Monitor interval and a delete action is indicated, the file might be retried immediately instead of waiting for the next interval. This could result in a retry loop until the next interval is reached. This has been fixed in 7.1.1.
  • When performing very large scans of folders the scan operation is likely to be split into sub-scans to permit resuming the scan process in case the scan is interrupted. It is possible that a multi-hour scan could purge some of the split scans before they were executed. This has been fixed in 7.1.1.
  • If a file copy takes more than 5 minutes and the source file location is not the machine running the Job, the connection to agents accessed by the Job may be dropped unnecessarily. The path would recover but should not have been dropped. This has been fixed in 7.1.1.
  • If a file change is detected during open file processing a retry is queued. For a Schedule, the way these retries were being processed could prevent a Schedule from terminating. This has been fixed in 7.1.1.
  • A number of changes were made to error recovery logic to prevent unnecessary path drops. This has been fixed in 7.1.1.
  • If copying a file with multiple streams using standard open file support in a Schedule, it was possible to receive a “File was opened by another user. The file or folder was not processed and will be retried on the next scheduled run.” The file would not be copied on each execution but would repeat the warning. This has been fixed in 7.1.1.
  • Running a Job with the “disconnect users from open destination files” option enabled on the Open Files tab of the Rule, could result in the Communications Agent service crashing with a “the process was terminated due to an internal error in the .NET Runtime at IP 000007FEEB1C09F5 (000007FEEAFD0000) with exit code 80131506.” This has been fixed in 7.1.1.

Preview

  • If you cancel a Preview, the events generated by that Preview could not be replayed if you attempt to run the job without a Preview. This has been fixed in 7.1.1.
  • If you performed a Preview that has deleted files you could get an exception in the Preview that relates to a bad index for the root path. This has been fixed in 7.1.1.

SyncLockStatus

  • In SyncLockStatus, it was possible to receive a “The calling thread cannot access this object because a different thread owns it” pop-up error message. This has been fixed in 7.1.1.

System Requirements

  • System requirements have changed. The need for older versions of the .NET Framework and C++ Runtimes have been eliminated. SureSync now requires only the .NET Framework 4.0 and the Microsoft Visual C++ Runtime 12.0.

User Interface

  • The SureSync Desktop has been updated with improved property sheets and wizard panels. These modern UI changes make the product easier and more intuitive to use.
  • When in a browse panel, it was possible to receive a “GetIconInfo failed in CreateBitmapFromHICON” error message. This has been fixed in 7.1.1.
  • When making certain changes in the Local Communications Agent Config Utility, a restart of the service is required for the changes to become active. When answering yes to the prompt, the utility would report the service had been restarted when it was not. This has been fixed in 7.1.1.
    10] A number of changes were made to error recovery logic to prevent unnecessary path drops. This has been fixed in 7.1.1.
  • If you selected a Rule under a Job in the SureSync Desktop and attempted to run that Job would receive a “System.InvalidCastException: Unable to case object of type ‘SyncDesktop7.Ruleset’ to type ‘SyncDesktop7.DesktopJobItem.’ This has been fixed in 7.1.1.
  • The “purge log entries older than” option accessible by clicking Home and then Options, was not being set to the correct default of 14 days when a new database was created. This could result in database bloat because of the log entries. This has been fixed in 7.1.1.

Return to top of page

SureSync 7.0.33 Known Issues

Version 7.0.33 was posted on December 18, 2013.

Communications Agent

  • Internal agent timeouts were increased to 90 seconds by default. On slow networks or networks with a lot of latency the timeouts previously were not long enough resulting in path drops. A user defined timeout value has also been added to the Agent configuration to allow for increasing or reducing this timeout as necessary. This has been fixed in 7.0.34.
  • If an agent drops references to a job without dropping the connection correctly, the job may hang or fail to release resources while it waits for a copy operation to complete that was already discarded by the agent. This has been fixed in 7.0.34.

File Synchronization

  • If a file was renamed multiple times before the file was processed by SureSync, a performance degradation could occur. Journal index entries were orphaned when new journal entries were processed for the same file name. This has been fixed in 7.0.34.
  • An operating system directory enumeration on a rename could generate an exception when the API error code is not returned when a file is not found. This could cause an IndexOutOfRangeException error to occur. This has been fixed in 7.0.34.
  • A long running operation, such as a very long file copy, may cause the Job to be terminated on an agent causing the operation to fail and may drop the connection. This has been fixed in 7.0.34.
  • If a Job was reading the Change Journal and the Change Journal exceeded its size limit before SureSync processed all the entries, a path drop could occur with an “attempted to read deleted USN entry” error. The correct action is to rescan and not drop the path. This has been fixed in 7.0.34.
  • On a Schedule processing a large directory structure, it was possible to encounter an internal “OutOfMemoryException” when trying to start a worker task thread. This would make the Schedule appear to hang. This has been fixed in 7.0.34.

