Monitoring

Percona Monitoring and Management (PMM) provides an excellent solution of monitoring Percona Server for MongoDB.

Note

Only PMM 2.x versions are supported by the Operator.

PMM is a client/server application. PMM Client runs on each node with the database you wish to monitor: it collects needed metrics and sends gathered data to PMM Server. As a user, you connect to PMM Server to see database metrics on a number of dashboards.

That’s why PMM Server and PMM Client need to be installed separately.

Installing PMM Server

PMM Server runs as a Docker image, a virtual appliance, or on an AWS instance. Please refer to the official PMM documentation for the installation instructions.

Installing PMM Client

The following steps are needed for the PMM client installation in your Kubernetes-based environment:

1. The PMM client installation is initiated by updating the pmm section in the deploy/cr.yaml file.

   • set pmm.enabled=true
   • set the pmm.serverHost key to your PMM Server hostname or IP address (it should be resolvable and reachable from within your cluster)

   • authorize PMM Client within PMM Server in one of two ways:

     with token-based authorization (recommended)with password-based authorization

     Acquire the API Key from your PMM Server and set PMM_SERVER_API_KEY in the deploy/secrets.yaml secrets file to this obtained API Key value. Keep in mind that you need an API Key with the “Admin” role. The API Key won’t be rotated automatically.

     check that the PMM_SERVER_USER key in the deploy/secrets.yaml secrets file contains your PMM Server user name (admin by default), and make sure the PMM_SERVER_PASSWORD key in the deploy/secrets.yaml secrets file contains the password specified for the PMM Server during its installation.

     Password-based authorization method is deprecated since the Operator 1.13.0.

     Note

     You use deploy/secrets.yaml file to create Secrets Object. The file contains all values for each key/value pair in a convenient plain text format. But the resulting Secrets contain passwords stored as base64-encoded strings. If you want to update password field, you’ll need to encode the value into base64 format. To do this, you can run echo -n "password" | base64 --wrap=0 (or just echo -n "password" | base64 in case of Apple macOS) in your local shell to get valid values. For example, setting the PMM Server API Key to new_key in the my-cluster-name-secrets object can be done with the following command:

     in Linuxin macOS

     $ kubectl patch secret/my-cluster-name-secrets -p '{"data":{"PMM_SERVER_API_KEY": '$(echo -n new_key | base64 --wrap=0)'}}'
     

     $ kubectl patch secret/my-cluster-name-secrets -p '{"data":{"PMM_SERVER_API_KEY": '$(echo -n new_key | base64)'}}'
     

     Apply changes with the kubectl apply -f deploy/secrets.yaml command.

   • Starting from the Operator version 1.12.0, MongoDB operation profiling is disabled by default, and you should enable it to make PMM Query Analytics work. You can pass options to MongoDB in several ways, for example in the configuration subsection of the deploy/cr.yaml:

     spec:
       ...
       replsets:
         - name: rs0
           size: 3
           configuration: |
             operationProfiling:
               slowOpThresholdMs: 200
               mode: slowOp
               rateLimit: 100
     

   • you can also use pmm.mongodParams and pmm.mongosParams keys to specify additional parameters for the pmm-admin add mongodb command for mongod and mongos Pods respectively, if needed.

     Note

     Please take into account that Operator automatically manages common MongoDB Service Monitoring parameters mentioned in the officiall pmm-admin add mongodb documentation, such like username, password, service-name, host, etc. Assigning values to these parameters is not recommended and can negatively affect the functionality of the PMM setup carried out by the Operator.

     When done, apply the edited deploy/cr.yaml file:

     $ kubectl apply -f deploy/cr.yaml
     

2. Check that corresponding Pods are not in a cycle of stopping and restarting. This cycle occurs if there are errors on the previous steps:

   $ kubectl get pods
   $ kubectl logs my-cluster-name-rs0-0 -c pmm-client
   

Contact Us

For free technical help, visit the Percona Community Forum.

To report bugs or submit feature requests, open a JIRA ticket.

For paid support and managed or consulting services , contact Percona Sales.

---

Last update: 2023-09-29