Skip to content

Percona Monitoring and Management 2.32.0

Release date: Nov 8, 2022
Installation: Installing Percona Monitoring and Management

Percona Monitoring and Management (PMM) is an open source database monitoring, management, and observability solution for MySQL, PostgreSQL, and MongoDB.

Important

We recommend using the latest version of PMM. This ensures that you have access to the latest PMM features and that your environment runs on the latest version of the underlying components, such as VictoriaMetrics, with all the bug fixes in place.

Release Highlights

The release highlights are also available in video format. Watch the What’s new in PMM 2.32 video

Backup Management for MongoDB is now generally available

We are happy to announce that the Backup Management for MongoDB has reached General Availability and is no longer in Technical Preview.

This change comes with significant functionality enhancements and added UX improvements:

  • Point-in-time recovery (PITR) restores for MongoDB backups: This enables you to restore Amazon S3-compatible backups from any checkpoint in the past.
  • Local client storage: In addition to Amazon S3-compatible, we have now also enabled the Local Client location option, which saves backups on the remote folder mounted to PMM client host.
  • Key UX improvements: You can now create on-demand and scheduled backups from the same window instead of using separate modals. In addition, we have added more icons and renamed tabs columns to better label the actions available behind important options.

For more information, see the Backup and Restore topic in the documentation.

State-of-the-art new Home dashboard

In our continued effort to enhance user experience, we have added another feather to our cap by replacing the existing home dashboard with a brand new Home dashboard in PMM 2.32.0.

We are releasing this dashboard as the default Home dashboard. This dashboard was released in PMM 2.30.0 as an experimental dashboard. After considering your feedback and working meticulously on this dashboard, we have developed an enhanced version in PMM 2.32.0. Read more in the blogpost.

Highlights of the new Home dashboard include:

  • Optimized performance to handle a large number of nodes concurrently
  • Scalable to accommodate varying loads at all times
  • Easy anomaly detection to check the health of your nodes from the Home dashboard
  • Color-coded states on panels for a quick visual representation of the problem areas
  • Comprehensive and easy-to-use dashboard to reduce cognitive load

For detailed information, see documentation.

Enhanced main menu

  • In PMM 2.32.0, you can easily access all dashboards for monitored services due to the contemporized navigation structure. All the dashboards in the services folder that were not accessible from the main menu are now accessible from the main menu via Other Dashboards. Also, your custom dashboards can be accessed easily from the main menu via Other Dashboards.

    For example, to access the dashboards for MySQL, go to MySQL > Other dashboards.

  • Only monitored services are now included in the main menu, enhancing the user experience and minimizing cognitive load to maximize usability.

    For example, you will only see MongoDB dashboards in the main menu if you monitor the MongoDB database rather than all the other dashboards.

New Features

  • PMM-9366: [Backup Management]: Added possibility to restore MongoDB PITR backups. This option is available under Backup > All Backups > Actions > Restore this backup for any PITR-type backups available in the table.
  • PMM-9368: [Backup Management]: Added support for local client storage: When adding storage locations under Backup > Storage Location, you can now also select Local Client option as the location type for storing future backups.
  • PMM-10599, PMM-10503: [Backup Management]: Refactored User Interface for creating backups to enable you to create on-demand and scheduled backups from the same window instead of using separate modals.
  • PMM-10470: [Backup Management]: Following the support for MongoDB physical backups introduced in PMM 2.31, we have now added the possibility to restore these backups. Restoring MongoDB physical backups requires additional manual operations after the restore and is only available with Percona Server for MongoDB.
  • PMM-10502, PMM-10831: [Backup Management]: Improved User Experience and labels on all four tabs of the Backup section:

    • Added icons to better illustrate failed and successful backups.
    • Added Service Name column to show the service from where the backup was retrieved.
    • Renamed DB Vendor tab to DB Technology.
    • Improved Actions menu with clearer labeled options.
  • PMM-10742: [Backup Management]: Extended PMM compatibility with Percona Backup for MongoDB (PBM) to support the latest version, v. 2.0 PBM.

  • PMM-10829: [Backup Management]: The backups table under Backup > All Backups now includes a new Location column which displays the service where the backup came from (S3 or Local Storage).

  • PMM-8792, PMM-9055, PMM-9056 - PMM 2.32.0 introduces the parameter --max-query-length in the pmm-admin command for MySQL, PostreSQL, and MongoDB databases. It can also be set up through the UI while adding the service. You can see a new input field on the UI, where you can set the value for max-query-length. With these changes, queries longer than 2048 characters are not truncated if this parameter is set to -1 (which means no limit), and the Explain and Examples tabs on the QAN dashboard do not throw an error (PMM-1899). These tabs provide contextual information pertaining to the query, thus enhancing user experience. For more information, see documentation.

  • PMM-9545: To monitor Kubernetes clusters, PMM now collects Kubernetes metrics from kube-state-metrics. For more information on kube-state-metrics, see documentation.

    Disclaimer

    This feature is still technical preview and is subject to change. We recommend that early adopters use this feature for testing purposes only.

  • PMM-10579: [Dashboard]: Only monitored services are displayed on the main menu for a clean and simple User Interface.

  • PMM-10512: [Alerting]: When creating new alert rules. you can now use two new labels (application_name and usename) for pg_stat_activity_ metrics.

Improvements

  • PMM-9946: [DBaaS]: To enhance user experience, once PMM is deployed to a Kubernetes cluster and a user wants to use DBaaS, PMM automatically registers that Kubernetes cluster to DBaaS.

