1.6.242.0813 Release Notes

Feature-level permissions arrived: decide which roles and which departments may use remote screen control and AI diagnosis. The new symptom list narrows by type, source and cause until you reach the origin. Remote screen control gained a special-key toolbar and quality that follows the line.

Feature permissions and target scope

Settings › Permissions

Decide which roles may use each feature, then narrow it further to specific departments or PCs. Outside that scope the server refuses to run the command and the button is locked in Manager. Between allow and deny there is an on-consent option, where remote control starts only after the person at that PC accepts. Leave the scope unset and everything stays allowed, so your current setup is unchanged.

Symptom list

Symptoms

Group symptoms by type, by where they occurred, or by what caused them, then pick a row to narrow to the next step. A small trend bar sits beside each row so you can see when they clustered. Search reads the underlying symptom records rather than what is on screen, so a type name still finds matches three levels deep.

Command results kept at scale

Command management

Fixed a case where result data could be lost when sending a command to many PCs. Results are now stored differently and survive regardless of how many machines were targeted; existing results carry over, so past records stay visible. Live and reopened views no longer disagree on the count.

Remote screen control

Node detail · Beta

A special-key toolbar puts Ctrl+Alt+Del, Hangul/English, Alt+Tab and screen lock one click away. Transmission quality now moves across four levels with line conditions, and the notice on the target PC appears in that PC's language, distinguishing viewing from controlling. Switching logon no longer drops the session.

AI diagnosis

Team Lead Jung

Seven more checks were added. While a response is pending you can see elapsed time and the reason for failure, and stop it partway. Read-only queries run immediately, while commands that change settings now require approval.

More accurate access blocking

Server

The server automatically blocks an address after repeated bad access attempts. We closed the paths where that judgement could be made incorrectly, so legitimate users are not blocked without cause and blocks cannot be evaded.

Recovered missing load figures

Load view

Late-arriving data is now collected again over several passes. Previously the hourly rollup ran once and stopped, so data that came in afterwards was never counted and figures could read lower than reality.

Response-time history recovers itself

Server monitor

When the server's internal store restarted, five-minute history recording stopped and did not resume until the service was restarted. It now recovers on its own.

Stability during install and removal

Package

Closed-network updates could delete a core operating-system component and break later installs; that is fixed. Installs that stalled on internal networks with restricted address schemes now complete, and removing the product now deletes only our own firewall rules.

Node management

Node management widget

Switch between grid and card views. A card reads in three tiers - who this is, how it is doing now, and when and with what - and right-clicking a node in the list opens controls such as agent restart or update and PC reboot or shutdown (visible only to accounts with permission).

Command management

Command management

Search, status and date filters now apply to every command rather than the first page already fetched. The 100-item cap on the template list is gone and scrolling continues. Selected commands can be saved as templates, deleted, or exported in one go.

File upload and connection state

Command console

File upload works again from the command console, accepting absolute paths, paths with spaces, wildcards and paths relative to the current folder. A PC no longer stays listed as connected on its old server after being moved.

Accurate dispatch and records

Server

Commands that never actually went out could still show as dispatched; that is fixed. The server now confirms the signal left and reports how many failed. Access-record search no longer disagrees with its own count.