diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
index 8ad0d47ec..4c45dfded 100644
--- a/.gitlab-ci.yml
+++ b/.gitlab-ci.yml
@@ -179,6 +179,7 @@ OSTree:
           # - openstack/fedora-34-x86_64
           - openstack/rhel-8-x86_64
           - openstack/rhel-8.5-x86_64
+          - openstack/rhel-9.0-beta-nightly-x86_64
 
 New OSTree:
   stage: test
@@ -191,6 +192,17 @@ New OSTree:
       - RUNNER:
           - openstack/rhel-8.5-x86_64-large
 
+New OSTree (RHEL9-BETA):
+  stage: test
+  extends: OSTree
+  script:
+    - schutzbot/deploy.sh
+    - /usr/libexec/tests/osbuild-composer/ostree-ng-og.sh
+  parallel:
+    matrix:
+      - RUNNER:
+          - openstack/rhel-9.0-beta-nightly-x86_64
+
 OSTree simplified installer:
   stage: test
   extends: OSTree
diff --git a/test/cases/ostree-ng-og.sh b/test/cases/ostree-ng-og.sh
new file mode 100644
index 000000000..91a21063d
--- /dev/null
+++ b/test/cases/ostree-ng-og.sh
@@ -0,0 +1,648 @@
+#!/bin/bash
+set -euo pipefail
+
+# Provision the software under test.
+/usr/libexec/osbuild-composer-test/provision.sh
+
+# Get OS data.
+source /etc/os-release
+ARCH=$(uname -m)
+
+# Colorful output.
+function greenprint {
+    echo -e "\033[1;32m${1}\033[0m"
+}
+
+# Start libvirtd and test it.
+greenprint "🚀 Starting libvirt daemon"
+sudo systemctl start libvirtd
+sudo virsh list --all > /dev/null
+
+# Set a customized dnsmasq configuration for libvirt so we always get the
+# same address on bootup.
+sudo tee /tmp/integration.xml > /dev/null << EOF
+<network>
+  <name>integration</name>
+  <uuid>1c8fe98c-b53a-4ca4-bbdb-deb0f26b3579</uuid>
+  <forward mode='nat'>
+    <nat>
+      <port start='1024' end='65535'/>
+    </nat>
+  </forward>
+  <bridge name='integration' stp='on' delay='0'/>
+  <mac address='52:54:00:36:46:ef'/>
+  <ip address='192.168.100.1' netmask='255.255.255.0'>
+    <dhcp>
+      <range start='192.168.100.2' end='192.168.100.254'/>
+      <host mac='34:49:22:B0:83:30' name='vm-bios' ip='192.168.100.50'/>
+      <host mac='34:49:22:B0:83:31' name='vm-uefi' ip='192.168.100.51'/>
+    </dhcp>
+  </ip>
+</network>
+EOF
+if ! sudo virsh net-info integration > /dev/null 2>&1; then
+    sudo virsh net-define /tmp/integration.xml
+fi
+if [[ $(sudo virsh net-info integration | grep 'Active' | awk '{print $2}') == 'no' ]]; then
+    sudo virsh net-start integration
+fi
+
+# Allow anyone in the wheel group to talk to libvirt.
+greenprint "🚪 Allowing users in wheel group to talk to libvirt"
+sudo tee /etc/polkit-1/rules.d/50-libvirt.rules > /dev/null << EOF
+polkit.addRule(function(action, subject) {
+    if (action.id == "org.libvirt.unix.manage" &&
+        subject.isInGroup("adm")) {
+            return polkit.Result.YES;
+    }
+});
+EOF
+
+# Set up variables.
+OSTREE_REF="test/rhel/9/${ARCH}/edge"
+OS_VARIANT="rhel9-unknown"
+TEST_UUID=$(uuidgen)
+IMAGE_KEY="osbuild-composer-ostree-test-${TEST_UUID}"
+BIOS_GUEST_ADDRESS=192.168.100.50
+UEFI_GUEST_ADDRESS=192.168.100.51
+PROD_REPO_URL=http://192.168.100.1/repo
+PROD_REPO=/var/www/html/repo
+STAGE_REPO_ADDRESS=192.168.200.1
+STAGE_REPO_URL="http://${STAGE_REPO_ADDRESS}/repo/"
+QUAY_REPO_URL="docker://quay.io/osbuild/testing-rhel-edge-push"
+QUAY_REPO_TAG=$(tr -dc a-z0-9 < /dev/urandom | head -c 4 ; echo '')
+ARTIFACTS="ci-artifacts"
+mkdir -p "${ARTIFACTS}"
+
+# Set up temporary files.
