diff --git a/sysadmin/grid5000/cassandra/03-deploy_nodes.sh b/sysadmin/grid5000/cassandra/03-deploy_nodes.sh
index a947f9f..be57a20 100755
--- a/sysadmin/grid5000/cassandra/03-deploy_nodes.sh
+++ b/sysadmin/grid5000/cassandra/03-deploy_nodes.sh
@@ -1,33 +1,35 @@
 #!/usr/bin/env bash
 
-set -eu
+set -eux
 
 SCRIPT_DIR="$(pwd $(dirname @0))"
 
 source "${SCRIPT_DIR}/environment.cfg"
 
 echo "########### Nodes:"
 uniq "${OAR_FILE_NODES}"
 echo "########### Installing os on nodes"
 
 INSTALLED_OS_STAMP="${OAR_JOB_ID}.os.stamp"
 
 if [ ! -e "${SCRIPT_DIR}/${INSTALLED_OS_STAMP}" ]; then
     ${SCRIPT_DIR}/_install_os.sh
     touch "${SCRIPT_DIR}/${INSTALLED_OS_STAMP}"
 fi
 
 uniq "${OAR_NODE_FILE}" > ${SCRIPT_DIR}/nodes.lst
 
-echo "${CASSANDRA_HOSTS}" | sed 's/ /,/' > ${SCRIPT_DIR}/cassandra_seeds.lst
+NODE_COUNT="$(wc -l ${SCRIPT_DIR}/nodes.lst | cut -f1 -d' ')"
 
-time parallel --halt now,fail=1 rsync -avP  . "${SSH_USER}"@{}:install < ${SCRIPT_DIR}/nodes.lst
+echo "${CASSANDRA_HOSTS}" | sed 's/ /,/g' > ${SCRIPT_DIR}/cassandra_seeds.lst
 
-time parallel --halt now,fail=1 -u ssh ${SSH_OPTIONS} "${SSH_USER}"@{} install/_provision_node.sh < ${SCRIPT_DIR}/nodes.lst
+time parallel -j${NODE_COUNT} rsync -avP  . "${SSH_USER}"@{}:install < ${SCRIPT_DIR}/nodes.lst
+
+time parallel -j${NODE_COUNT} -u ssh ${SSH_OPTIONS} "${SSH_USER}"@{} install/_provision_node.sh < ${SCRIPT_DIR}/nodes.lst
 
 echo "########### Cassandra installation done"
 touch ${SCRIPT_DIR}/nodes.installed
 
 # The script must not exit to avoid the oar job to be killed
 echo "########### Sleeping"
 sleep infinity
diff --git a/sysadmin/grid5000/cassandra/Vagrantfile b/sysadmin/grid5000/cassandra/Vagrantfile
index 4beb88e..9be1e32 100644
--- a/sysadmin/grid5000/cassandra/Vagrantfile
+++ b/sysadmin/grid5000/cassandra/Vagrantfile
@@ -1,82 +1,88 @@
 # -*- mode: ruby -*-
 # vi: set ft=ruby :
 
 vms = {
     "cassandra1" => {
         :ip          => "10.168.180.11",
         :memory      => 2048,
         :cpus        => 2,
         :type       => 'cassandra',
     },
     "cassandra2" => {
         :ip          => "10.168.180.12",
         :memory      => 2048,
         :cpus        => 2,
         :type       => 'cassandra',
     },
     "cassandra3" => {
         :ip          => "10.168.180.13",
         :memory      => 2048,
         :cpus        => 2,
         :type       => 'cassandra',
     },
     "swh-storage1" => {
         :ip          => "10.168.180.14",
         :memory      => 1024,
         :cpus        => 2,
         :type       => 'swh-storage',
     },
+    "monitoring1" => {
+        :ip          => "10.168.180.15",
+        :memory      => 1024,
+        :cpus        => 2,
+        :type       => 'monitoring',
+    },
 }  
 
 # Images/remote configuration
 $global_debian10_box = "debian10-20210517-1348"
 $global_debian10_box_url = "https://annex.softwareheritage.org/public/isos/libvirt/debian/swh-debian-10.9-amd64-20210517-1348.qcow2"
 
 vms.each { | vm_name, vm_props |
 
     Vagrant.configure("2") do |global_config|
         unless Vagrant.has_plugin?("libvirt")
             $stderr.puts <<-MSG
             vagrant-libvirt plugin is required for this.
             To install: `$ sudo apt install vagrant-libvirt
         MSG
             exit 1
         end
         
         global_config.vm.define vm_name do |config|
             config.vm.box                     = $global_debian10_box
             config.vm.box_url                 = $global_debian10_box_url
             config.vm.box_check_update        = false
             config.vm.hostname                = vm_name
             config.vm.network   :private_network, ip: vm_props[:ip], netmask: "255.255.0.0"
 
