Percona Monitoring and Management 2.2.0

Percona Monitoring and Management (PMM) is a free and open-source platform for managing and monitoring MySQL, MongoDB, and PostgreSQL performance. You can run PMM in your own environment for maximum security and reliability. It provides thorough time-based analysis for MySQL, MongoDB, and PostgreSQL servers to ensure that your data works as efficiently as possible.

Main improvements in this release are:

• Alternative installation methods available for PMM 1.x are re-implemented for PMM 2: now PMM Server can be installed as a virtual appliance, or run using AWS Marketplace
• AWS RDS and remote instances monitoring re-added in this release include AWS RDS MySQL / Aurora MySQL instances, and remote PostgreSQL, MySQL, MongoDB, and ProxySQL ones
• The new Settings dashboard allows configuring PMM Server via the graphical interface

For PMM install instructions, see Installing PMM Server and Installing PMM Client.

⚠️ Caution
PMM 2 is designed to be used as a new installation — please don’t try to upgrade your existing PMM 1 environment.

Improvements and new features

• PMM-4575: The new PMM Settings dashboard allows users to configure various PMM Server options: setting metrics resolution and data retention, enabling or disabling send usage data statistics back to Percona and checking for updates; this dashboard is now the proper place to upload your public key for the SSH login and to download PMM Server logs for diagnostics
• PMM-4907 and PMM-4767: The user’s AMI Instance ID is now used to setup running PMM Server using AWS Marketplace as an additional verification on the user, based on the Amazon Marketplace rules
• PMM-4950 and PMM-3094: Alternative AWS partitions are now supported when adding an AWS RDS MySQL or Aurora MySQL Instance to PMM
• PMM-4976: Home dashboard clean-up: “Systems under monitoring” and “Network IO” singlestats were refined to be based on the host variable; also avoiding using color as an indicator of state; “All” row elements were relinked to the “Nodes Overview” dashboard with regards to the selected host.
• PMM-4800: The pmm-admin add mysql command has been modified to make help text more descriptive: now when you enable tablestats you will get more detail on if they’re enabled for your environment and where you stand with respect to the auto-disable limit
• PMM-4969: Update Grafana to version 6.5.1
• PMM-5053: A tooltip was added to the Head Block graph on the Prometheus dashboard
• PMM-5068: Drill-down links were added to the Node Summary dashboard graphs
• PMM-5050: Drill-down links were added to the graphs on all Services Compare dashboards
• PMM-5037: Drill-down links were added to all graphs on the Services Overview dashboards
• PMM-4988: Filtering in Query Analytics have undergone improvements to make group selection more intuitive: Labels unavailable under the current selection are shown as gray/disabled, and the percentage values are dynamically recalculated to reflect Labels available within the currently applied filters
• PMM-4966: All passwords are now substituted with asterisk signs in the exporter logs for security reasons when not in debug mode
• PMM-527: node_exporter is now providing hardware monitoring information such as CPU temperatures and fan statuses; while this information is being collected by PMM Server, it will not be shown until a dedicated dashboard is added in a future release
• PMM-3198: Instead of showing All graphs for all services by default, MySQL Command/Handler Counters Compare dashboard now shows the predefined set of ten most informative ones, to reduce load on PMM Server at its first open

Fixed bugs

• PMM-4978: The “Top MySQL Questions” singlestat on the MySQL Instances Overview dashboard was changed to show ops instead of percentage
• PMM-4917: The “Systems under monitoring” and “Monitored DB Instances” singlestats on the Home dashboard now have a sparkline to make situation more clear with recently shut down nodes/instances
• PMM-4979: Set decimal precision 2 for all the elements, including charts and singlestats, on all dashboards
• PMM-4980: Fix “Load Average” singlestat on the Node Summary dashboard to show decimal value instead of percent
• PMM-4981: Disable automatic color gradient in filled graphs on all dashboards
• PMM-4941: Some charts were incorrectly showing empty fragments with high time resolution turned on
• PMM-5022: Fix outdated drill-down links on the Prometheus Exporters Overview and Nodes Overview dashboards
• PMM-5023: Make the All instances uptime singlestat on the Home dashboard to show Min values instead of Avg
• PMM-5029: Option to upload dashboard snapshot to Percona was disappearing after upgrade to 2.1.x
• PMM-4946: Rename singlestats on the Home dashboard for better clarity: “Systems under monitoring” to “Nodes under monitoring” and “Monitored DB Instances” to “Monitored DB Services”, and make the last one to count remote DB instances also
• PMM-5015: Fix format of Disk Page Buffers singlestat on the Compare dashboard for PostgreSQL to have two digits precision for the consistency with other singlestats
• PMM-5014: LVM logical volumes were wrongly sized on a new AWS deployment, resulting in “no space left on device” errors.
• PMM-4804: Incorrect parameters validation required both service-name and service-id parameters of the pmm-admin remove command to be presented, while the command itself demanded only one of them to identify the service.
• PMM-3298: Panic errors were present in the rds_exporter log after adding an RDS instance from the second AWS account
• PMM-5089: The serialize-javascript package was updated to version 2.1.1 because of the possibility of regular expressions cross-site scripting vulnerability in it (CVE-2019-16769). Please note PMM versions were not affected by this vulnerability, as serialize-javascript package is used as a build dependency only.
• PMM-5149: Disk Space singlestat was unable to show data for RDS instances because of not taking into account sources with unknown file system type