Developing a microservice on your laptop and want to quickly size the application container using a test load ? Run the Kruize container locally and point it to your application container. Kruize monitors the app container using Prometheus and provides recommendations as a Grafana dashboard (Prometheus and Grafana containers are automatically downloaded when you run kruize).
$ ./deploy.sh -c docker
### Installing kruize for docker...
Info: Checking pre requisites for Docker...
...
Waiting for kruize container to come up
######################## Starting App Monitor loop #########################
Kruize recommendations available on the grafana dashboard at: http://localhost:3000
Info: Press CTRL-C to exit
cadvisor: found. Adding to list of containers to be monitored.
grafana: found. Adding to list of containers to be monitored.
kruize: found. Adding to list of containers to be monitored.
prometheus: found. Adding to list of containers to be monitored.
Now edit manifests/docker/kruize-docker.yaml to add the names of the containers that you need kruize to monitor.
$ cat manifests/docker/kruize-docker.yaml
---
# Add names of the containers that you want kruize to monitor, one per line in double quotes
containers:
- name: "cadvisor"
- name: "grafana"
- name: "kruize"
- name: "prometheus"
- name: "acmeair-mono-app1"
- name: "acmeair-db1"
In the above example, kruize is monitoring the application containers acmeair-mono-app1 and acmeair-db1 as well as its own set of containers. You should now see the "App Monitor loop" listing the new containers to be monitored
cadvisor: found. Adding to list of containers to be monitored.
grafana: found. Adding to list of containers to be monitored.
kruize: found. Adding to list of containers to be monitored.
prometheus: found. Adding to list of containers to be monitored.
acmeair-mono-app1: found. Adding to list of containers to be monitored.
acmeair-db1: found. Adding to list of containers to be monitored.
You can now access the grafana dashboard at http://localhost:3000. Login as admin/admin and click on the pre-installed Kruize Dashboard. Select the application name from the Deployment drop down and you are all set !
Kruize can be deployed to a supported Kubernetes cluster. We currently support Minikube, IBM Cloud Private (ICP) and OpenShift.
$ ./deploy.sh -c minikube
### Installing kruize for minikube
Info: Checking pre requisites for minikube...
Info: kruize needs cadvisor/prometheus/grafana to be installed in minikube
Download and install these software to minikube(y/n)? y <----- Say yes to install cadvisor/prometheus/grafana
Info: Downloading cadvisor git
...
Info: Downloading prometheus git
Info: Installing prometheus
...
Info: Waiting for all Prometheus Pods to get spawned......done
Info: Waiting for prometheus-k8s-1 to come up...
prometheus-k8s-1 2/3 Running 0 5s
Info: prometheus-k8s-1 deploy succeeded: Running
prometheus-k8s-1 2/3 Running 0 6s
Info: One time setup - Create a service account to deploy kruize
serviceaccount/kruize-sa created
clusterrole.rbac.authorization.k8s.io/kruize-cr created
clusterrolebinding.rbac.authorization.k8s.io/kruize-crb created
servicemonitor.monitoring.coreos.com/kruize created
prometheus.monitoring.coreos.com/prometheus created
Info: Deploying kruize yaml to minikube cluster
deployment.apps/kruize created
service/kruize created
Info: Waiting for kruize to come up...
kruize-695c998775-vv4dn 0/1 ContainerCreating 0 4s
kruize-695c998775-vv4dn 1/1 Running 0 9s
Info: kruize deploy succeeded: Running
kruize-695c998775-vv4dn 1/1 Running 0 9s
Info: Access grafana dashboard to see kruize recommendations at http://localhost:3000 <--- Click on this link to access grafana dashboards
Info: Run the following command first to access grafana port
$ kubectl port-forward -n monitoring grafana-58dc7468d7-rn7nx 3000:3000 <---- But run this command first
After the installation completes successfully, run the port-forward command as shown. This is needed to access the grafana service. Now click on the http://localhost:3000 to access grafana. Login as admin/admin, navigate to Create from the left bar, then click on Import. Click on Upload .json file and point it to the dashboard file that is part of this repo. Once installed, select Kruize Dashboard. Select the application name from the Deployment drop down and you are all set !
Note: Kruize only monitors application with a specific label. See Add Application Label below for more info.
$ ./deploy.sh -c openshift
### Installing kruize for OpenShift
WARNING: This will create a Kruize ServiceMonitor object in the openshift-monitoring namespace
WARNING: This is currently not recommended for production
Create ServiceMonitor object and continue installation?(y/n)? y
Info: Checking pre requisites for OpenShift...done
Info: Logging in to OpenShift cluster...