  • PMM-10733: [DBaaS]: We have moved another step forward in simplifying the use of DBaaS by simplifying the selection of the database version.

  • PMM-10723: [DBaaS]: To make the content legible, we have changed the color of the notification block.

  • PMM-10156: [Dashboard]: Starting with PMM 2.32.0, we have an enhanced PMM dashboard that gives a birds-eye view of the services, infrastructure, and critical issues (if any).

  • PMM-10612: [Dashboard]: You can now easily access all dashboards for monitored services. Dashboards in the services folder (MySQL dashboards) that were not accessible from the main menu are now accessible from Other Dashboards.

  • PMM-10695: [Dashboard]: For precision, we have added a unit (%) for the CPU Anomalies panel.

  • PMM-10487: [Dashboard]: Our Experimental MongoDB dashboards are now equipped with tooltips to help you understand the information provided in the panels.

  • PMM-10692: [Dashboard]: Enhanced the High CPU Usage, High Disk Queue, and High Memory Usage panels to make the troubleshooting process easier and more actionable. Just click the node name on the panel to see the details of the problem.

  • PMM-10751: [Dashboard]: For the Vacuum dashboard, when filtering by Service name, the first panel that shows the service name and resolution was hidden due to the Disclaimer. It has now been moved above the Disclaimer to make it readable.

  • PMM-10690: [Dashboard]: Starting with PMM 2.32.0, we have enabled sparklines for Disk Read and Write panels to visually represent and show trends in the data.

  • PMM-10697: [Dashboard]: We have improved the visualization of the Used Memory Anomaly Panel on the Home Dashboard so that all the anomalies are visible.

  • PMM-10694: [Dashboard]: Added and refined tooltips for all the panels on the dashboard.

  • PMM-10721: On the Explore dashboard, Metrics panel, removed the links pointing to the Grafana discussion board.

  • PMM-10662, PMM-10633, PMM-10631, PMM-10600: An additional MySQL and MongoDB related datapoints have been added to Telemetry.

  • PMM-10926: [Backup Management]: Updated the documentation with information about required user permissions for creating Mongo BM backups. For more information, see the MongoDB topic.

  • PMM-10393: The panel for total time taken for scraping metrics by Postgres exporter is improved. Instead of showing coarse-grained measurements of the total time taken by the exporter, it now shows the total time taken per collector.

  • PMM-10555: Runtime for collectors in MongoDB exporter is pushed via Telemetry to PMM. This information detects considerable scrape time for collectors and improves collector performance in future releases.

Bugs Fixed

  • PMM-10944: [Backup Management]: Fixed an issue that was preventing QAN to retrieve data after a restore for MongoDB with authentication enabled.

  • PMM-10840: [Alerting]: Fixed embedded templates that are using expressions like x == 0 or x != 0. PMM servers connected to Percona Platform already include this fix, since Percona Platform delivered it via Automatic Template Updates, a couple of weeks ago.

  • PMM-10759: [Alerting]: Renamed the Alerting option in the PMM Settings to Percona Alerting. Since this option refers to Percona Alert Rule Templates, this new label reflects better the fact that this option is different from the basic, built-in Alerting functionality.

  • PMM-10810: [Alerting]: Fired alerts now show the correct summary, which includes the node or service name specified in the template annotations.

  • PMM-10500: [DBaaS]: DBaaS: Fixed an issue where a database cluster for an unregistered k8s cluster was still being displayed on the UI.

  • PMM-10640: [DBaaS 1.10, 1.11]: Fixed an issue where a MongoDB 5.0.x cluster was not getting created.

  • PMM-9874: [Dashboard]: Fixed an issue where the current ReplSet State stat is unreadable after modifying or restarting the database cluster for MongoDB Dashboard.

  • PMM-10339: [Dashboard]: Fixed the duplicate values displayed on the CPU usage charts for MongoDB and MySQL dashboards.

  • PMM-10802: [Advisors]: Fixed MongoDBGetDiagnosticData and MongoDBReplSetGetStatus query type for Advisors Checks V2.

  • PMM-10688: This bug fix restores the ability to copy existing kubeconfig and register a new cluster by provided copy. However, AWS secrets will be hidden in the UI.

Known Issues

PMM-11029: pmm-agent: killed when –max-query-length <=3

If the value of the parameter max-query-length is set to 3, 2 or 1, the PMM agent will get terminated.

Solution

The value of the parameter should be greater than 3.

PMM-11126: PMM server cannot be upgraded to version 2.32.0

A bug in PMM Server ansible scripts caused PMM to upgrade Nginx’s dependencies without updating Nginx itself. Due to this, PMM throws an error while upgrading and cannot upgrade to a newer version.

Important

This issue persists on versions prior to PMM 2.32.0.

Solution

Log in to PMM server and execute the following command:

sed -i 's/- nginx/- nginx*/' /usr/share/pmm-update/ansible/playbook/tasks/update.yml

[PMM-10858](https://jira.percona.com/browse/PMM-10858): **PMM server doesnt show the latest versions available with the instances created from AWS**


In specific environments, including AWS, some EPEL repository mirrors did not respond within the time limit defined by `pmm-update,` currently set to 30 seconds. It was causing `supervisord `to kill `pmm-update-checker`, the component responsible for determining if a newer PMM Server is available for upgrade.

!!! caution alert alert-warning "Important"
    This issue persists on PMM 2.32.0 and all versions prior to PMM 2.32.0.

**Solution**

Log in to the PMM Server and run the following command as a root user:

```sh
   $ yum-config-manager --setopt=epel.timeout=1 --save

Last update: 2023-02-09