GlusterFS Containers with Docker, Kubernetes and Openshift

Thought of sharing consolidated news on GlusterFS containers efforts here. Below is the snip of the email which I sent few days back to gluster-users and gluster-devel mailing list. Hope it gives a summary, if not please let me know.


I would like to provide you a status update on the developments with GlusterFS containers and its presence in projects like docker, kubernetes and Openshift.

We have containerized GlusterFS with base image of CentOS and Fedora and its available at Docker Hub[1] to consume.

The Dockerfile of the Image can be found at github[2].

You can pull the image with


# docker pull gluster/gluster-centos
# docker pull gluster/gluster-fedora

The exact steps to be followed to run GlusterFS container is mentioned here[3].

We can deploy GlusterFS pods in Kubernetes Environment and an example blog about this setup can be found here [4].

There is GlusterFS volume plugin available in Kubernetes and openshift v3 which provides Persistent Volume
to the containers in the Environment, How to use GlusterFS containers for Persistent Volume and Persistent Volume Claim in Openshift has been recorded at [5].

[1]https://hub.docker.com/r/gluster/
[2]https://github.com/gluster/docker/
[3]http://tinyurl.com/jupgene
[4]http://tinyurl.com/zsrz36y
[5]http://tinyurl.com/hne8g7o

Please let us know if you have any comments/suggestions/feedback.

Persistent Volume and Claim in OpenShift and Kubernetes using GlusterFS Volume Plugin

OpenShift is a platform as a service product from Red Hat. The software that runs the service is open-sourced under the name OpenShift Origin, and is available on GitHub.

OpenShift v3 is a layered system designed to expose underlying Docker and Kubernetes concepts as accurately as possible, with a focus on easy composition of applications by a developer. For example, install Ruby, push code, and add MySQL.

Docker is an open platform for developing, shipping, and running applications. With Docker you can separate your applications from your infrastructure and treat your infrastructure like a managed application. Docker does this by combining kernel containerization features with workflows and tooling that help you manage and deploy your applications. Docker containers wrap up a piece of software in a complete filesystem that contains everything it needs to run: code, runtime, system tools, system libraries – anything you can install on a server. Available on GitHub.

Kubernetes is an open-source system for automating deployment, operations, and scaling of containerized applications. It groups containers that make up an application into logical units for easy management and discovery. Kubernetes builds upon a decade and a half of experience of running production workloads at Google, combined with best-of-breed ideas and practices from the community. Available on GitHub.

GlusterFS is a scalable network filesystem. Using common off-the-shelf hardware, you can create large, distributed storage solutions for media streaming, data analysis, and other data- and bandwidth-intensive tasks. GlusterFS is free and open source software. Available on GitHub.

Hope you know a little bit of all the above Technologies, now we jump right into our topic which is Persistent Volume and Persistent volume claim in Kubernetes and Openshift v3 using GlusterFS volume. So what is Persistent Volume? Why do we need it? How does it work using GlusterFS Volume Plugin?

In Kubernetes, Managing storage is a distinct problem from managing compute. The PersistentVolume subsystem provides an API for users and administrators that abstracts details of how storage is provided from how it is consumed. To do this we introduce two new API resources in kubernetes: PersistentVolume and PersistentVolumeClaim.

A PersistentVolume (PV) is a piece of networked storage in the cluster that has been provisioned by an administrator. It is a resource in the cluster just like a node is a cluster resource. PVs are volume plugins like Volumes, but have a lifecycle independent of any individual pod that uses the PV. This API object captures the details of the implementation of the storage, be that NFS, iSCSI, or a cloud-provider-specific storage system.

A PersistentVolumeClaim (PVC) is a request for storage by a user. It is similar to a pod. Pods consume node resources and PVCs consume PV resources. Pods can request specific levels of resources (CPU and Memory). Claims can request specific size and access modes (e.g, can be mounted once read/write or many times read-only).

In simple words, Containers in Kubernetes Cluster need some storage which should be persistent even if the container goes down or no longer needed. So Kubernetes Administrator creates a Storage(GlusterFS storage, In this case) and creates a PV for that storage. When a Developer (Kubernetes cluster user) needs a Persistent Volume in a container, creates a Persistent Volume claim. Persistent Volume Claim will contain the options which Developer needs in the pods. So from list of Persistent Volume the best match is selected for the claim and Binded to the claim. Now the developer can use the claim in the pods.


Prerequisites:

Need a Kubernetes or Openshift cluster, My setup is one master and three nodes.