             config.vm.synced_folder ".", "/vagrant", type: 'nfs', nfs_version: 4
 
             config.vm.provision :ansible do |ansible|
                 ansible.verbose = true
                 ansible.become = true
                 ansible.playbook = "ansible/playbook.yml"
                 ansible.inventory_path = "ansible/hosts.yml"
                 ansible.raw_arguments  = [
                     "-v",
                     "--connection=paramiko",
                     "--private-key=/home/.../.vagrant/machines/.../private_key",
                     "--extra-vars=@.credentials",
                     "--vault-password-file=.vault_password"
                   ]
             end
 
             config.vm.provider :libvirt do |provider|
                 provider.memory = vm_props[:memory]
                 provider.cpus = vm_props[:cpus]
                 provider.driver = 'kvm'
                 if vm_props[:type] == "cassandra"
                     provider.storage :file, :size => '1G'
                     provider.storage :file, :size => '1G'
                     provider.storage :file, :size => '1G'
                 end
     
             end
         end
     end
 }
diff --git a/sysadmin/grid5000/cassandra/_install_os.sh b/sysadmin/grid5000/cassandra/_install_os.sh
index 0dc03a2..5e05fd4 100755
--- a/sysadmin/grid5000/cassandra/_install_os.sh
+++ b/sysadmin/grid5000/cassandra/_install_os.sh
@@ -1,6 +1,6 @@
 #!/usr/bin/env bash
 
 set -eux
 
 # Install the OS
-kadeploy3 -e debian10-x64-base -f "${OAR_FILE_NODES}" -k ~/.ssh/id_rsa.pub
+kadeploy3 -e debian10-x64-big -f "${OAR_FILE_NODES}" -k ~/.ssh/id_rsa.pub
diff --git a/sysadmin/grid5000/cassandra/_provision_node.sh b/sysadmin/grid5000/cassandra/_provision_node.sh
index a1234fe..0770351 100755
--- a/sysadmin/grid5000/cassandra/_provision_node.sh
+++ b/sysadmin/grid5000/cassandra/_provision_node.sh
@@ -1,11 +1,12 @@
 #!/usr/bin/env bash
 
 set -eux
 
 apt update
 apt install -y ansible
 
 cd /root/install/ansible
 CASSANDRA_SEEDS="$(cat ../cassandra_seeds.lst)"
+NODES="$(cat ../nodes.lst | tr '\n' ',')"
 
-ansible-playbook -i hosts.yml -e @/root/install/.credentials --vault-password-file=/root/install/.vault_password -l "$(hostname)" playbook.yml --extra-vars "cassandra_seed_ips=${CASSANDRA_SEEDS}"
+ansible-playbook -i hosts.yml -e @/root/install/.credentials --vault-password-file=/root/install/.vault_password -l "$(hostname)" playbook.yml --extra-vars "cassandra_seed_ips=${CASSANDRA_SEEDS}" --extra-vars "nodes=${NODES}"
diff --git a/sysadmin/grid5000/cassandra/ansible/_install_prometheus_exporter.yml b/sysadmin/grid5000/cassandra/ansible/_install_prometheus_exporter.yml
new file mode 100644
index 0000000..6aa96e3
--- /dev/null
+++ b/sysadmin/grid5000/cassandra/ansible/_install_prometheus_exporter.yml
@@ -0,0 +1,5 @@
+---
+- name: Install node_exporter package
+  apt:
+    name:
+      - prometheus-node-exporter
diff --git a/sysadmin/grid5000/cassandra/ansible/cassandra.yml b/sysadmin/grid5000/cassandra/ansible/cassandra.yml
index b2c828f..9e5da53 100644
--- a/sysadmin/grid5000/cassandra/ansible/cassandra.yml
+++ b/sysadmin/grid5000/cassandra/ansible/cassandra.yml
@@ -1,59 +1,71 @@
 ---
 # - name: "Get public ipv4 address"
 #   set_fact:
 #     cassandra_seed_ips: "{{ansible_facts[item]['ipv4']['address']}}"
 #   with_items:
 #     - "{{cassandra_listen_interface }}"
 
 - name: Install cassandra signing key
   apt_key:
     url: https://downloads.apache.org/cassandra/KEYS
     state: present
 
 - name: Install cassandra apt repository
   apt_repository:
     repo: deb http://downloads.apache.org/cassandra/debian 40x main
     state: present
     filename: cassandra.sources
 
 - name: Install cassandra packages
   apt:
     update_cache: true # force an apt update before
     name:
       ## TODO: check other version than jdk11
-      - openjdk-11-jdk
       - cassandra
       - dstat
-      - smartmontools
       - facter
+      - openjdk-11-jdk
+      - smartmontools
       - tcpdump
 
+- name: install prometheus node exporter
+  include: _install_prometheus_exporter.yml
+
 - name: Create datadirs
   file:
     state: directory
     path: "{{ item }}"
     owner: "cassandra"
     group: "cassandra"
     mode: "0755"
+    recurse: true
   with_items:
     - "{{ cassandra_data_dir_base }}"
     - "{{ cassandra_data_dir_system }}"
     - "{{ cassandra_data_dir }}"
     - "{{ cassandra_commitlogs_dir }}"
 