+TEMPDIR=$(mktemp -d)
+BLUEPRINT_FILE=${TEMPDIR}/blueprint.toml
+KS_FILE=${TEMPDIR}/ks.cfg
+COMPOSE_START=${TEMPDIR}/compose-start-${IMAGE_KEY}.json
+COMPOSE_INFO=${TEMPDIR}/compose-info-${IMAGE_KEY}.json
+
+# SSH setup.
+SSH_OPTIONS=(-o StrictHostKeyChecking=no -o UserKnownHostsFile=/dev/null -o ConnectTimeout=5)
+SSH_DATA_DIR=$(/usr/libexec/osbuild-composer-test/gen-ssh.sh)
+SSH_KEY=${SSH_DATA_DIR}/id_rsa
+SSH_KEY_PUB=$(cat "${SSH_KEY}".pub)
+
+if [[ "${ID}-${VERSION_ID}" != "rhel-9.0" ]]; then
+        echo "unsupported distro: ${ID}-${VERSION_ID}"
+        exit 1
+fi
+
+CONTAINER_TYPE=edge-container
+CONTAINER_FILENAME=container.tar
+INSTALLER_TYPE=edge-installer
+INSTALLER_FILENAME=installer.iso
+
+
+# Get the compose log.
+get_compose_log () {
+    COMPOSE_ID=$1
+    LOG_FILE=${ARTIFACTS}/osbuild-${ID}-${VERSION_ID}-${COMPOSE_ID}.log
+
+    # Download the logs.
+    sudo composer-cli compose log "$COMPOSE_ID" | tee "$LOG_FILE" > /dev/null
+}
+
+# Get the compose metadata.
+get_compose_metadata () {
+    COMPOSE_ID=$1
+    METADATA_FILE=${ARTIFACTS}/osbuild-${ID}-${VERSION_ID}-${COMPOSE_ID}.json
+
+    # Download the metadata.
+    sudo composer-cli compose metadata "$COMPOSE_ID" > /dev/null
+
+    # Find the tarball and extract it.
+    TARBALL=$(basename "$(find . -maxdepth 1 -type f -name "*-metadata.tar")")
+    tar -xf "$TARBALL" -C "${TEMPDIR}"
+    rm -f "$TARBALL"
+
+    # Move the JSON file into place.
+    cat "${TEMPDIR}"/"${COMPOSE_ID}".json | jq -M '.' | tee "$METADATA_FILE" > /dev/null
+}
+
+# Build ostree image.
+build_image() {
+    blueprint_name=$1
+    image_type=$2
+
+    # Get worker unit file so we can watch the journal.
+    WORKER_UNIT=$(sudo systemctl list-units | grep -o -E "osbuild.*worker.*\.service")
+    sudo journalctl -af -n 1 -u "${WORKER_UNIT}" &
+    WORKER_JOURNAL_PID=$!
+    # Stop watching the worker journal when exiting.
+    trap 'sudo pkill -P ${WORKER_JOURNAL_PID}' EXIT
+
+    # Start the compose.
+    greenprint "🚀 Starting compose"
+    if [ $# -eq 3 ]; then
+        repo_url=$3
+        sudo composer-cli --json compose start-ostree --ref "$OSTREE_REF" --url "$repo_url" "$blueprint_name" "$image_type" | tee "$COMPOSE_START"
+    else
+        sudo composer-cli --json compose start-ostree --ref "$OSTREE_REF" "$blueprint_name" "$image_type" | tee "$COMPOSE_START"
+    fi
+    COMPOSE_ID=$(jq -r '.body.build_id' "$COMPOSE_START")
+
+    # Wait for the compose to finish.