Note: you can use kubectl in place of oc, oc is openshift controller which is a wrapper around kubectl. I am not sure about the difference.

2) Have a GlusterFS cluster setup, Create a GlusterFS Volume and start the GlusterFS volume.

3) All nodes in kubernetes cluster must have GlusterFS-Client Package installed.

Now we have the prerequisites \o/ …

In Kube-master administrator has to write required yaml file which will be given as input to the kube cluster.

There are three files to be written by administrator and one by Developer.

Service
Service Keeps the endpoint to be persistent or active.
Endpoint
Endpoint is the file which points to the GlusterFS cluster location.
PV
PV is Persistent Volume where the administrator will define the gluster volume name, capacity of volume and access mode.
PVC
PVC is persistent volume claim where developer defines the type of storage as needed.

STEP 1: Create a service for the gluster volume.

Verify:

STEP 2: Create an Endpoint for the gluster service

The ip here is the glusterfs cluster ip.

STEP 3: Create a PV for the gluster volume.

Note : path here is the gluster volume name. Access mode specifies the way to access the volume. Capacity has the storage size of the GlusterFS volume.

STEP 4: Create a PVC for the gluster PV.

Note: the Developer request for 8 Gb of storage with access mode rwx.

Here the pvc is bounded as soon as created, because it found the PV that satisfies the requirement. Now lets go and check the pv status

See now the PV has been bound to “default/glusterfs-claim”. In this state developer has the Persistent Volume Claim bounded successfully, now the developer can use the pv claim like below.

STEP 5: Use the persistent Volume Claim in a Pod defined by the Developer.

The above pod definition will pull the humble/gluster-client image(some private image) and start init script. The gluster volume will be mounted on the host machine by the GlusterFS volume Plugin available in the kubernetes and then bind mounted to the container’s /home. So all the Kubernetes cluster nodes must have glusterfs-client packages.

Lets try running.

Wow its running… lets go and check where it is running.

Found the Pod running successfully on one of the Kubernetes node.

On the host:

I can see the gluster volume being mounted on the host \o/. Lets check inside the container. Note the random number is the container-id from the docker ps command.

Yippy the GlusterFS volume has been mounted inside the container on /home as mentioned in the pod definition. Lets try writing something to it

Since the AccessMode is RWX I am able to write to the mount point.

That’s all Folks.

GlusterFS containers in Kubernetes Cluster for persistent data store !!

Everything is containerized, so Gluster . As you know, Gluster Container images are available for long time ( for both CentOS and Fedora ) in Docker hub. In previous blog posts, we saw how to build/run Gluster Containers. In this setup, we will try to setup a kubernetes cluster with Gluster containers. If you dont know much about kubernetes , please go through this . In short, kubernetes is an orchestration software for container environment which brings the services like scheduling, service discovery..etc. We will deploy a kubernetes cluster in couple of atomic nodes. Then run Gluster containers on these atomic hosts via kubernetes. Once the gluster containers are running, we will form a trusted pool out of these gluster containers and export a volume, so that other application containers can make use of this volume to store its data in persistent way!! .

kubernetes

Sounds interesting ? Yes, let us start.

NOTE: This article also discuss the steps to configure etcd server ( a key value store).
. For this particular setup we may not need to configure etcd. However your environment may need, for example to configure flannel.

Setup

Three centos ( You can also use fedora/RHEL) atomic hosts :


centos-atomic-KubeMaster
centos-atomic-Kubenode1
centos-atomic-Kubenode2

To configure/install CentOS atomic hosts, please follow the steps mentioned here.
and the atomic images can be downloaded from here

Then start the atomic installation, if cloud init is configured, it will come into play and ask for “atomic host” login.

username: centos
password: atomic

Note: Above is based on the cloud-init configuration. If you have customized the cloud init configuration for different username and password, please supply the same. (wait till the vm to completely load meta-data and user-data. else it will throw invalid login till its completely loaded)

At this stage we have three atomic hosts.:


10.70.42.184 centos-atomic-KubeMaster
10.70.42.29 centos-atomic-Kubenode1
10.70.43.88 centos-atomic-Kubenode2

If you already have this setup, make sure all the machines are able to talk to each other.

First things first,

-bash-4.2# atomic host upgrade

Upgrade your system to latest docker, etcd, kubernetes..etc, in all nodes.
With the three systems in place, the next thing is to set up Kubernetes. Setting up Kubernetes on the Master, select any system to be master.