+- name: Download prometheus jmx exporter
+  get_url:
+    url: https://repo1.maven.org/maven2/io/prometheus/jmx/jmx_prometheus_javaagent/0.15.0/jmx_prometheus_javaagent-0.15.0.jar
+    dest: /opt/jmx_prometheus_javaagent-0.15.0.jar
+
+- name: Download prometheus exporter configuration
+  get_url:
+    url: https://raw.githubusercontent.com/prometheus/jmx_exporter/master/example_configs/cassandra.yml
+    dest: /opt/jmx_exporter.yml
+
 - name: Configure cassandra
   template:
     src: "templates/{{item}}"
     dest: "{{cassandra_config_dir}}/{{item}}"
-  with_items: [cassandra.yaml, jvm.options]
+  with_items: [cassandra.yaml, jvm11-server.options]
   register: cassandra_configuration_files
 
 - name: Restart cassandra service
   service:
     name: cassandra
     state: restarted
   when: cassandra_configuration_files.changed
 
-
-
   # TODO test different read ahead
diff --git a/sysadmin/grid5000/cassandra/ansible/files/g5k-setup-docker b/sysadmin/grid5000/cassandra/ansible/files/g5k-setup-docker
new file mode 100644
index 0000000..d44a8e7
--- /dev/null
+++ b/sysadmin/grid5000/cassandra/ansible/files/g5k-setup-docker
@@ -0,0 +1,47 @@
+#! /usr/bin/env ruby
+
+require 'optparse'
+
+def main
+  options = {}
+  OptionParser.new do |opts|
+    opts.banner = "Usage: g5k-setup-docker [-it]\nInstall Docker locally and make Docker commands available without user needing to log out"
+    opts.on('-i', '--install', 'Just install Docker locally (user need to log out)') do |i|
+      options[:install] = i
+    end
+    opts.on('-t', '--tmp', 'Bind /tmp/docker to /var/lib/docker to provide more storage space for pulled images (WARNING : A reboot will erase it)') do |t|
+      options[:tmp] = t
+    end
+  end.parse!
+
+  ret = true
+  ret &= sh('sudo-g5k').first if Process.euid != 0
+  if options[:tmp]
+    ret &= sh('sudo mkdir /tmp/docker').first
+    ret &= sh('sudo mkdir /var/lib/docker').first
+    ret &= sh('sudo mount --bind /tmp/docker /var/lib/docker').first
+  end
+  cpu_arch = `uname -m`.strip
+  # Docker is not officially supported on ppc64
+  if cpu_arch == 'ppc64le'
+    ret &= sh('echo "deb http://packages.grid5000.fr/deb/docker-ce/buster /" | sudo tee /etc/apt/sources.list.d/docker-ce.list').first
+    ret &= sh('sudo apt-get update && sudo apt-get install -y docker-ce').first
+  else
+    ret &= sh('curl -sSL https://get.docker.com/ | sh').first
+  end
+  ret &= sh('sudo mkdir -p /etc/docker')
+  ret &= sh('echo "{ \"registry-mirrors\": [\"http://docker-cache.grid5000.fr\"] }" | sudo tee /etc/docker/daemon.json')
+  ret &= sh('sudo systemctl restart docker')
+  ret &= sh('sudo chmod o+rw /var/run/docker.sock').first if !options[:install]
+  ret
+end
+
+def sh(cmd)
+  puts(cmd)
+  output = `#{cmd}`.chomp
+  status = ($?.exitstatus == 0)
+  [status, output]
+end
+
+# MAIN PROGRAM
+exit(main)
diff --git a/sysadmin/grid5000/cassandra/ansible/hosts.yml b/sysadmin/grid5000/cassandra/ansible/hosts.yml
index 7845051..068eddb 100644
--- a/sysadmin/grid5000/cassandra/ansible/hosts.yml
+++ b/sysadmin/grid5000/cassandra/ansible/hosts.yml
@@ -1,126 +1,152 @@
 # Global configuration
 swh-storage:
   hosts:
     parasilo-[20:28].rennes.grid5000.fr:
     paranoia-[1:8].rennes.grid5000.fr:
+    # paravance-[].rennes.grid5000.fr:
     # local vagrant hosts
     swh-storage1:
   vars:
       ansible_connection: local
       journal:
         brokers:
           - broker1.journal.softwareheritage.org:9093
           - broker2.journal.softwareheritage.org:9093
           - broker3.journal.softwareheritage.org:9093
           - broker4.journal.softwareheritage.org:9093   
         consumer:
           user: swh-vse
           group: swh-vse-grid5000-1
         replayer_count:
           content: 0
           skipped_content: 0
           directory: 50
           origin: 0
           origin_visit: 0
           origin_visit_status: 0
           release: 0
           revision: 0
           snapshot: 0
 
 
 cassandra:
   hosts:
     dahu-[1:32].grenoble.grid5000.fr:
     parasilo-[1:19].rennes.grid5000.fr:
     # local vagrant hosts
     cassandra[1:9]:
   vars:
     ansible_connection: local
     cassandra_config_dir: /etc/cassandra
     cassandra_data_dir_base: /srv/cassandra
     cassandra_data_dir_system: "{{cassandra_data_dir_base}}/system"
     cassandra_data_dir: "{{ cassandra_data_dir_base }}/data"
     cassandra_commitlogs_dir: "{{ cassandra_data_dir_base }}/commitlogs"
 