Preview

  • Running a Job from Preview would not always execute correctly. This has been fixed in 7.0.34.
  • When running a Job from Preview, an “NullReferenceException” could happen internally that would result in the Job failing to complete. This has been fixed in 7.0.34.
  • Preview overrides were not always working correctly. This has been fixed in 7.0.34.

Return to top of page

SureSync 7.0.32 Known Issues

Version 7.0.32 was posted on December 12, 2013.

Communications Agent

  • It was possible to receive a “System.NullReferenceException: Object reference not set to an instance of an object” error out of the FileIO service on a remote agent. The error was related to a status message. This has been fixed in 7.0.33.

File Locking

  • With file locking enabled, if only one action was pending it was possible for the item to end up in our delayed processing queue and not be processed. This would result in our temporary folder temporarily not being deleted and the lock for the file not being released. Any other file action would trigger the processing so this is only seen in a completely idle environment. This has been fixed in 7.0.33.
  • It was possible to receive a “System.InvalidOperationException: Collection was modified after the enumerator was instantiated” error from the synchronization process when locking was enabled. This has been fixed in 7.0.33.

User Interface

  • The log viewer would not always display the log entries for a Job, Schedule or Monitor when using a SQL database. This error was related to changes in 7.0.32. This has been fixed in 7.0.33.

Return to top of page

SureSync 7.0.31 Known Issues

Version 7.0.31 was posted on November 12, 2013.

Communications Agent

  • On Windows 2003 machines, it was possible to receive an “Unable to obtain SID value” error message that caused the Communications Agent to fail to start. This has been fixed in 7.0.32.
    11] In the SureSync Desktop, on the right click menus a horizontal divider line was displayed when not needed. This made it appear like additional options should be displayed when they were not. This has been fixed in 7.0.32.

File Synchronization

  • When using AOFS with a Schedule, it was possible for files processed with VSS to not be copied. This occurred because SureSync attempts a non-VSS copy first and when this fails a retry is scheduled to use VSS for efficiency purposes. If the Schedule ended before the 30 seconds required for the retry passed, the copy would be rescheduled to run on the next Scheduled execution. Now, this action is queued for an immediate retry. The second retry can still occur in some cases which will result in the rescheduled for next execution message but the file is copied. This has been fixed in 7.0.32.
  • New throttling options have been introduced related to maximum disk busy rates and maximum number of active file copies from a single Communications Agent. These options help ensure solid system performance for all processes and prevent SureSync from overwhelming a machine. This has been fixed in 7.0.32.
  • If a Change Journal on a system was too small, SureSync attempts to automatically expand it to a more appropriate size. This resize operation could result in a scan loop that was not productive. This has been fixed in 7.0.32.
  • If you deleted more than 200 folders at once, it was possible to have the Job in question hang. This would prevent further processing of events until the Job, Schedule or Real-Time Monitor was canceled and restarted. This has been fixed in 7.0.32.

Logging

  • The SureSync Job Log Viewer has been renamed to the Log Viewer. This has been fixed in 7.0.32.
  • The Job Viewer has been enhanced. When running a Schedule or Real-Time Monitor with multiple Jobs under it, it is sometimes useful to see only actions related to a specific Job. This can now be done with the ‘Filter Job’ option under the ‘Filter Details’ Ribbon bar button. This has been fixed in 7.0.32.

Real-Time Monitors

  • Real-Time Monitor Interval statistics were often logged as zeroes incorrectly. This has been fixed in 7.0.32.