+    greenprint "⏱ Waiting for compose to finish: ${COMPOSE_ID}"
+    while true; do
+        sudo composer-cli --json compose info "${COMPOSE_ID}" | tee "$COMPOSE_INFO" > /dev/null
+        COMPOSE_STATUS=$(jq -r '.body.queue_status' "$COMPOSE_INFO")
+
+        # Is the compose finished?
+        if [[ $COMPOSE_STATUS != RUNNING ]] && [[ $COMPOSE_STATUS != WAITING ]]; then
+            break
+        fi
+
+        # Wait 30 seconds and try again.
+        sleep 5
+    done
+
+    # Capture the compose logs from osbuild.
+    greenprint "💬 Getting compose log and metadata"
+    get_compose_log "$COMPOSE_ID"
+    get_compose_metadata "$COMPOSE_ID"
+
+    # Kill the journal monitor immediately and remove the trap
+    sudo pkill -P ${WORKER_JOURNAL_PID}
+    trap - EXIT
+
+    # Did the compose finish with success?
+    if [[ $COMPOSE_STATUS != FINISHED ]]; then
+        echo "Something went wrong with the compose. 😢"
+        exit 1
+    fi
+}
+
+# Wait for the ssh server up to be.
+wait_for_ssh_up () {
+    SSH_STATUS=$(sudo ssh "${SSH_OPTIONS[@]}" -i "${SSH_KEY}" admin@"${1}" '/bin/bash -c "echo -n READY"')
+    if [[ $SSH_STATUS == READY ]]; then
+        echo 1
+    else
+        echo 0
+    fi
+}
+
+# Clean up our mess.
+clean_up () {
+    greenprint "🧼 Cleaning up"
+    # Remove tag from quay.io repo
+    skopeo delete --creds "${QUAY_USERNAME}:${QUAY_PASSWORD}" "${QUAY_REPO_URL}:${QUAY_REPO_TAG}"
+
+    # Clear vm
+    if [[ $(sudo virsh domstate "${IMAGE_KEY}-uefi") == "running" ]]; then
+        sudo virsh destroy "${IMAGE_KEY}-uefi"
+    fi
+    sudo virsh undefine "${IMAGE_KEY}-uefi" --nvram
+    # Remove qcow2 file.
+    sudo rm -f "$LIBVIRT_UEFI_IMAGE_PATH"
+
+    # Remove any status containers if exist
+    sudo podman ps -a -q --format "{{.ID}}" | sudo xargs --no-run-if-empty podman rm -f
+    # Remove all images
+    sudo podman rmi -f -a
+
+    # Remove prod repo
+    sudo rm -rf "$PROD_REPO"
+
+    # Remomve tmp dir.
+    sudo rm -rf "$TEMPDIR"
+
+    # Stop prod repo http service
+    sudo systemctl disable --now httpd
+}
+
+# Test result checking
+check_result () {
+    greenprint "🎏 Checking for test result"
+    if [[ $RESULTS == 1 ]]; then
+        greenprint "💚 Success"
+    else
+        greenprint "❌ Failed"
+        clean_up
+        exit 1
+    fi
+}
+
+###########################################################
+##
+## Prepare edge prod and stage repo
+##
+###########################################################
+greenprint "🔧 Prepare edge prod repo"
+# Start prod repo web service
+# osbuild-composer-tests have mod_ssl as a dependency. The package installs
+# an example configuration which automatically enabled httpd on port 443, but
+# that one is already in use. Remove the default configuration as it is useless
+# anyway.
+sudo rm -f /etc/httpd/conf.d/ssl.conf
+sudo systemctl enable --now httpd.service
+
+# Have a clean prod repo
+sudo rm -rf "$PROD_REPO"
+sudo mkdir -p "$PROD_REPO"
+sudo ostree --repo="$PROD_REPO" init --mode=archive
+sudo ostree --repo="$PROD_REPO" remote add --no-gpg-verify edge-stage "$STAGE_REPO_URL"
+
+# Prepare stage repo network
+greenprint "🔧 Prepare stage repo network"
+sudo podman network inspect edge >/dev/null 2>&1 || sudo podman network create --driver=bridge --subnet=192.168.200.0/24 --ip-range=192.168.200.0/24 --gateway=192.168.200.254 edge
+
+##########################################################
+##
+## edge container image for building installer image
+##
+##########################################################
+
+# Write a blueprint for ostree image.
