Skip to content

Latest commit

 

History

History
28 lines (17 loc) · 1.37 KB

File metadata and controls

28 lines (17 loc) · 1.37 KB

In this directory, you can deploy EFK (elasticsearch, fluentd, kibana) log monitoring and visualization system with one command. With EFK, you can monitor the application log of Train Ticket.

The specific deployment process is as follows:

Prerequisites:

You need to have set up a Kubernetes cluster and deployed the train ticket system in it.
Then, set the virtual machine size of all nodes in the cluster, just execute the command:
sysctl -w vm.max_map_count=655300

Deploy:

OK, let's officially start deploying EFK with one command.
After entering the directory where this file is located, we only need to execute the runEFK.sh script:
bash runEFK.sh

Note: You may make sure that these pods are in running status. Otherwise you may need to deal with these problems (such as manually pulling the image), then uninstall EFK, and then redeploy EFK.

Access Kibana UI:

Then you can see that the kibana service has been exposed on port 8086. Just wait a few minutes, and we can access the interface of kibana through the browser at
http://ipAddress:8086/api/v1/namespaces/kube-system/services/kibana-logging/proxy .

Uninstall:

When you want to delete EFK in the cluster,you just need to execute the deleteEFK.sh script:
bash deleteEFK.sh