USE CEPH AS PERSISTENT STORAGE FOR NOMAD

SUMMARY :

To orchestrate containers, Nomad is a a more and more popular choice, it supports CSI plugins so ceph-csi can directly be used.

We’ll see how to use Ceph (RBD) to store container’s data.

CONTENT

REQUIREMENTS

• Nomad cluster (v1.1.1)
• Ceph cluster (octopus 15.2.10)
• ceph-csi (3.3.1)
• Docker (20.10.7)

CONFIGURE NOMAD

We’ll use docker and need to allow containers to use priviliged mode.

Edit /etc/nomad.d/nomad.hcl and it’s plugin section.

plugin "docker" {
  config {
    allow_privileged = true
  }
}

Restart nomad systemctl restart nomad

CREATE CEPH CSI CONTROLLER AND PLUGIN NODE

The ceph-csi plugin requieres two components:

• The controller plugin communicates with provider’s API.
• The node plugin executes tasks on client like managing mounting points.

CEPH-CSI-PLUGIN-CONTROLLER.NOMAD

You need to replace :

• e585be01-47aa-44bc-9a58-55415b3f4225 by your ceph fsuid
• Monitors IP address by yours

job "ceph-csi-plugin-controller" {
  datacenters = ["dc1"]
group "controller" {
    network {
      port "metrics" {}
    }
    task "ceph-controller" {
template {
        data        = <<EOF
[{
    "clusterID": "e585be01-47aa-44bc-9a58-55415b3f4225",
    "monitors": [
        "192.168.112.5",
  "192.168.112.6",
  "192.168.112.7"
    ]
}]
EOF
        destination = "local/config.json"
        change_mode = "restart"
      }
      driver = "docker"
      config {
        image = "quay.io/cephcsi/cephcsi:v3.3.1"
        volumes = [
          "./local/config.json:/etc/ceph-csi-config/config.json"
        ]
        mounts = [
          {
            type     = "tmpfs"
            target   = "/tmp/csi/keys"
            readonly = false
            tmpfs_options = {
              size = 1000000 # size in bytes
            }
          }
        ]
        args = [
          "--type=rbd",
          "--controllerserver=true",
          "--drivername=rbd.csi.ceph.com",
          "--endpoint=unix://csi/csi.sock",
          "--nodeid=${node.unique.name}",
    "--instanceid=${node.unique.name}-controller",
          "--pidlimit=-1",
    "--logtostderr=true",
          "--v=5",
          "--metricsport=$${NOMAD_PORT_metrics}"
        ]
      }
   resources {
        cpu    = 500
        memory = 256
      }
      service {
        name = "ceph-csi-controller"
        port = "metrics"
        tags = [ "prometheus" ]
      }
csi_plugin {
        id        = "ceph-csi"
        type      = "controller"
        mount_dir = "/csi"
      }
    }
  }
}

CEPH-CSI-PLUGIN-NODES.NOMAD

job "ceph-csi-plugin-nodes" {
  datacenters = ["dc1"]
  type        = "system"
  group "nodes" {
    network {
      port "metrics" {}
    }

    task "ceph-node" {
      driver = "docker"
      template {
        data        = <<EOF
[{
    "clusterID": "e585be01-47aa-44bc-9a58-55415b3f4225",
    "monitors": [
        "192.168.112.5",
  "192.168.112.6",
  "192.168.112.7"
    ]
}]
EOF
        destination = "local/config.json"
        change_mode = "restart"
      }
      config {
        image = "quay.io/cephcsi/cephcsi:v3.3.1"
        volumes = [
          "./local/config.json:/etc/ceph-csi-config/config.json"
        ]
        mounts = [
          {
            type     = "tmpfs"
            target   = "/tmp/csi/keys"
            readonly = false
            tmpfs_options = {
              size = 1000000 # size in bytes
            }
          }
        ]
        args = [
          "--type=rbd",
          "--drivername=rbd.csi.ceph.com",
          "--nodeserver=true",
          "--endpoint=unix://csi/csi.sock",
          "--nodeid=${node.unique.name}",
          "--instanceid=${node.unique.name}-nodes",
          "--pidlimit=-1",
    "--logtostderr=true",
          "--v=5",
          "--metricsport=$${NOMAD_PORT_metrics}"
        ]
        privileged = true
      }
   resources {
        cpu    = 500
        memory = 256
      }
      service {
        name = "ceph-csi-nodes"
        port = "metrics"
        tags = [ "prometheus" ]
      }
csi_plugin {
        id        = "ceph-csi"
        type      = "node"
        mount_dir = "/csi"
      }
    }
  }
}

We can now start them

nomad job run ceph-csi-plugin-controller.nomad
nomad job run ceph-csi-plugin-nodes.nomad

We can then directly check for the plugin status. It can take few minutes as it has to download image.

root@nomad112 ~/nomadceph $ nomad plugin status ceph-csi
ID                   = ceph-csi
Provider             = rbd.csi.ceph.com
Version              = 3.3.1
Controllers Healthy  = 1
Controllers Expected = 1
Nodes Healthy        = 1
Nodes Expected       = 1

Allocations
ID        Node ID   Task Group  Version  Desired  Status   Created    Modified
23b4db0c  a61ef171  nodes       4        run      running  3h26m ago  3h25m ago
fee74115  a61ef171  controller  6        run      running  3h26m ago  3h25m ago

To be able to use ceph, our host must have the rbd module added to the kernel.

sudo lsmod |grep rbd
rbd                    94208  2
libceph               364544  1 rbd

If it’s not the case, just load it by running sudo modprobe rbd

Since nomad 1.1.0 we can directly create RBD images with nomad.