 
 # Per cluster specificities
 dahu_cluster_hosts:
   hosts:
     dahu[1:32].grenoble.grid5000.fr
   vars:
     cassandra_listen_interface: enp24s0f0
 
 parasilo_cluster_hosts:
   hosts:
     parasilo-[1:28].rennes.grid5000.fr:
-    paranoia-[1:8].rennes.grid5000.fr:
   vars:
     cassandra_listen_interface: eno1
     zfs_pools:
       commitlogs:
         disks:
           - sdf
         datasets:
           commitlogs: /srv/cassandra/commitlogs
       data:
         disks:
           - sdb
           - sdc
           - sdd
           - sde
         datasets:
           data: /srv/cassandra/data
 
+paravance_cluster_hosts:
+  hosts:
+    paravance-[1:78].rennes.grid5000.fr:
+  vars:
+    cassandra_listen_interface: eno1
+    zfs_pools:
+      - data:
+          disks:
+            - sdb
+          datasets:
+            data: /srv/cassandra
+
+monitoring:
+  hosts:
+    monitoring1:
+    paravance-[1:78].rennes.grid5000.fr:
+  vars:
+    ansible_connection: local
+    install_docker_install_script: false
+
 # Vagrant configuration
 vagrant_nodes:
   hosts:
     cassandra1:
       ansible_host: 10.168.180.11
       ansible_user: vagrant
       ansible_ssh_private_key_file: .vagrant/machines/cassandra1/libvirt/private_key
     cassandra2:
       ansible_host: 10.168.180.12
       ansible_user: vagrant
       ansible_ssh_private_key_file: .vagrant/machines/cassandra2/libvirt/private_key
     cassandra3:
       ansible_host: 10.168.180.13
       ansible_user: vagrant
       ansible_ssh_private_key_file: .vagrant/machines/cassandra3/libvirt/private_key
     swh-storage1:
       ansible_host: 10.168.180.14
       ansible_user: vagrant
       ansible_ssh_private_key_file: .vagrant/machines/swh-storage/libvirt/private_key
+    monitoring1:
+      ansible_host: 10.168.180.15
+      ansible_user: vagrant
+      ansible_ssh_private_key_file: .vagrant/machines/monitoring/libvirt/private_key
   vars:
     ansible_connection: ssh
+    install_docker_install_script: true
     journal:
       brokers:
       # staging
         - broker0.journal.staging.swh.network:9093
       consumer:
         user: swh-vse
         group: swh-vse-grid5000-1
       replayer_count:
         content: 0
         skipped_content: 0
         directory: 5
         origin: 0
         origin_visit: 0
         origin_visit_status: 0
         release: 0
         revision: 0
         snapshot: 0
     cassandra_listen_interface: eth1
     # passed through --extra-vars on grid5000
     cassandra_seed_ips: 10.168.180.11,10.168.180.12,10.168.180.13
+    nodes: 10.168.180.11,10.168.180.12,10.168.180.13,10.168.180.14,10.168.180.15
     zfs_pools:
       commitlogs:
         disks:
           - vdb
         datasets:
           commitlogs: /srv/cassandra/commitlogs
       data:
         disks:
           - vdc
           - vdd
         datasets:
           data: /srv/cassandra/data
diff --git a/sysadmin/grid5000/cassandra/ansible/monitoring.yml b/sysadmin/grid5000/cassandra/ansible/monitoring.yml
new file mode 100644
index 0000000..2539b95
--- /dev/null
+++ b/sysadmin/grid5000/cassandra/ansible/monitoring.yml
@@ -0,0 +1,45 @@
+---
+- name: Create grid5000 tools directories
+  file:
+    state: directory
+    path: "{{ item }}"
+    owner: root
+    group: root
+    mode: "0755"
+  with_items:
+    - /grid5000
+    - /grid5000/code
+    - /grid5000/code/bin
+  when: install_docker_install_script
+
+- name: Install docker installation script
+  copy:
+    src: "files/g5k-setup-docker"
+    dest: "/grid5000/code/bin/g5k-setup-docker"
+    owner: root
+    group: root
+    mode: "0755"
+  when: install_docker_install_script
+
+- name: Install docker
+  command:
+    cmd: "/grid5000/code/bin/g5k-setup-docker"
+
+- name: Create prometheus data directory
+  file:
+    state: directory
+    path: /tmp/prometheus
+    owner: nobody
+    group: nogroup
+
+- name: install prometheus node exporter
+  include: _install_prometheus_exporter.yml
+
+- name: Create prometheus configuration
+  template:
+    src: "templates/prometheus/prometheus.yml"
+    dest: "/etc/prometheus.yml"
+
+- name: Start prometheus
+  command:
+    cmd: "docker run -d -p 9090:9090 -v /etc/prometheus.yml:/etc/prometheus/prometheus.yml -v /tmp/prometheus:/prometheus --name prometheus prom/prometheus"
diff --git a/sysadmin/grid5000/cassandra/ansible/playbook.yml b/sysadmin/grid5000/cassandra/ansible/playbook.yml
index b8e8a54..a295930 100644
--- a/sysadmin/grid5000/cassandra/ansible/playbook.yml
+++ b/sysadmin/grid5000/cassandra/ansible/playbook.yml
@@ -1,11 +1,16 @@
 ---
 - name: Install cassandra
   hosts: cassandra
   tasks:
     - include: zfs.yml
     - include: cassandra.yml
 
