github/kubernetes
1
0
Fork 0
mirror of https://github.com/k3s-io/kubernetes.git synced 2025-07-29 22:46:12 +00:00
Code Issues Projects Releases Wiki Activity

fix sentence in overview.md

Browse Source
This commit is contained in:
Phaneendra Chiruvella 2015-03-16 09:44:13 +05:30
parent cb66e553d2
commit 6b474e6100
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
docs/overview.md
View File

@ -4,7 +4,7 @@ Kubernetes is an open-source system for managing containerized applications (cur
Today, Kubernetes focuses on continuously-running stateless (e.g. web server or in-memory object cache) and "cloud native" stateful applications (e.g. NoSQL datastores), but in the near future it will support all the other workload types commonly found in production cluster environments, such as batch, stream processing, and traditional databases.
In Kubernetes, all containers run inside [pods](pods.md). A pod can host a single container, or multiple cooperating containers; in the latter case, the containers in the pod are guaranteed to be co-located on the same machine and can share resources. A pod can also contain zero or more [volumes](volumes.md), which are directories that are private to a container or shared across containers in a pod. For each pod the user creates, the system finds a machine that is healthy and that has sufficiently availability capacity, and starts up the corresponding container(s) there. If a container fails it can be automatically restarted by Kubernetes' node agent, called the Kubelet. But if the pod or its machine fails, it is not automatically moved or restarted unless the user also defines a [replication controller](replication-controller.md), which we discuss next.
In Kubernetes, all containers run inside [pods](pods.md). A pod can host a single container, or multiple cooperating containers; in the latter case, the containers in the pod are guaranteed to be co-located on the same machine and can share resources. A pod can also contain zero or more [volumes](volumes.md), which are directories that are private to a container or shared across containers in a pod. For each pod the user creates, the system finds a machine that is healthy and that has sufficient availabile capacity, and starts up the corresponding container(s) there. If a container fails it can be automatically restarted by Kubernetes' node agent, called the Kubelet. But if the pod or its machine fails, it is not automatically moved or restarted unless the user also defines a [replication controller](replication-controller.md), which we discuss next.
Users can create and manage pods themselves, but Kubernetes drastically simplifies system management by allowing users to delegate two common pod-related activities: deploying multiple pod replicas based on the same pod configuration, and creating replacement pods when a pod or its machine fails. The Kubernetes API object that manages these behaviors is called a [replication controller](replication-controller.md). It defines a pod in terms of a template, that the system then instantiates as some number of pods (specified by the user). The replicated set of pods might constitute an entire application, a micro-service, or one layer in a multi-tier application. Once the pods are created, the system continually monitors their health and that of the machines they are running on; if a pod fails due to a software problem or machine failure, the replication controller automatically creates a new pod on a healthy machine, to maintain the set of pods at the desired replication level. Multiple pods from the same or different applications can share the same machine. Note that a replication controller is needed even in the case of a single non-replicated pod if the user wants it to be re-created when it or its machine fails.