Resolved Issues

This section lists the issue that is resolved with this release.

• PAALS-1923: Monitor creation via the API slowed progressively as active monitor count increased, causing delays of 2–3 minutes in production during bulk provisioning.

  The fix resolves the underlying performance issue; testing at 5,000 active monitors confirmed an average monitor creation response time of approximately 9 seconds.

• PAALS-1883: TWAMP data displayed in the GUI export could continue to display samples for a stream even after ingestion into the PostgreSQL metrics tables had stopped, creating a discrepancy between what the GUI reported and what was actually stored in the database. This was caused by the metrics service failing to recover its connection to Kafka/Zookeeper after a disconnect, silently halting data ingestion while the GUI continued to reflect stale data.

  The fix improves error handling so the metrics service reliably reconnects and resumes ingestion after a connection loss, ensuring GUI and database values stay consistent.

• PAALS-1889: The Action text configured in an alarm template was not applied to monitors that used the template, causing email notifications to omit the suggested action text even though it displayed correctly when set directly on a monitor.

  The fix ensures action text from alarm templates is correctly included in monitor alarm notifications.

• PAALS-1916:Hourly (minute60) metrics generation intermittently stopped for approximately 2–3 hours at a time, while minute1 and minute5 metrics continued updating normally. In some cases data resumed automatically. In others, a manual restart of the metrics service was required, and data for the affected hours was not always backfilled. The issue was traced to the timescaledb exporter failing to insert records with values exceeding the standard integer column range, causing the insert transaction to roll back.

  The fix resolves the underlying data type handling, ensuring hourly metrics generation completes reliably without gaps.