 - name: Install SWH Storage
   hosts: swh-storage
   tasks:
     - include: swh-storage.yml
+
+- name: Install Monitoring
+  hosts: monitoring
+  tasks:
+    - include: monitoring.yml
diff --git a/sysadmin/grid5000/cassandra/ansible/swh-storage.yml b/sysadmin/grid5000/cassandra/ansible/swh-storage.yml
index e270fc1..dd84201 100644
--- a/sysadmin/grid5000/cassandra/ansible/swh-storage.yml
+++ b/sysadmin/grid5000/cassandra/ansible/swh-storage.yml
@@ -1,120 +1,125 @@
 ---
 - name: Add Backports repository
   apt_repository:
     repo: deb http://deb.debian.org/debian/ buster-backports main contrib non-free
     filename: backports.sources
 
 - name: swhstorage group
   group:
     name: swhstorage
     
 - name: swhstorage user
   user:
     name: swhstorage
     group: swhstorage
+    home: /var/lib/swhstorage # *big images mount homes via nfs so the user creation failed
 
 - name: Add SWH repository
   apt_repository:
     repo: deb [trusted=yes] https://debian.softwareheritage.org/ buster-swh main
     state: present
     filename: cassandra.sources
 
 - name: Install packages
   apt:
     name:
       - daemonize
       - dstat
       - facter
+      - prometheus-statsd-exporter
       - python3
       - python3-gunicorn
       - tcpdump
 
 - name: Install packages from backports
   apt:
     name:
       - python3-typing-extensions
       - gunicorn3
     default_release: buster-backports
 
 - name: Install swh storage packages
   apt:
     name:
       - python3-swh.storage
       - python3-swh.journal
 
+- name: install prometheus node exporter
+  include: _install_prometheus_exporter.yml
+
 - name: Create directories
   file:
     state: directory
     path: "{{ item }}"
     owner: root
     group: root
     mode: "0755"
   with_items:
     - /etc/gunicorn
     - /etc/gunicorn/instances
     - /run/gunicorn
     - /run/gunicorn/swh-storage
     - /etc/softwareheritage
     - /etc/softwareheritage/storage
     - /etc/softwareheritage/replayer
 
 - name: Create swh-storage directories
   file:
     state: directory
     path: "{{ item }}"
     owner: swhstorage
     group: swhstorage
     mode: "0755"
   with_items:
     - /run/gunicorn/swh-storage/
     - /run/replayer
 
 - name: Configure gunicorn - default service
   template:
     src: "templates/gunicorn/gunicorn.service"
     dest: "/etc/systemd/system/gunicorn.service"
 
 - name: Configure gunicorn - log configuration
   template:
     src: "templates/gunicorn/logconfig.ini"
     dest: "/etc/gunicorn/logconfig.ini"
 
 - name: swh-storage gunicorn instance configuration
   template:
     src: "templates/gunicorn/gunicorn-instance.cfg"
     dest: "/etc/gunicorn/instances/swh-storage.cfg"
 
 - name: swh-storage configuration directories
   template:
     src: "templates/swhstorage/storage.yml"
     dest: "/etc/softwareheritage/storage/storage.yml"
 
 - name: swh-storage service configuration
   template:
     src: "templates/gunicorn/gunicorn-instance-service.cfg"
     dest: "/etc/systemd/system/gunicorn-swh-storage.service" # TODO variabilize
 
 - name: swh-storage service
   service:
     name: gunicorn-swh-storage
     enabled: true
     state: started
 
 - name: swh-storage init cassandra script
   template:
     src: templates/swhstorage/init-cassandra.sh
     dest: /usr/local/bin/swh-storage-init-cassandra.sh
     mode: 0755
 
 - name: Configure replayer services
   include: _configure_replayer_services.yml obj={{ item }}
   loop:
     - content
     - skipped_content
     - directory
     - origin
     - origin_visit
     - origin_visit_status
     - release
     - revision
     - snapshot
diff --git a/sysadmin/grid5000/cassandra/ansible/templates/gunicorn/gunicorn-instance-service.cfg b/sysadmin/grid5000/cassandra/ansible/templates/gunicorn/gunicorn-instance-service.cfg
index 63ae9cf..2183a53 100644
--- a/sysadmin/grid5000/cassandra/ansible/templates/gunicorn/gunicorn-instance-service.cfg
+++ b/sysadmin/grid5000/cassandra/ansible/templates/gunicorn/gunicorn-instance-service.cfg
@@ -1,25 +1,26 @@
 [Unit]
 Description=Gunicorn instance swh-storage
 ConditionPathExists=/etc/gunicorn/instances/swh-storage.cfg
 PartOf=gunicorn.service
 ReloadPropagatedFrom=gunicorn.service
 Before=gunicorn.service
 