+tee "$BLUEPRINT_FILE" > /dev/null << EOF
+name = "container"
+description = "A base rhel-edge container image"
+version = "0.0.1"
+modules = []
+groups = []
+
+[[packages]]
+name = "python3"
+version = "*"
+
+[[customizations.user]]
+name = "admin"
+description = "Administrator account"
+password = "\$6\$GRmb7S0p8vsYmXzH\$o0E020S.9JQGaHkszoog4ha4AQVs3sk8q0DvLjSMxoxHBKnB2FBXGQ/OkwZQfW/76ktHd0NX5nls2LPxPuUdl."
+key = "${SSH_KEY_PUB}"
+home = "/home/admin/"
+groups = ["wheel"]
+EOF
+
+greenprint "📄 container blueprint"
+cat "$BLUEPRINT_FILE"
+
+# Prepare the blueprint for the compose.
+greenprint "📋 Preparing container blueprint"
+sudo composer-cli blueprints push "$BLUEPRINT_FILE"
+sudo composer-cli blueprints depsolve container
+
+# Build container image.
+build_image container "${CONTAINER_TYPE}"
+
+# Download the image
+greenprint "📥 Downloading the container image"
+sudo composer-cli compose image "${COMPOSE_ID}" > /dev/null
+
+# Clear stage repo running env
+greenprint "🧹 Clearing stage repo running env"
+# Remove any status containers if exist
+sudo podman ps -a -q --format "{{.ID}}" | sudo xargs --no-run-if-empty podman rm -f
+# Remove all images
+sudo podman rmi -f -a
+
+# Deal with stage repo image
+greenprint "🗜 Pushing image to quay.io"
+IMAGE_FILENAME="${COMPOSE_ID}-${CONTAINER_FILENAME}"
+skopeo copy --dest-creds "${QUAY_USERNAME}:${QUAY_PASSWORD}" "oci-archive:${IMAGE_FILENAME}" "${QUAY_REPO_URL}:${QUAY_REPO_TAG}"
+greenprint "Downloading image from quay.io"
+sudo podman login quay.io --username "${QUAY_USERNAME}" --password "${QUAY_PASSWORD}"
+sudo podman pull "${QUAY_REPO_URL}:${QUAY_REPO_TAG}"
+sudo podman images
+greenprint "🗜 Running the image"
+sudo podman run -d --name rhel-edge --network edge --ip "$STAGE_REPO_ADDRESS" "${QUAY_REPO_URL}:${QUAY_REPO_TAG}"
+# Clear image file
+sudo rm -f "$IMAGE_FILENAME"
+
+# Wait for container to be running
+until [ "$(sudo podman inspect -f '{{.State.Running}}' rhel-edge)" == "true" ]; do
+    sleep 1;
+done;
+
+# Sync installer edge content
+greenprint "📡 Sync installer content from stage repo"
+sudo ostree --repo="$PROD_REPO" pull --mirror edge-stage "$OSTREE_REF"
+
+# Clean compose and blueprints.
+greenprint "🧽 Clean up container blueprint and compose"
+sudo composer-cli compose delete "${COMPOSE_ID}" > /dev/null
+sudo composer-cli blueprints delete container > /dev/null
+
+########################################################
+##
+## rhel-edge installer image building from container image
+##
+########################################################
+
+# Write a blueprint for installer image.
+tee "$BLUEPRINT_FILE" > /dev/null << EOF
+name = "installer"
+description = "A rhel-edge installer image"
+version = "0.0.1"
+modules = []
+groups = []
+EOF
+
+greenprint "📄 installer blueprint"
+cat "$BLUEPRINT_FILE"
+
+# Prepare the blueprint for the compose.
+greenprint "📋 Preparing installer blueprint"
+sudo composer-cli blueprints push "$BLUEPRINT_FILE"
+sudo composer-cli blueprints depsolve installer
+
+# Build installer image.
