Server Maintenance

These options are dedicated to server maintenance and server health.

This maintenance panel offers granular control over various resources within your instance. Each option has certain scenarios that justify their use case, which can be reviewed below. Maintenance is executed automatically on your Xano instance, and it is typically not necessary to perform these actions manually. If you have any questions, please reach out to support for more information.

Clear Internal Cache

Clearing your internal cache can be helpful if you are seeing any of the following errors or behaviors.

• Invalid App errors when calling your APIs

• 404 errors when calling your APIs

• Slow performance when working inside of Xano

Database Maintenance

These are specific options related to database maintenance, and can be helpful if you are seeing any of the following errors or behaviors.

• Database storage is not reporting accurately

• Database table row counts are not reporting accurately

• If you have recently deleted a large amount of records from your database and the available storage has not yet updated to reflect this

ANALYZE

ANALYZE gathers statistics about the data distribution in tables, enabling the query optimizer to generate efficient execution plans. It updates the query planner's knowledge of the data, improving query performance by enabling better index selection and join strategies. It will recalculate the storage utilization and row count for your database tables.

This operation is automatically executed daily.

PARTIAL VACUUM

This command analyzes and cleans up the database, but it does not necessarily reclaim all available disk space. It marks the space previously occupied by deleted rows as reusable for future inserts and updates. VACUUM also updates statistics used by the query planner to improve query performance.

FULL VACUUM

Full Vacuum requires at least 50% free database storage to execute safely.

This command performs a more thorough cleanup compared to partial. It reclaims all available disk space by rewriting the entire table and indexes from scratch. This process can be more resource-intensive and time-consuming, as it involves copying the data to a new file and rebuilding the indexes. Full VACUUM can significantly improve disk space utilization but may cause downtime for larger databases.

Server Maintenance

Restarting your instance is typically quick but can take several minutes to complete. These actions can lead to downtime. You should only run these actions if you have a need to do so.

Restarting your instance does not reset API or Function Stack performance-related issues, and optimization of your API endpoint and/or Function Stack should be troubleshooted via our troubleshooting guide.

Restart Backend

Restart Backend resets the instance’s server. All records and endpoints stay the same. This action terminates all backend processes and starts the server anew. Tasks are not affected.

Typical Scenario: You are seeing an unexpected traffic spike related to some bad actors, and your instance becomes inaccessible because of this. Restarting your backend can interrupt their access temporarily, giving you a window to modify your database or function stacks to halt their efforts.

Restart Database

Restart Database allows you to reset your database resources. This does not reset your database values but will cease ongoing database transactions: all of your records will stay the same, but while the database is restarting, nothing can be read/written to it.

Typical Scenario: You just executed a Bulk Add Records function, and realize that the records being created are incorrect, and want to halt execution of the database operation.

Restart Frontend

Restart Frontend resets the UI elements of your instance. This is useful if you experience any visual bugs. Typically, you won’t ever need to utilize this option.

Restart Node

Restart Node resets the resources dedicated to Lambda functions.

Typical Scenario: You are seeing unexplained errors when running Lambda functions.

Restart Redis

Restart Node resets the resources dedicated to Redis caching and clears all values.

Typical Scenario: You are seeing errors such as Invalid App when calling your APIs, and clearing your instance cache did not resolve the issue.

Restart Task

Restart Task resets the resources dedicated to background tasks. Background tasks that are scheduled will stay scheduled. Tasks in the middle or running will stop and restart when the Restart Task action has been completed.

Typical Scenario: You have a background task running that seems to be “stuck”, or behaving unexpectedly, and you want to halt execution before it completes.

History Database

From this selection, you can manually delete all data from your request history. This option can be useful if you have experienced an elevated volume of requests and need to quickly free up available storage space.

You can choose to clear the request history from custom functions, middleware, APIs, tasks, and database triggers.

Using the Force option will restart your backend during this process, and can result in some downtime while processing.

History Cache

When requests are made to objects that log request history, such as APIs, the history is not immediately entered into the database; it is held in a cache and processed at a fast, regular cadence.

In some rare circumstances, such as slow performance inside of Xano or errors accessing your instance, you may have encountered a full request cache, and clearing it here can resolve those issues.

Please note that if items are cleared from the cache, they will not be logged in your request history when browsing inside of Xano.