Review recent Confluence configuration changes

This recommendation appears detect recent configuration changes, such as Confluence properties, database connection pool settings, feature flags, or Marketplace apps that were installed, updated, or uninstalled, that degraded the performance of Confluence.

Signals used

For us to detect this issue, the following conditions need to occur together:

  • At least one key experience, such as viewing or editing a page, shows degraded ready-for-user (RFU) page load times.

  • During the same time window, we've detected one or more of the following configuration changes:

    • Confluence configuration changes

    • Marketplace app changes

    • Feature flag changes

    • Database configuration changes

How we detect it

Signal

Details

Ready-for-user (RFU) degradation

We look for RFU degradation, where key Confluence experiences take significantly longer than their historical baseline. This helps us confirm that users are actually experiencing slower page loads, not just a one-off spike.

Confluence configuration changes

We look for changes to Confluence application properties and system properties within the same time window. We compare the property values at the start and end of the window to detect additions, removals, or modifications.

Marketplace app changes

We detect Marketplace apps that were installed, updated, or uninstalled in the same time window. These changes can introduce new performance regressions.

Feature flag changes

We compare feature flags between the start and end of the time window to detect flags that were enabled, disabled, or added.

Database configuration changes

We detect changes to the database configuration, such as maximum pool size, minimum idle connections, or idle timeout, by comparing settings between Confluence startup event.

Filters

When this issue occurs multiple times, we group occurrences under different time filters, for example, Last 24 hours or Last 7 days.

  • The Degraded experiences section of the recommendation captures the highest degradation from all occurrences in the selected time filter.

  • The Identified changes section summarizes changes from all occurrences in the selected time filter.

This helps you see which configuration changes are most closely aligned with periods of degraded performance.

Frequency

The Frequency section in the recommendation shows how often this issue has occurred over time. This can help you identify if:

  • Performance degradations often follow configuration changes.

  • Clusters of changes (for example, multiple app updates or property changes rolled out at once) line up with repeated recommendations.

The actions shown in the recommendation are conditional. You'll only see the changes that were detected.

Review and revert risky Confluence configuration changes

Download the CSV file with the list of changes and review them. Look for configuration changes that affect indexing, search, caching, or other performance-sensitive settings. If you suspect a specific change contributed to the regression, consider reverting it to the previous value, especially if it is not a documented or supported setting.

Review Marketplace app changes

Download the CSV file with the list of changes and review them. Focus on newly installed apps or apps that were updated just before the degradation started. If you suspect an app is contributing to the regressions, temporarily disable or revert it to the previous version.

Review feature flag changes

Download the CSV file with the list of changes and review them. If you enabled specific feature flags, consider disabling them to check if performance returns to normal.

Review database configuration changes

Database connection settings can significantly impact Confluence performance if misconfigured. Compare your current settings with Atlassian’s recommended database configuration for your Confluence version and database. If a recent change diverges from those recommendations, consider reverting it.

Still need help?

The Atlassian Community is here for you.