Skip to content

🔄 Preview: [TT-17927] Update documentation for master - #2797

Draft
probelabs[bot] wants to merge 486 commits into
productionfrom
update/TT-17927/release-main-docs
Draft

🔄 Preview: [TT-17927] Update documentation for master#2797
probelabs[bot] wants to merge 486 commits into
productionfrom
update/TT-17927/release-main-docs

Merge main into update/TT-17927/release-main-docs

1eb36b6
Select commit
Loading
Failed to load commit list.
probelabs / Visor: performance succeeded Aug 27, 2026 in 46s

✅ Check Passed (Warnings Found)

performance check passed. Found 2 warnings, but fail_if condition was not met.

Details

📊 Summary

  • Total Issues: 2
  • Warning Issues: 2

🔍 Failure Condition Results

Passed Conditions

  • global_fail_if: Condition passed

Issues by Category

Performance (2)

  • ⚠️ snippets/dashboard-config.mdx:164 - The default value of 0 (unlimited) for max_open_connections can lead to database connection exhaustion under load. This setting is documented for storage.main, storage.analytics, storage.logs, and storage.uptime. In a production environment with multiple Tyk Dashboard replicas, an unlimited number of connections can overwhelm the PostgreSQL server, causing severe performance degradation. The documentation should warn users about this risk.
  • ⚠️ snippets/dashboard-config.mdx:176 - The default behavior of reusing connections forever (connection_max_lifetime is empty) can cause intermittent request failures in environments with stateful firewalls or load balancers that close idle connections. This applies to all four PostgreSQL storage configurations (main, analytics, logs, uptime). This can lead to poor performance and errors when the Dashboard attempts to use a connection that has been silently terminated by a network device.

Powered by Visor from Probelabs

💡 TIP: You can chat with Visor using /visor ask <your question>

Annotations

Check warning on line 168 in snippets/dashboard-config.mdx

See this annotation in the file changed.

@probelabs probelabs / Visor: performance

performance Issue

The default value of 0 (unlimited) for `max_open_connections` can lead to database connection exhaustion under load. This setting is documented for `storage.main`, `storage.analytics`, `storage.logs`, and `storage.uptime`. In a production environment with multiple Tyk Dashboard replicas, an unlimited number of connections can overwhelm the PostgreSQL server, causing severe performance degradation. The documentation should warn users about this risk.
Raw output
For each `max_open_connections` field, add a recommendation to set a specific limit for production deployments. For example: 'For production environments, it is strongly recommended to set a limit based on your database capacity and number of Dashboard replicas to prevent connection exhaustion.'

Check warning on line 180 in snippets/dashboard-config.mdx

See this annotation in the file changed.

@probelabs probelabs / Visor: performance

performance Issue

The default behavior of reusing connections forever (`connection_max_lifetime` is empty) can cause intermittent request failures in environments with stateful firewalls or load balancers that close idle connections. This applies to all four PostgreSQL storage configurations (`main`, `analytics`, `logs`, `uptime`). This can lead to poor performance and errors when the Dashboard attempts to use a connection that has been silently terminated by a network device.
Raw output
For each `connection_max_lifetime` field, add a note advising users to set a lifetime shorter than any network infrastructure timeouts. For example: 'It is recommended to set this to a value like `30m` to prevent errors caused by network devices closing idle connections.'