User Interface

  • When switching between databases, it was possible for the SureSync Desktop to report a “System.InvalidCastException: Unable to cast object of type” error message. This has been fixed in 7.0.32.
  • With the SureSync Desktop, if you clicked on a Rule expanded under a Job which was under a Schedule or Real-Time Monitor, the Enterprise Status button would be disabled incorrectly. This has been fixed in 7.0.32.
  • Additional changes were made to the rule subfolder browse for writing inclusions and exclusions involving special folders such as My Music or My Documents. This has been fixed in 7.0.32.
  • It was possible to receive a “System.Data.OleDb.OleDbException: Record is deleted” error when leaving the SureSync Desktop launched on a machine for long periods of time. This has been fixed in 7.0.32.
  • Within SureSync’s user interface components, dates were not always being localized correctly. This has been fixed in 7.0.32.
  • In the SureSync Desktop, adding a drive letter path using the local agent would be presented as C:\ and from a remote agent would be presented as C:. Both produce the same result but now all show as C: for consistency. Adding a folder to the path will result in the \Path being added. This has been fixed in 7.0.32.
  • Changes were made to the validation routines on the “User ID for UNC Path Access” field in Path Options. This field was enforcing domain\user format when it should not. Non-Windows path access is often done with accounts not in that format. This has been fixed in 7.0.32.
  • If a license that required activation passed the 14 day grace period, the license panel was incorrectly disabling the “activation” checkboxes preventing activation. This has been fixed in 7.0.32.

Return to top of page

SureSync 7.0.30 Known Issues

Version 7.0.30 was posted on October 27, 2013.

Communications Agent

  • In certain rare circumstances, it was possible for a Job, Schedule or Monitor to lose connection to the Status service within the Communications Agent. Changes have been made to recover the connection to the Status service and a fix was made to avoid incorrectly dropping that connection. This has been fixed in 7.0.31.

File Synchronization

  • It was possible to receive “access denied” error messages when attempting to access the NTFS Change Journal or when validating root paths which would result in a “path not found” type error. This would occur even with a valid path that could be accessed via Windows Explorer. A required privilege was not being established correctly within the Communications Agent. This has been fixed in 7.0.31.
  • When copying a file with deltas to multiple destinations, it was possible for a hang to occur in the file copy. This has been fixed in 7.0.31.

User Interface

  • When configuring a Communications Agent machine, the “Test Connection” button on the Computers tab was not being correctly enabled. This required clicking on another tab to enable the button adding a step. This has been fixed in 7.0.31.
  • The rule subfolder browse panel was not always writing the exclusion correctly when dealing with special folders like My Documents. This has been fixed in 7.0.31.
  • When activating a license on a machine that had previously been both activated and then deactivate, it was possible in rare circumstances for the 2nd activation for the machine to fail. This has been fixed in 7.0.31.
  • It was possible to receive a “object reference not set to an instance of an object” error on the launch of the SureSync Desktop which would prevent the Desktop from fully loading. This has been fixed in 7.0.31.

Return to top of page

SureSync 7.0.29 Known Issues

Version 7.0.29 was posted on October 18, 2013.

Database

  • An error “an OleDbException: Unspecified error” could occur with an Access database. A change was made to retry this error. The retry succeeds and resolves the issues. This has been fixed in 7.0.30.

Desktop

  • If you changed the Communications Agent for a path by editing the root path in the Job, the ServerID could be set incorrectly. This could result in a cryptographic error when sending credentials to a remote server and no file access would be allowed. This has been fixed in 7.0.30.
  • The SureSync 7 Desktop was incorrectly allowing a SureSync 6 license file to be imported. Jobs, Schedules and Real-Time Monitors would fail saying they were unlicensed. To avoid confusion, importing a license containing SureSync 6 licenses will be rejected. This has been fixed in 7.0.30.
  • A Communications Agent Configuration Utility could display duplicate identical connections. The configuration was correct but the display was incorrect. This has been fixed in 7.0.30.

File Synchronization

  • Changes were made to minimize the amount of errors reported during a connection loss. Previously, each thread would end up reporting an error as it encountered the path drop. This resulted in logs filled with errors surrounding one event. The number of messages is now greatly reduced. This has been fixed in 7.0.30.
  • A “System.IndexOutOfRangeException: Index was outside the bounds of the array” error could occur that would cancel a Real-Time Monitor. This has been fixed in 7.0.30.
  • If a Multi-Mirror Rule was configured to run without file history, the Job encountered a short path loss and had a very short retry period specified then it was possible for an incorrect path to be selected as the “changed path.” This would result in the wrong action being performed for the file. This has been fixed in 7.0.30.

Return to top of page

SureSync 7.0.28 Known Issues

Version 7.0.28 was posted on October 16, 2013.