 [Service]
 User=swhstorage
 Group=swhstorage
 PIDFile=/run/swh-storage.pid
 RuntimeDirectory=gunicorn/swh-storage
 WorkingDirectory=/run/gunicorn/swh-storage
 Environment=SWH_CONFIG_FILENAME=/etc/softwareheritage/storage/storage.yml
 Environment=SWH_LOG_TARGET=journal
 Environment=SWH_MAIN_PACKAGE=swh.storage
+Environment=STATSD_PORT=9125
 ExecStart=/usr/bin/gunicorn3 -p /run/gunicorn/swh-storage/pidfile -c /etc/gunicorn/instances/swh-storage.cfg swh.storage.api.server:make_app_from_configfile()
 ExecStop=/bin/kill -TERM $MAINPID
 ExecReload=/bin/kill -HUP $MAINPID
 
 Restart=always
 RestartSec=10
 
 [Install]
 WantedBy=multi-user.target
diff --git a/sysadmin/grid5000/cassandra/ansible/templates/jvm.options b/sysadmin/grid5000/cassandra/ansible/templates/jvm11-server.options
similarity index 98%
rename from sysadmin/grid5000/cassandra/ansible/templates/jvm.options
rename to sysadmin/grid5000/cassandra/ansible/templates/jvm11-server.options
index 7e78467..3680e3a 100644
--- a/sysadmin/grid5000/cassandra/ansible/templates/jvm.options
+++ b/sysadmin/grid5000/cassandra/ansible/templates/jvm11-server.options
@@ -1,103 +1,105 @@
 ###########################################################################
 #                         jvm11-server.options                            #
 #                                                                         #
 # See jvm-server.options. This file is specific for Java 11 and newer.    #
 ###########################################################################
 
 #################
 #  GC SETTINGS  #
 #################
 
 
 
 ### CMS Settings
 -XX:+UseConcMarkSweepGC
 -XX:+CMSParallelRemarkEnabled
 -XX:SurvivorRatio=8
 -XX:MaxTenuringThreshold=1
 -XX:CMSInitiatingOccupancyFraction=75
 -XX:+UseCMSInitiatingOccupancyOnly
 -XX:CMSWaitDuration=10000
 -XX:+CMSParallelInitialMarkEnabled
 -XX:+CMSEdenChunksRecordAlways
 ## some JVMs will fill up their heap when accessed via JMX, see CASSANDRA-6541
 -XX:+CMSClassUnloadingEnabled
 
 
 
 ### G1 Settings
 ## Use the Hotspot garbage-first collector.
 #-XX:+UseG1GC
 #-XX:+ParallelRefProcEnabled
 
 #
 ## Have the JVM do less remembered set work during STW, instead
 ## preferring concurrent GC. Reduces p99.9 latency.
 #-XX:G1RSetUpdatingPauseTimePercent=5
 #
 ## Main G1GC tunable: lowering the pause target will lower throughput and vise versa.
 ## 200ms is the JVM default and lowest viable setting
 ## 1000ms increases throughput. Keep it smaller than the timeouts in cassandra.yaml.
 #-XX:MaxGCPauseMillis=500
 
 ## Optional G1 Settings
 # Save CPU time on large (>= 16GB) heaps by delaying region scanning
 # until the heap is 70% full. The default in Hotspot 8u40 is 40%.
 #-XX:InitiatingHeapOccupancyPercent=70
 
 # For systems with > 8 cores, the default ParallelGCThreads is 5/8 the number of logical cores.
 # Otherwise equal to the number of cores when 8 or less.
 # Machines with > 10 cores should try setting these to <= full cores.
 #-XX:ParallelGCThreads=16
 # By default, ConcGCThreads is 1/4 of ParallelGCThreads.
 # Setting both to the same value can reduce STW durations.
 #-XX:ConcGCThreads=16
 
 
 ### JPMS
 
 -Djdk.attach.allowAttachSelf=true
 --add-exports java.base/jdk.internal.misc=ALL-UNNAMED
 --add-exports java.base/jdk.internal.ref=ALL-UNNAMED
 --add-exports java.base/sun.nio.ch=ALL-UNNAMED
 --add-exports java.management.rmi/com.sun.jmx.remote.internal.rmi=ALL-UNNAMED
 --add-exports java.rmi/sun.rmi.registry=ALL-UNNAMED
 --add-exports java.rmi/sun.rmi.server=ALL-UNNAMED
 --add-exports java.sql/java.sql=ALL-UNNAMED
 
 --add-opens java.base/java.lang.module=ALL-UNNAMED
 --add-opens java.base/jdk.internal.loader=ALL-UNNAMED
 --add-opens java.base/jdk.internal.ref=ALL-UNNAMED
 --add-opens java.base/jdk.internal.reflect=ALL-UNNAMED
 --add-opens java.base/jdk.internal.math=ALL-UNNAMED
 --add-opens java.base/jdk.internal.module=ALL-UNNAMED
 --add-opens java.base/jdk.internal.util.jar=ALL-UNNAMED
 --add-opens jdk.management/com.sun.management.internal=ALL-UNNAMED
 