CEPH-VOLUME.HCL

You need to replace :

• e585be01-47aa-44bc-9a58-55415b3f4225 by your ceph fsuid vim ceph-csi-plugin-controller.nomad
• Monitors IP address by yours
• admin with your ceph user
• AQAVXIlgv6krBhAATxGcAZZgOwW7mW9HdSPJxQ== with the user key

id = "ceph-mysql"
name = "ceph-mysql"
type = "csi"
plugin_id = "ceph-csi"
capacity_max = "200G"
capacity_min = "100G"

capability {
  access_mode     = "single-node-writer"
  attachment_mode = "file-system"
}

secrets {
  userID  = "admin"
  userKey = "AQAVXIlgv6krBhAATxGcAZZgOwW7mW9HdSPJxQ=="
}

parameters {
  clusterID = "e585be01-47aa-44bc-9a58-55415b3f4225"
  pool = "rbd"
  imageFeatures = "layering"
}

Then we create it.

nomad volume create ceph-volume.hcl
Created external volume 0001-0024-e585be01-47aa-44bc-9a58-55415b3f4225-0000000000000002-100fc248-d011-11eb-b9f0-0242ac110002 with ID ceph-mysql

We’ll reuse the job from Hashicorp learn but edit it to use our RBD volume.

MYSQL.NOMAD

job "mysql-server" {
  datacenters = ["dc1"]
  type        = "service"

  group "mysql-server" {
    count = 1

    volume "ceph-mysql" {
      type      = "csi"
        attachment_mode = "file-system"
        access_mode     = "single-node-writer"
      read_only = false
      source    = "ceph-mysql"
    }

    network {
      port "db" {
        static = 3306
      }
    }

    restart {
      attempts = 10
      interval = "5m"
      delay    = "25s"
      mode     = "delay"
    }

    task "mysql-server" {
      driver = "docker"

      volume_mount {
        volume      = "ceph-mysql"
        destination = "/srv"
        read_only   = false
      }

      env {
        MYSQL_ROOT_PASSWORD = "password"
      }

      config {
        image = "hashicorp/mysql-portworx-demo:latest"
        args  = ["--datadir", "/srv/mysql"]
        ports = ["db"]
      }

      resources {
        cpu    = 500
        memory = 1024
      }

      service {
        name = "mysql-server"
        port = "db"

        check {
          type     = "tcp"
          interval = "10s"
          timeout  = "2s"
        }
      }
    }
  }
}

We can now start our job

nomad job run mysql.nomad

Check for job status to get it’s alloc, then list the items table.

nomad job status mysql-server
...
Status        = running
...
Allocations
ID        Node ID   Task Group    Version  Desired  Status   Created  Modified
38070da7  9ad01c63  mysql-server  0        run      running  6s ago   3s ago

nomad alloc exec 38070da7 sh
# mysql -u root -p -D itemcollection
Enter password:
...
mysql> select * from items;
+----+----------+
| id | name     |
+----+----------+
|  1 | bike     |
|  2 | baseball |
|  3 | chair    |
+----+----------+

Let’s add some items

mysql> INSERT INTO items (name) VALUES ('glove');
mysql> INSERT INTO items (name) VALUES ('hat');
mysql> INSERT INTO items (name) VALUES ('keyboard');
mysql> select * from items;
+----+----------+
| id | name     |
+----+----------+
|  1 | bike     |
|  2 | baseball |
|  3 | chair    |
|  4 | glove    |
|  5 | hat      |
|  6 | keyboard |
+----+----------+

To ensure our DB is on persistent storage, we’ll delete the container and rebuild it with the same job.

nomad stop -purge mysql-server
nomad job run mysql.nomad

We check if data we are added are still here.

nomad alloc exec d8a401a0 sh
mysql -u root -p -D itemcollection
mysql> select * from items;
+----+----------+
| id | name     |
+----+----------+
|  1 | bike     |
|  2 | baseball |
|  3 | chair    |
|  4 | glove    |
|  5 | hat      |
|  6 | keyboard |
+----+----------+

Like we created the image with nomad, we can delete it.

We first need to delete and purge the job.

nomad stop -purge mysql-server

Then we can delete the volume.

nomad volume delete ceph-mysql
Successfully deleted volume "ceph-mysql"!

In this example we used RBD but we could have use CephFS in the same way.

Ressources

• https://www.nomadproject.io/docs/commands/volume/create
• https://github.com/ceph/ceph-csi/blob/devel/docs/deploy-rbd.md
• https://learn.hashicorp.com/tutorials/nomad/stateful-workloads-csi-volumes?in=nomad/stateful-workloads
• https://itnext.io/provision-volumes-from-external-ceph-storage-on-kubernetes-and-nomad-using-ceph-csi-7ad9b15e9809
• Florian Apolloner for helping me on Nomad’s gitter.