Communications Agent

  • On a Communications Agent, it was possible to encounter an error “System.NullReferenceException: Object reference not set to an instance of an object” that would be logged in the Application Event Viewer. When this error occurred, the Communications Agent would stop responding to requests from other Agents causing the path to drop offline until the service was restarted. This has been fixed in 7.0.29.
  • A journal enabled Schedule or a Real-Time Monitor could encounter a “no response was received from the Software Pursuits Communication Service for a Change Journal request” error that would cause a path to drop. This has been fixed in 7.0.29.
  • The error “System.ArgumentException: Attempting to create a client connection to own local agent” error could occur. This would result in the file being retried until the retry count was exhausted and then the file would be skipped. This has been fixed in 7.0.29.

Return to top of page

SureSync 7.0.27 Known Issues

Version 7.0.27 was posted on October 9, 2013.

Communications Agent

  • Jobs could lose connections during the initial startup of our services when the Job source machine was routing connections and a direct connection between remote agents was established in the middle of a file copy. This has been fixed in 7.0.28.
  • In rare circumstances, it was possible for callbacks used in communications to be misrouted. This could result in frequent path drops or cause Schedules or Real-Time Monitors to appear to stop processing. This has been fixed in 7.0.28.
  • The Software Pursuits VSS Manager for SureSync 7 service has been eliminated. The functionality has been integrated into the Communications Agent which is more efficient. This has been fixed in 7.0.28.

Real-Time Monitors

  • When running a Real-Time Monitor with File Locking, if a path was lost while files were locked it was possible for those locked files to not be processed until the next full scan occurred. This has been fixed in 7.0.28.

Scheduler

  • In environments where multiple machines share a SID (this could happen when using drive imaging programs to clone machines), it was possible to have issues with the Scheduler service initializing. The Scheduler would get confused about which machine actually hosted the database and you could receive a message stating “The Scheduler Service on [machinename] is running, but the Scheduler Service failed to initialize.” This has been fixed in 7.0.28.

User Interface

  • The default setting for purging job log entries was supposed to be set to 14 days and was not. This has been fixed in 7.0.28.

Return to top of page

SureSync 7.0.26 Known Issues

Version 7.0.26 was posted on September 27, 2013.

Communications Agent

  • In rare circumstances, an agent to agent connection could be dropped and go unnoticed by the Job for 48 hours. This has been fixed in 7.0.27.
  • In rare circumstances, it was possible for internal Communications Agent service components to lose a connection and then produce an error during internal cleanup that would cause the Communications Agent service to crash. This has been fixed in 7.0.27.

E-mail Alerts

  • Encryption support has been added to the e-mail alert engine. Many public SMTP servers and those provided to companies by third-party providers require this functionality. This has been fixed in 7.0.27.

File Synchronization

  • SyncFiles.exe was not asking for proper elevation on a UAC enabled machine when Jobs, Schedules or Monitors were run manually. This elevation happens automatically with the Scheduler service running. This has been fixed in 7.0.27.
  • Minor licensing corrections have been made. This has been fixed in 7.0.27.
  • If a Change Journal event time matches the start time for a subfolder scan, it was possible for the Change Journal event to be ignored because the Job incorrectly thought the file would be processed by the subfolder scan. The missed file was not processed until the next re-scan of the folder. This has been fixed in 7.0.27.
  • A “System.NullReferenceException: Object reference not set to an instance of an object” error could occur in SyncFilesRoot.NodeItem.NextFile(). This has been fixed in 7.0.27.
  • With a Schedule or a Real-Time Monitor that was processing deltas, it was possible for the job to stop processing during communication issues with a path. This has been fixed in 7.0.27.
  • Support has been added for ReFS volumes. This includes Change Journal support. This has been fixed in 7.0.27.

User Interface

  • The Communications Agent Configuration Utility in the SureSync Desktop was incorrectly allowing a user to save a Communications Agent configuration without a username and password. This has been fixed in 7.0.27.
  • If you delete a Communications Agent configuration, the agent connection was not properly deleted and could be reused in the next Communications Agent defined in the panel. This would result in the “Test” button to the new Agent failing. This has been fixed in 7.0.27.
  • If you attempted to uncheck the “Run a Communications Agent on this machine” checkbox in the Communications Agent configuration panel, it was possible to receive a “System.Data.DeletedRownInaccessibleException: Deleted row information cannot be accessed through the row” error that prevented the change. This has been fixed in 7.0.27.
  • Corrections were made to the Rule Subfolder exclusion browse panel to correctly handle user profile folders. This has been fixed in 7.0.27.
  • A “System.NullReferenceException: Object reference not set to an instance of an object” error could occur in SyncDesktopLib.SyncJob.ValidatePath(). This has been fixed in 7.0.27.