 
 ### GC logging options -- uncomment to enable
 
 # Java 11 (and newer) GC logging options:
 # See description of https://bugs.openjdk.java.net/browse/JDK-8046148 for details about the syntax
 # The following is the equivalent to -XX:+PrintGCDetails -XX:+UseGCLogFileRotation -XX:NumberOfGCLogFiles=10 -XX:GCLogFileSize=10M
 #-Xlog:gc=info,heap*=trace,age*=debug,safepoint=info,promotion*=trace:file=/var/log/cassandra/gc.log:time,uptime,pid,tid,level:filecount=10,filesize=10485760
 
 # Notes for Java 8 migration:
 #
 # -XX:+PrintGCDetails                   maps to -Xlog:gc*:... - i.e. add a '*' after "gc"
 # -XX:+PrintGCDateStamps                maps to decorator 'time'
 #
 # -XX:+PrintHeapAtGC                    maps to 'heap' with level 'trace'
 # -XX:+PrintTenuringDistribution        maps to 'age' with level 'debug'
 # -XX:+PrintGCApplicationStoppedTime    maps to 'safepoint' with level 'info'
 # -XX:+PrintPromotionFailure            maps to 'promotion' with level 'trace'
 # -XX:PrintFLSStatistics=1              maps to 'freelist' with level 'trace'
 
 ### Netty Options
 
 # On Java >= 9 Netty requires the io.netty.tryReflectionSetAccessible system property to be set to true to enable
 # creation of direct buffers using Unsafe. Without it, this falls back to ByteBuffer.allocateDirect which has
 # inferior performance and risks exceeding MaxDirectMemory
 -Dio.netty.tryReflectionSetAccessible=true
 
+-javaagent:/opt/jmx_prometheus_javaagent-0.15.0.jar=7070:/opt/jmx_exporter.yml
+
 # The newline in the end of file is intentional
diff --git a/sysadmin/grid5000/cassandra/ansible/templates/prometheus/prometheus.yml b/sysadmin/grid5000/cassandra/ansible/templates/prometheus/prometheus.yml
new file mode 100644
index 0000000..d63724b
--- /dev/null
+++ b/sysadmin/grid5000/cassandra/ansible/templates/prometheus/prometheus.yml
@@ -0,0 +1,54 @@
+# my global config                                                                                                                                                                                                 
+global:                                                                                                                                                                                                            
+  scrape_interval:     15s # Set the scrape interval to every 15 seconds. Default is every 1 minute.                                                                                                               
+  evaluation_interval: 15s # Evaluate rules every 15 seconds. The default is every 1 minute.                                                                                                                       
+  # scrape_timeout is set to the global default (10s).                                                                                                                                                             
+                                                                                                                                                                                                                   
+# Alertmanager configuration                                                                                                                                                                                       
+alerting:                                                                                                                                                                                                          
+  alertmanagers:                                                                                                                                                                                                   
+  - static_configs:                                                                                                                                                                                                
+    - targets:                                                                                                                                                                                                     
+      # - alertmanager:9093                                                                                                                                                                                        
+                                                                                                                                                                                                                   
+# Load rules once and periodically evaluate them according to the global 'evaluation_interval'.                                                                                                                    
+rule_files:                                                                                                                                                                                                        
+  # - "first_rules.yml"                                                                                                                                                                                            
+  # - "second_rules.yml"                                                                                                                                                                                           
+                                                                                                                                                                                                                   
+# A scrape configuration containing exactly one endpoint to scrape:                                                                                                                                                
+# Here it's Prometheus itself.                                                                                                                                                                                     
+scrape_configs:                                                                                                                                                                                                    
+  # The job name is added as a label `job=<job_name>` to any timeseries scraped from this config.                                                                                                                  
+  - job_name: 'prometheus'                                                                                                                                                                                         
+                                                                                                                                                                                                                   
+    # metrics_path defaults to '/metrics'                                                                                                                                                                          
+    # scheme defaults to 'http'.                                                                                                                                                                                   
+                                                                                                                                                                                                                   
+    static_configs:                                                                                                                                                                                                
+    - targets: ['localhost:9090'] 
+
+  - job_name: 'cassandra exporter'
+
+    static_configs:
+    - targets: [
+{% for cassandra in cassandra_seed_ips.split(',') %}
+        '{{cassandra}}:7070',
+{% endfor %}
+    ]
+
+  - job_name: 'Node exporter'
+    static_configs:
+    - targets: [
+{% for node in nodes.split(',') %}
+      '{{node}}:9100',
+{% endfor %}
+      ]
+
+  - job_name: 'Statsd exporter'
+    static_configs:
+    - targets: [
+{% for node in nodes.split(',') %}
+      '{{node}}:9102',
+{% endfor %}
+      ]
diff --git a/sysadmin/grid5000/cassandra/ansible/templates/swhstorage/replayer@.service b/sysadmin/grid5000/cassandra/ansible/templates/swhstorage/replayer@.service
index 05989c9..03e77d9 100644
--- a/sysadmin/grid5000/cassandra/ansible/templates/swhstorage/replayer@.service
+++ b/sysadmin/grid5000/cassandra/ansible/templates/swhstorage/replayer@.service
@@ -1,20 +1,21 @@
 [Unit]
 Description=swh storage {{ item }} replayer
 ConditionPathExists=/etc/softwareheritage/replayer/replayer-{{ item }}.yml
 PartOf=replayer-{{item}}.target
 