+# Test --url arg following by URL with tailling slash for bz#1942029
+build_image installer "${INSTALLER_TYPE}" "${PROD_REPO_URL}/"
+
+# Download the image
+greenprint "📥 Downloading the installer image"
+sudo composer-cli compose image "${COMPOSE_ID}" > /dev/null
+ISO_FILENAME="${COMPOSE_ID}-${INSTALLER_FILENAME}"
+sudo mv "${ISO_FILENAME}" /var/lib/libvirt/images
+
+# Clean compose and blueprints.
+greenprint "🧹 Clean up installer blueprint and compose"
+sudo composer-cli compose delete "${COMPOSE_ID}" > /dev/null
+sudo composer-cli blueprints delete installer > /dev/null
+
+########################################################
+##
+## install rhel-edge image with installer(ISO)
+##
+########################################################
+
+# Ensure SELinux is happy with our new images.
+greenprint "👿 Running restorecon on image directory"
+sudo restorecon -Rv /var/lib/libvirt/images/
+
+# Create qcow2 file for virt install.
+greenprint "🖥 Create qcow2 file for virt install"
+LIBVIRT_BIOS_IMAGE_PATH=/var/lib/libvirt/images/${IMAGE_KEY}-bios.qcow2
+LIBVIRT_UEFI_IMAGE_PATH=/var/lib/libvirt/images/${IMAGE_KEY}-uefi.qcow2
+sudo qemu-img create -f qcow2 "${LIBVIRT_BIOS_IMAGE_PATH}" 20G
+sudo qemu-img create -f qcow2 "${LIBVIRT_UEFI_IMAGE_PATH}" 20G
+
+# Write kickstart file for ostree image installation.
+greenprint "📑 Generate kickstart file"
+tee "$KS_FILE" > /dev/null << STOPHERE
+text
+network --bootproto=dhcp --device=link --activate --onboot=on
+
+zerombr
+clearpart --all --initlabel --disklabel=msdos
+autopart --nohome --noswap --type=plain
+ostreesetup --nogpg --osname=rhel-edge --remote=rhel-edge --url=file:///ostree/repo --ref=${OSTREE_REF}
+poweroff
+
+%post --log=/var/log/anaconda/post-install.log --erroronfail
+
+# no sudo password for user admin
+echo -e 'admin\tALL=(ALL)\tNOPASSWD: ALL' >> /etc/sudoers
+
+# delete local ostree repo and add external prod edge repo
+ostree remote delete rhel-edge
+ostree remote add --no-gpg-verify --no-sign-verify rhel-edge ${PROD_REPO_URL}
+%end
+STOPHERE
+
+##################################################
+##
+## Install and test Edge image on BIOS VM
+##
+##################################################
+# Install ostree image via anaconda.
+greenprint "💿 Install ostree image via installer(ISO) on BIOS VM"
+sudo virt-install  --initrd-inject="${KS_FILE}" \
+                   --extra-args="inst.ks=file:/ks.cfg console=ttyS0,115200" \
+                   --name="${IMAGE_KEY}-bios" \
+                   --disk path="${LIBVIRT_BIOS_IMAGE_PATH}",format=qcow2 \
+                   --ram 3072 \
+                   --vcpus 2 \
+                   --network network=integration,mac=34:49:22:B0:83:30 \
+                   --os-type linux \
+                   --os-variant ${OS_VARIANT} \
+                   --location "/var/lib/libvirt/images/${ISO_FILENAME}" \
+                   --nographics \
+                   --noautoconsole \
+                   --wait=-1 \
+                   --noreboot
+
+# Start VM.
+greenprint "📟 Start BIOS VM"
+sudo virsh start "${IMAGE_KEY}-bios"
+
+# Check for ssh ready to go.
+greenprint "🛃 Checking for SSH is ready to go"
+for LOOP_COUNTER in $(seq 0 30); do
+    RESULTS="$(wait_for_ssh_up $BIOS_GUEST_ADDRESS)"
+    if [[ $RESULTS == 1 ]]; then
+        echo "SSH is ready now! 🥳"
+        break
+    fi
+    sleep 10
+done
+
+# Check image installation result
+check_result
+
+# Get ostree commit value.
