Table of Contents

Skyline DataMiner Labs Monitoring (Technical Reference)

About

This connector ingests historical "New client registered" alarm events using SLNet history queries. It aggregates logins per user, client type, and date, and it keeps a rolling window based on a retention threshold.

Workflow

The connector uses the following workflow:

  1. A lookup start date is configured (Start Lookup Datetime).

  2. The Refresh action (button or hourly timer) requests the alarm history for the "New client registered" event.

  3. Each event is parsed to extract the username, device, and client.

  4. A composite key (User | Normalized Client | Date) is used:

    • If the key exists: The Last Login timestamp and Times Accessed parameters are updated.
    • If the key does not exist yet: A new row is created.

  5. A daily cleanup (24h timer) removes rows whose last login date exceeds the Days Threshold value.

Main User Activity Overview Columns

Column Description
Event Type Always "New client registered" (current scope).
User Name DataMiner user full name.
User Device Device/host reported in the event.
User Client Normalized client type (impersonation suffix removed).
User Category Internal/Local/External.
Last Login Timestamp Timestamp of most recent login that day (updates on repeated logins).
Times Accessed Count of logins for that day and client combination.

Manual Controls

  • Refresh: Retrieves new events since the last lookup.
  • Clear: Removes all entries immediately.
  • Days Threshold: Allows you to adjust how long data is kept. The threshold will be applied during the next cleanup.

Data Acquisition

Mechanism Details
Source Alarm history (GetAlarmDetailsFromDbMessage)
Filter ParameterID = 64503
Interval Hourly timer (Slow Timer) or manual Refresh button
Range From last processed OLE automation date (lookupLastDatetime) to Now
Aggregation Key = User Name + Normalized Client + Calendar Date

Normalization Logic

  • Impersonation suffix: Any " - impersonated for ..." segment in the client string is removed.
  • User categories:
    • Internal: Email ends with @skyline.be.
    • Local: No email address.
    • External: All others.

Retention/Cleanup

  • The Days Threshold parameter defines the maximum allowed age of each entry (i.e., the difference between the current date and the last login date).
  • A daily timer invokes the cleanup (QAction 3), deleting rows older than the threshold.
  • Rows with unparsable timestamps are considered invalid and removed.

Table Behavior

  • One row is added per day per user and client type.
  • When multiple logins occur on the same day:
    • The timestamp is adjusted to the latest timestamp.
    • The "Times Accessed" parameter value increases.
  • The display key combines Device, Category, and Event Type (per NamingFormat).

QActions

QAction Role
110 Event retrieval, aggregation, persistence.
3 Cleanup logic (triggered manually or by timer).

Timers

Timer Interval Purpose
Slow Timer (1h) 3600000 ms Periodic refresh (Group 110).
Very Slow Timer (24h) 86400000 ms Periodic cleanup (Group 3).

Known Limitations

  • Only processes one event type ("New client registered").
  • The daily aggregation loses intra-day chronological order beyond the last timestamp.

Troubleshooting

Symptom Possible cause Action
No new rows lookupStartDatetime not set or no events Configure start or verify event generation.
Rows not removed Days Threshold too high Adjust the Days Threshold value.
Unexpected client strings Normalization rule not matching Review the impersonation suffix.
Counter does not increment Events share different client strings Confirm the normalization outcome.
Improve this page