1. Etcd configuration:
Edit the /etc/etcd/etcd.conf. The etcd service needs to be configured to listen on all interfaces to ports 2380. (ETCD_LISTEN_PEER_URLS) and port 2379 (ETCD_LISTEN_CLIENT_URLS), and listen on 2380 on localhost (ETCD_LISTEN_PEER_URLS)

-bash-4.2# cat /etc/etcd/etcd.conf | grep -v "#"
ETCD_NAME=default
ETCD_DATA_DIR="/var/lib/etcd/default.etcd"
ETCD_LISTEN_PEER_URLS="http://0.0.0.0:2380"
ETCD_LISTEN_CLIENT_URLS="http://0.0.0.0:2379"
ETCD_ADVERTISE_CLIENT_URLS="http://0.0.0.0:2379"

2. Kubernetes Configuration:

Edit the /etc/kubernetes/config file and change the KUBE_MASTER line to identify the location of your master server (it points to 127.0.0.1, by default). Leave other settings as they are.

KUBE_MASTER="--master=http://10.70.42.184:8080"

3. Kubernetes apiserver Configuration:

Edit the /etc/kubernetes/apiserver and add a new KUBE_ETCD_SERVERS line (as shown below), then review and change other lines in the apiserver configuration file. Change KUBE_API_ADDRESS to listen on all network addresses(0.0.0.0), instead of just localhost. Set an address range for the KUBE_SERVICE_ADDRESS that Kubernetes can use to assign to services (see a description of this address below). Finally, remove the term “ServiceAccount” from the KUBE_ADMISSION_CONTROL instruction.


-bash-4.2# cat /etc/kubernetes/apiserver | grep -v "#"
KUBE_API_ADDRESS="--address=0.0.0.0"
KUBE_ETCD_SERVERS="--etcd_servers=http://10.70.42.184:2379"
KUBE_SERVICE_ADDRESSES="--service-cluster-ip-range=10.254.100.0/24"
KUBE_ADMISSION_CONTROL="--admission_control=NamespaceLifecycle,NamespaceExists,LimitRanger,SecurityContextDeny,ResourceQuota"
KUBE_API_ARGS=""

4. Start master services:

To run the Kubernetes master services, you need to enable and start several systemd services. From the master, run the following for loop to start and enable Kubernetes systemd services on the master:


-bash-4.2# for SERVICES in etcd kube-apiserver kube-controller-manager kube-scheduler; do
systemctl restart $SERVICES;
systemctl enable $SERVICES;
systemctl status $SERVICES;
done

5. Setting up Kubernetes on the Nodes

On each of the two Kubernetes nodes, you need to edit several configuration files and start and enable several Kubernetes systemd services:

1.Edit /etc/kubernetes/config:

Edit the KUBE_MASTER line in this file to identify the location of your master (it is 127.0.0.1, by default). allow_privileged must be set to true. Leave other settings as they are.


KUBE_ALLOW_PRIV="--allow_privileged=true"
KUBE_MASTER="--master=http://10.70.42.184:8080"

2.Edit /etc/kubernetes/kubelet:

In this file on each node, modify KUBELET_ADDRESS (0.0.0.0 to listen on all network interfaces), KUBELET_HOSTNAME (replace hostname_override with the hostname or IP address of the local system). You may leave this blank to use the actual hostname, set KUBELET_ARGS, and KUBELET_API_SERVER as below. --host-network-sources=* is specified to use the host networking option of docker(–net=host). You can use any networking mode of docker. However in this setup, we use --net=host option to make sure we get maximum performance.

-bash-4.2# cat /etc/kubernetes/kubelet | grep -v "#"
KUBELET_ADDRESS="--address=0.0.0.0"
KUBELET_HOSTNAME="--hostname_override="
KUBELET_API_SERVER="--api_servers=http://10.70.42.184:8080"
KUBELET_ARGS="--register-node=true --host-network-sources=*"

3. Edit /etc/kubernetes/proxy:
No settings are required in this file. If you have set KUBE_PROXY_ARGS, you can comment it out:

-bash-4.2# cat /etc/kubernetes/proxy
###
# kubernetes proxy config
# default config should be adequate
# Add your own!
#KUBE_PROXY_ARGS="--master=http://master.example.com:8080"

4. Start the Kubernetes nodes systemd services:

On each node, you need to start several services associated with a Kubernetes node:


-bash-4.2# for SERVICES in docker kube-proxy.service kubelet.service; do
systemctl restart $SERVICES;
systemctl enable $SERVICES;
systemctl status $SERVICES; done

5. Check the services:
Run the netstat command on each of the three systems to check which ports the services are running on. The etcd service should only be running on the master.