+greenprint "🕹 Get ostree install commit value"
+INSTALL_HASH=$(curl "${PROD_REPO_URL}/refs/heads/${OSTREE_REF}")
+
+# Run Edge test on BIOS VM
+# Add instance IP address into /etc/ansible/hosts
+sudo tee "${TEMPDIR}"/inventory > /dev/null << EOF
+[ostree_guest]
+${BIOS_GUEST_ADDRESS}
+
+[ostree_guest:vars]
+ansible_python_interpreter=/usr/bin/python3
+ansible_user=admin
+ansible_private_key_file=${SSH_KEY}
+ansible_ssh_common_args="-o StrictHostKeyChecking=no -o UserKnownHostsFile=/dev/null"
+EOF
+
+# Test IoT/Edge OS
+greenprint "📼 Run Edge tests on BIOS VM"
+sudo ANSIBLE_STDOUT_CALLBACK=debug ansible-playbook -v -i "${TEMPDIR}"/inventory -e image_type=rhel-edge -e ostree_commit="${INSTALL_HASH}" /usr/share/tests/osbuild-composer/ansible/check_ostree.yaml || RESULTS=0
+check_result
+
+# Clean up BIOS VM
+greenprint "🧹 Clean up BIOS VM"
+if [[ $(sudo virsh domstate "${IMAGE_KEY}-bios") == "running" ]]; then
+    sudo virsh destroy "${IMAGE_KEY}-bios"
+fi
+sudo virsh undefine "${IMAGE_KEY}-bios"
+sudo rm -f "$LIBVIRT_BIOS_IMAGE_PATH"
+
+##################################################
+##
+## Install, upgrade and test Edge image on UEFI VM
+##
+##################################################
+# Install ostree image via anaconda.
+greenprint "💿 Install ostree image via installer(ISO) on UEFI VM"
+sudo virt-install  --initrd-inject="${KS_FILE}" \
+                   --extra-args="inst.ks=file:/ks.cfg console=ttyS0,115200" \
+                   --name="${IMAGE_KEY}-uefi"\
+                   --disk path="${LIBVIRT_UEFI_IMAGE_PATH}",format=qcow2 \
+                   --ram 3072 \
+                   --vcpus 2 \
+                   --network network=integration,mac=34:49:22:B0:83:31 \
+                   --os-type linux \
+                   --os-variant ${OS_VARIANT} \
+                   --location "/var/lib/libvirt/images/${ISO_FILENAME}" \
+                   --boot uefi,loader_ro=yes,loader_type=pflash,nvram_template=/usr/share/edk2/ovmf/OVMF_VARS.fd,loader_secure=no \
+                   --nographics \
+                   --noautoconsole \
+                   --wait=-1 \
+                   --noreboot
+
+# Start VM.
+greenprint "💻 Start UEFI VM"
+sudo virsh start "${IMAGE_KEY}-uefi"
+
+# Check for ssh ready to go.
+greenprint "🛃 Checking for SSH is ready to go"
+for LOOP_COUNTER in $(seq 0 30); do
+    RESULTS="$(wait_for_ssh_up $UEFI_GUEST_ADDRESS)"
+    if [[ $RESULTS == 1 ]]; then
+        echo "SSH is ready now! 🥳"
+        break
+    fi
+    sleep 10
+done
+
+# Check image installation result
+check_result
+
+##################################################
+##
+## upgrade rhel-edge with new upgrade commit
+##
+##################################################
+
+# Write a blueprint for ostree image.
+tee "$BLUEPRINT_FILE" > /dev/null << EOF
+name = "upgrade"
+description = "An upgrade rhel-edge container image"
+version = "0.0.2"
+modules = []
+groups = []
+
+[[packages]]
+name = "python3"
+version = "*"
+
+[[packages]]
+name = "wget"
+version = "*"
+
+[[customizations.user]]
+name = "admin"
+description = "Administrator account"
+password = "\$6\$GRmb7S0p8vsYmXzH\$o0E020S.9JQGaHkszoog4ha4AQVs3sk8q0DvLjSMxoxHBKnB2FBXGQ/OkwZQfW/76ktHd0NX5nls2LPxPuUdl."