 [Service]
 Type=forking
 User=swhstorage
 Group=swhstorage
 PIDFile=/run/replayer/replayer-{{item}}-%i.pid
 Environment=SWH_CONFIG_FILENAME=/etc/softwareheritage/replayer/replayer-{{ item }}.yml
 Environment=SWH_LOG_TARGET=journal
+Environment=STATSD_PORT=9125
 ExecStart=daemonize -p /run/replayer/replayer-{{item}}-%i.pid /usr/bin/swh storage replay
 ExecStop=/bin/kill -TERM $MAINPID
 ExecReload=/bin/kill -HUP $MAINPID
 # Restart=Always
 # RestartSec=10
 
 [Install]
 WantedBy=multi-user.target
diff --git a/sysadmin/grid5000/cassandra/ansible/zfs.yml b/sysadmin/grid5000/cassandra/ansible/zfs.yml
index a03a77b..735bc7f 100644
--- a/sysadmin/grid5000/cassandra/ansible/zfs.yml
+++ b/sysadmin/grid5000/cassandra/ansible/zfs.yml
@@ -1,70 +1,70 @@
 ---
 - name: Install contrib and non-free repositories
   apt_repository:
     repo: deb http://deb.debian.org/debian/ buster-backports main contrib non-free
     filename: backports.sources
 
 - name: Install zfs packages
   apt:
     update_cache: true # force an apt update before
     name:
     - linux-image-amd64
     - linux-headers-amd64
     - libnvpair1linux
     - libuutil1linux
     - libzfs2linux
     - libzpool2linux
     - zfs-dkms
     - zfs-zed
     - zfsutils-linux
   ignore_errors: True
 
 - name: Ensure zfs initialized
   shell: /usr/sbin/modprobe zfs
 
 - name: Finalize zfs packages installation
   apt:
     update_cache: true # force an apt update before
     name:
     - linux-image-amd64
     - linux-headers-amd64
     - libnvpair1linux
     - libuutil1linux
     - libzfs2linux
     - libzpool2linux
     - zfs-dkms
     - zfs-zed
     - zfsutils-linux
 
 - name: Install zfs packages after modprobe
   apt:
     update_cache: true # force an apt update before
     name:
     - zfsutils-linux
     - zfs-zed
 
 - name: test if zfs pools are configured
   command: zfs list -o name
   register: pool_list
 
 - name: zfs pool import
   shell: "zpool import {{ item.key }}"
   loop:  "{{ lookup('dict', zfs_pools) }}"
   when: item.key not in pool_list.stdout_lines
   ignore_errors: True
 
 - name: test if zfs pools are configured
   command: zfs list -o name
   register: pool_list
 
 - name: create zfs pools
   shell: "zpool create -f {{ item.key }} {{ item.value.disks | join(' ') }}"
-  loop:  "{{ lookup('dict', zfs_pools) }}"
+  loop:  "{{ lookup('dict', zfs_pools, wantlist=True) }}"
   when: item.key not in pool_list.stdout_lines
 
 - name: call dataset creation
   include: _zfs_create_dataset.yml obj={{ outside_item }}
-  loop: "{{ lookup('dict', zfs_pools) }}"
+  loop: "{{ lookup('dict', zfs_pools, wantlist=True) }}"
   loop_control:
     loop_var: outside_item
   when: outside_item.key not in pool_list.stdout_lines
diff --git a/sysadmin/grid5000/cassandra/environment.cfg b/sysadmin/grid5000/cassandra/environment.cfg
index ad60d85..60087e7 100644
--- a/sysadmin/grid5000/cassandra/environment.cfg
+++ b/sysadmin/grid5000/cassandra/environment.cfg
@@ -1,16 +1,16 @@
 export G5K_SITE='rennes.grid5000.fr'
 
 export CASSANDRA_HOSTS="parasilo-2 parasilo-3 parasilo-4 parasilo-5"
 export CASSANDRA_DISKS_COUNT=5
-export STORAGE_HOSTS="parasilo-20"
+export STORAGE_HOSTS="paravance-1"
 export JOURNAL_CLIENT_HOSTS=""
-export MONITORING_HOSTS=""
+export MONITORING_HOSTS="paravance-51"
 
 # export CASSANDRA_NB_DISK_PER_NODE=5
 export DISK_RESERVATION_DURATION=$((1 * 24)) # 1day during tests
 export NODE_RESERVATION_DURATION=00:50 # in hours
 
 
 export SSH_USER=root
 # export SSH_OPTIONS="-o UserKnownHostsFile=/dev/null -o StrictHostKeyChecking=no"
 export SSH_OPTIONS=""