Return to top of page

SureSync 7.0.25 Known Issues

Version 7.0.25 was posted on September 19, 2013.

File Synchronization

  • When encountering a 8.3 format file name that conflicted with a long file name’s auto generated short file name, the Schedule or Monitor would encounter an “object reference not set to an instance of an object” error that would cancel processing. This has been fixed in 7.0.26.
  • It was possible for a Job within a Schedule to cause the Schedule to “hang” on “waiting for completion.” You could cancel the Schedule but it would be waiting on the close of the Job until canceled. This has been fixed in 7.0.26.
  • In rare circumstances, after a file copy error occurred such a timeout the wrong Communications Agent credential could be sent to the wrong machine. This would result in a CryptographicException occurring under the covers that could hang processing on the Schedule or Monitor. This has been fixed in 7.0.26.
  • When running SureSync standard (no Communications Agent licensing) to perform UNC path based or FTP based synchronization, the licensing was incorrectly requiring a Communications Agent license for the machine running the Job. This has been fixed in 7.0.26.
  • A “System.OverflowException: Arithmetic operation resulted in an overflow” error could occur when performing a file resume on a file with multiple streams. This has been fixed in 7.0.26.
  • When looking up a Node for an event or scan request, the wrong NodeItem was selected by Relation ID instead of by Relation Index. This would result in a System.NullReferenceException: Object reference not set to an instance of an object that would cancel the Schedule or Monitor.

User Interface

  • In the SureSync Desktop on the Roots tab of a Rule, you were unable to select individual paths. The only option that would allow successful selection was the “(All Root Paths)” in a multi-directional Rule or the “(All Paths Except Source)” in a one-way Rule. This has been fixed in 7.0.26.
  • When refreshing the SureSync Desktop it was possible to receive a “System.NotSupportedException: This type of CollectionView does not support changes to its SourceCollection from a thread different from the Dispatcher thread” error. This has been fixed in 7.0.26.
  • In the SureSync Desktop, when building Rule Subfolder(s) exclusions on the General tab of the Rule using the UI panel accessed by clicking the “…” button you would receive a “your path exceeds the maximum number of characters of 255” message if your exclusion exceeded 255 characters. This check was incorrect because exclusions can be any length. This has been fixed in 7.0.26.
  • When attempting to save a FTP Profile under Options, an “object reference not set to an instance of an object” error would occur and the save would fail. This has been fixed in 7.0.26.
  • When modifying a Rule, the Last Modified On and Last Modified By statistics on the Statistics tab of the parent Job were not being updated correctly. This has been fixed in 7.0.26.
  • Changes were made to address “object reference not set to an instance of an object” errors that could occur in various locations within the SureSync Desktop. This has been fixed in 7.0.26.
  • When attempting to add a machine to Enterprise Status it was possible to receive a “the given key was not present in the dictionary” error. This has been fixed in 7.0.26.
  • When launching a Trial Run from the SureSync Desktop, the Preview was being incorrectly called. This has been fixed in 7.0.26.
  • When running a Preview, it was possible to receive a “Object reference not set to an instance of an object error. This has been fixed in 7.0.26.
  • When setting a custom path for Trial Run output on the General tab of Options, a “System.InvalidOperationException: Collection was modified enumeration operation may not execute” error would be produced preventing the save. This has been fixed in 7.0.26

Return to top of page

SureSync 7.0.24 Known Issues

Version 7.0.24 was posted on September 16, 2013.

Database

  • During a database upgrade it was possible to receive a “System.Data.ConstraintException: Failed to enable constraints. One or more rows contain values violating non-null, unique or foreign-key constraint” errors. This was caused by bad history that will now be repaired. This has been fixed in 7.0.25.

Desktop

  • System.NullReferenceException: Object reference not set” errors could occur in the SureSync Desktop when performing actions such as a test on a Communications Agent connection. This has been fixed in 7.0.25.

Real-Time Monitors

  • Folder renames in Real-Time Monitors could produce a “SystemArgumentException: An entry with the same key already exists” error that would cancel the Real-Time Monitor. This has been fixed in 7.0.25.
  • When you had multiple Jobs in a Real-Time Monitor and a connection to that machine was lost it was possible for the recovery process to encounter a loop. This has been fixed in 7.0.25.

Schedules

  • When a folder could not be deleted because it contained files that were excluded by the Rule, it was possible for the Schedule to stick on processing the close of the Schedule run. This has been fixed in 7.0.25.

Return to top of page