Authentication required for https://aaa.bbb.com:6443 (openshift)
Username: kubeadmin
Password:
Login successful.
You have access to 52 projects, the list has been suppressed. You can list all projects with 'oc projects'
Using project "kube-system".
Info: Setting Prometheus URL as https://prometheus-k8s-openshift-monitoring.apps.kaftans.os.fyre.ibm.com
Info: Deploying kruize yaml to OpenShift cluster
Now using project "openshift-monitoring" on server "https://api.kaftans.os.fyre.ibm.com:6443".
deployment.extensions/kruize configured
service/kruize unchanged
Info: Waiting for kruize to come up...
kruize-5cd5967d97-tz2cb 0/1 ContainerCreating 0 6s
kruize-5cd5967d97-tz2cb 0/1 ContainerCreating 0 13s
kruize-5cd5967d97-tz2cb 1/1 Running 0 20s
Info: kruize deploy succeeded: Running
kruize-5cd5967d97-tz2cb 1/1 Running 0 24s
Now you need to install the Kruize Dashboard, see the Install the Grafana dashboard section for more details.
Note: Kruize only monitors application with a specific label. See Add Application Label below for more info.
$ ./deploy.sh -c icp
Info: Checking pre requisites for ICP...done
Info: Logging in to ICP cluster...
API endpoint: https://aaa.bbb.com:8443
Username> admin
Password>
Authenticating...
OK
Targeted account mycluster Account (id-mycluster-account)
Select a namespace:
1. cert-manager
2. default
3. istio-system
4. kube-public
5. kube-system <---- Choose the kube-system namespace.
6. platform
7. services
Enter a number> 5
Targeted namespace kube-system
Configuring kubectl ...
Property "clusters.mycluster" unset.
Property "users.mycluster-user" unset.
Property "contexts.mycluster-context" unset.
Cluster "mycluster" set.
User "mycluster-user" set.
Context "mycluster-context" created.
Switched to context "mycluster-context".
OK
Configuring helm: /home/dino/.helm
OK
Info: Setting Prometheus URL as https://aaa.bbb.com:8443/prometheus
Info: Deploying kruize yaml to ICP cluster
deployment.extensions/kruize configured
service/kruize unchanged
kruize-866d48ddd-5qw2v 1/1 Running 0 10s
From the ICP main menu navigate to Grafana (Eg Menu -> Platform -> Monitoring). Now you need to install the Kruize Dashboard, see Install the Grafana dashboard for more details.
Note: Kruize only monitors application with a specific label. See Add Application Label below for more info.
kruize is now ready to monitor applications in your cluster ! Note however that kruize only monitors applications with the label app.kubernetes.io/name: "myapp" currently.
spec:
replicas: 1
template:
metadata:
labels:
app: myapp
name: myapp
app.kubernetes.io/name: "myapp" <--- Add this label to your app yaml
Login to your Grafana dashboard and navigate to Create from the left bar, then click on Import. Click on Upload .json file and point it to the dashboard file that is part of this repo.
Once imported, the grafana dashboard should look something like this.
Once installed, select Kruize Dashboard. Select the application name from the Deployment drop down and you are all set !
Kruize uses slf4j and the log4j-slf4j binding for its logging. The log levels used are:
| Logging Level | Description |
|---|---|
ERROR |
Error events that stop application from running correctly. |
WARN |
Designates potentially harmful situations. Includes ERROR logs. |
INFO |
Informational messages that highlight the progress of the application. Includes ERROR and WARN logs. The default logging level. |
DEBUG |
Designates fine-grained informational events that are most useful to debug an application. Includes logs from INFO level. |
ALL |
Turn on all logging. |
By default, the log level is set to INFO. To change the logging level, set the ENV LOGGING_LEVEL in manifests/kruize.yaml_template to any of the above levels.
- name: MONITORING_AGENT
value: "prometheus"
- name: MONITORING_SERVICE
value: "{{ MONITORING_SERVICE }}"
- name: LOGGING_LEVEL
value: "INFO" <--- Change this to any of the above levels
While submitting issues, we recommend users to attach the logs with log level set to DEBUG.
$ ./build.sh
Tag it appropriately and push it to a docker registry that is accessible to the kubernetes cluster. Don't forget to update the manifest yaml's to point to your newly built image !