+key = "${SSH_KEY_PUB}"
+home = "/home/admin/"
+groups = ["wheel"]
+EOF
+
+greenprint "📄 upgrade blueprint"
+cat "$BLUEPRINT_FILE"
+
+# Prepare the blueprint for the compose.
+greenprint "📋 Preparing upgrade blueprint"
+sudo composer-cli blueprints push "$BLUEPRINT_FILE"
+sudo composer-cli blueprints depsolve upgrade
+
+# Build upgrade image.
+build_image upgrade  "${CONTAINER_TYPE}" "$PROD_REPO_URL"
+
+# Download the image
+greenprint "📥 Downloading the upgrade image"
+sudo composer-cli compose image "${COMPOSE_ID}" > /dev/null
+
+# Clear stage repo running env
+greenprint "🧹 Clearing stage repo running env"
+# Remove any status containers if exist
+sudo podman ps -a -q --format "{{.ID}}" | sudo xargs --no-run-if-empty podman rm -f
+# Remove all images
+sudo podman rmi -f -a
+
+# Deal with stage repo container
+greenprint "🗜 Extracting image"
+IMAGE_FILENAME="${COMPOSE_ID}-${CONTAINER_FILENAME}"
+sudo podman pull "oci-archive:${IMAGE_FILENAME}"
+sudo podman images
+# Clear image file
+sudo rm -f "$IMAGE_FILENAME"
+
+# Run edge stage repo
+greenprint "🛰 Running edge stage repo"
+# Get image id to run image
+EDGE_IMAGE_ID=$(sudo podman images --filter "dangling=true" --format "{{.ID}}")
+sudo podman run -d --name rhel-edge --network edge --ip "$STAGE_REPO_ADDRESS" "$EDGE_IMAGE_ID"
+# Wait for container to be running
+until [ "$(sudo podman inspect -f '{{.State.Running}}' rhel-edge)" == "true" ]; do
+    sleep 1;
+done;
+
+# Pull upgrade to prod mirror
+greenprint "⛓ Pull upgrade to prod mirror"
+sudo ostree --repo="$PROD_REPO" pull --mirror edge-stage "$OSTREE_REF"
+sudo ostree --repo="$PROD_REPO" static-delta generate "$OSTREE_REF"
+sudo ostree --repo="$PROD_REPO" summary -u
+
+# Get ostree commit value.
+greenprint "🕹 Get ostree upgrade commit value"
+UPGRADE_HASH=$(curl "${PROD_REPO_URL}/refs/heads/${OSTREE_REF}")
+
+# Clean compose and blueprints.
+greenprint "🧽 Clean up upgrade blueprint and compose"
+sudo composer-cli compose delete "${COMPOSE_ID}" > /dev/null
+sudo composer-cli blueprints delete upgrade > /dev/null
+
+# Upgrade image/commit.
+greenprint "🗳 Upgrade ostree image/commit"
+sudo ssh "${SSH_OPTIONS[@]}" -i "${SSH_KEY}" admin@${UEFI_GUEST_ADDRESS} 'sudo rpm-ostree upgrade'
+sudo ssh "${SSH_OPTIONS[@]}" -i "${SSH_KEY}" admin@${UEFI_GUEST_ADDRESS} 'nohup sudo systemctl reboot &>/dev/null & exit'
+
+# Sleep 10 seconds here to make sure vm restarted already
+sleep 10
+
+# Check for ssh ready to go.