From master:


-bash-4.2# netstat -tulnp | grep -E "(kube)|(etcd)"
tcp 0 0 127.0.0.1:10251 0.0.0.0:* LISTEN 17805/kube-schedule
tcp 0 0 127.0.0.1:10252 0.0.0.0:* LISTEN 17764/kube-controll
tcp6 0 0 :::6443 :::* LISTEN 17833/kube-apiserve
tcp6 0 0 :::2379 :::* LISTEN 17668/etcd
tcp6 0 0 :::2380 :::* LISTEN 17668/etcd
tcp6 0 0 :::8080 :::* LISTEN 17833/kube-apiserve

From nodes:


-bash-4.2# netstat -tulnp | grep kube
tcp 0 0 127.0.0.1:10248 0.0.0.0:* LISTEN 104398/kubelet
tcp 0 0 127.0.0.1:10249 0.0.0.0:* LISTEN 104331/kube-proxy
tcp6 0 0 :::10250 :::* LISTEN 104398/kubelet
tcp6 0 0 :::57421 :::* LISTEN 104331/kube-proxy
tcp6 0 0 :::10255 :::* LISTEN 104398/kubelet
tcp6 0 0 :::34269 :::* LISTEN 104331/kube-proxy
tcp6 0 0 :::58239 :::* LISTEN 104331/kube-proxy
tcp6 0 0 :::4194 :::* LISTEN 104398/kubelet
Continue reading "GlusterFS containers in Kubernetes Cluster for persistent data store !!"

“Running with unpopulated /etc” – Failing to run systemd based container ?

Recently I experienced, systemd based container fails to run in certain version of distros.

For ex: If I run my container with systemd I get below messages.

#docker run –rm -t -i –privileged -v /sys/fs/cgroup:/sys/fs/cgroup:ro

systemd 219 running in system mode. (+PAM +AUDIT +SELINUX +IMA -APPARMOR +SMACK +SYSVINIT +UTMP +LIBCRYPTSETUP +GCRYPT +GNUTLS +ACL +XZ -LZ4 -SECCOMP +BLKID +ELFUTILS +KMOD +IDN)
Detected virtualization docker.
Detected architecture x86-64.
Running with unpopulated /etc.
....
Set hostname to .
Initializing machine ID from random generator.
Populated /etc with preset unit settings.
Unit etc-hosts.mount is bound to inactive unit dev-mapper-X.X.root.device. Stopping, too.
Unit etc-hostname.mount is bound to inactive unit dev-mapper-X.X.root.device. Stopping, too....
Unit etc-resolv.conf.mount is bound to inactive unit dev-mapper-X.X.root.device. Stopping, too.
Cannot add dependency job for unit display-manager.service, ignoring: Unit display-manager.service failed to load: No such file or directory.
Startup finished in 106ms.
Failed to create unit file /run/systemd/generator.late/network.service: File exists
Failed to create unit file /run/systemd/generator.late/netconsole.service: File exists

The line ‘Running with unpopulated /etc’ looked suspicious to me, after some attempts we were able to conclude that, the things were going wrong in absense of ‘/etc/machine-id’ file which used to be there. If you came across similar to this situation, make an entry in your docker file to create /etc/machine-id as shown below and give a try!

#RUN touch /etc/machine-id

now, build your image and start the container from new image.. Let me know how it goes.

Docker gotchas ( 1. ENTRYPOINT Vs CMD in Dockerfile)

Well, most of the folks are really confused about the difference between ENTRYPOINT and CMD, so you.
Its simple. Whatever you fill in CMD will be given to the ENTRYPOINT as an argument.

For ex: If you have a dockerfile like this.

CMD ["-tupln"]
ENTRYPOINT ["/usr/bin/netstat"]

and when you run the container without any other command ( For ex:#docker run ….. )
The above instruction basically make it something like “netstat -tulpn” for the container to execute.
Also there is a default entrypoint for each container which is nothing but “/bin/sh -c”
. However note that, the CMD used at end of the ” #docker run … .. ” will overwrite the default command.

You also need to pay some attention on the syntax you use for CMD or ENTRYPOINT. The difference when you use array syntax ( CMD [” “] or shell syntax ( CMD ls -l) is nothing but, if you are not using the array syntax the arguments are prepended with the default entry point which is nothing but “/bin/sh -c”. It may generate unexpected result. So, its always adviceable to use array syntax when using CMD or ENTRYPOINT.