+greenprint "🛃 Checking for SSH is ready to go"
+# shellcheck disable=SC2034  # Unused variables left for readability
+for LOOP_COUNTER in $(seq 0 30); do
+    RESULTS="$(wait_for_ssh_up $UEFI_GUEST_ADDRESS)"
+    if [[ $RESULTS == 1 ]]; then
+        echo "SSH is ready now! 🥳"
+        break
+    fi
+    sleep 10
+done
+
+# Check ostree upgrade result
+check_result
+
+# Add instance IP address into /etc/ansible/hosts
+sudo tee "${TEMPDIR}"/inventory > /dev/null << EOF
+[ostree_guest]
+${UEFI_GUEST_ADDRESS}
+
+[ostree_guest:vars]
+ansible_python_interpreter=/usr/bin/python3
+ansible_user=admin
+ansible_private_key_file=${SSH_KEY}
+ansible_ssh_common_args="-o StrictHostKeyChecking=no -o UserKnownHostsFile=/dev/null"
+EOF
+
+# Test IoT/Edge OS
+sudo ANSIBLE_STDOUT_CALLBACK=debug ansible-playbook -v -i "${TEMPDIR}"/inventory -e image_type=rhel-edge -e ostree_commit="${UPGRADE_HASH}" /usr/share/tests/osbuild-composer/ansible/check_ostree_nort.yaml || RESULTS=0
+check_result
+
+# Final success clean up
+clean_up
+
+exit 0
diff --git a/test/data/ansible/check_ostree_nort.yaml b/test/data/ansible/check_ostree_nort.yaml
new file mode 100644
index 000000000..cb0803fcb
--- /dev/null
+++ b/test/data/ansible/check_ostree_nort.yaml
@@ -0,0 +1,530 @@
+---
+- hosts: ostree_guest
+  become: no
+  vars:
+    workspace: "{{ lookup('env', 'WORKSPACE') }}"
+    total_counter: "0"
+    failed_counter: "0"
+
+  tasks:
+    # current target host's IP address
+    - debug: var=ansible_all_ipv4_addresses
+
+    # default kernel or rt kernel
+    - name: check installed kernel
+      command: uname -r
+      register: result_kernel
+
+    # first installed or upgraded
+    # first installed has one commit, but upgraded has two
+    - name: determin which stage the checking is running on
+      shell: rpm-ostree status --json | jq '.deployments | length'
+      register: result_stage
+
+    - set_fact:
+        checking_stage: "{{ result_stage.stdout }}"
+
+    # case: check ostree commit correctly updated
+    - name: get deployed ostree commit
+      command: rpm-ostree status --json
+      register: result_commit
+
+    - name: make a json result
+      set_fact:
+        deploy_commit: "{{ result_commit.stdout | from_json | json_query('deployments[0].checksum') }}"
+
+    - name: check commit deployed and built
+      block:
+        - assert:
+            that:
+              - deploy_commit == ostree_commit
+            fail_msg: "deployed ostree commit is not commit built by osbuild-composer"
+            success_msg: "successful building and deployment"
+      always:
+        - set_fact:
+            total_counter: "{{ total_counter | int + 1 }}"
+      rescue:
+        - name: failed count + 1
+          set_fact:
+            failed_counter: "{{ failed_counter | int + 1 }}"
+
+    # case from bug: https://bugzilla.redhat.com/show_bug.cgi?id=1848453
+    - name: check ostree-remount status
+      command: systemctl is-active ostree-remount.service
+      register: result_remount
+
+    - name: ostree-remount should be started
+      block:
+        - assert:
+            that:
+              - result_remount.stdout == "active"
+            fail_msg: "ostree-remount is not started by default"
+            success_msg: "starting ostree-remount successful"
+      always:
+        - set_fact:
+            total_counter: "{{ total_counter | int + 1 }}"
+      rescue:
+        - name: failed count + 1
+          set_fact:
+            failed_counter: "{{ failed_counter | int + 1 }}"
+
+    # bios or uefi
+    - name: bios or uefi
+      stat:
+        path: /sys/firmware/efi
+      register: result_uefi
+
+    - set_fact:
+        device_name: /dev/vda2
+
+    - set_fact:
+        device_name: /dev/vda3
+      when: result_uefi.stat.exists
+
+    # for edge-simplified-installer and edge-raw-image, /sysroot is mounted at /dev/vda4, need to set device_name to /dev/vda4
+    - name: check if it is simplified-installer or raw-image
+      command: df -h
+      register: result_df
+    
+    - set_fact:
+        device_name: /dev/vda4
+      when: "'/dev/vda4' in result_df.stdout"
+
+    # case: check /sysroot moutn point
+    - name: check /sysroot mount point
+      command: findmnt -r -o SOURCE -n /sysroot
+      register: result_sysroot_mount_point
+
+    - name: "/sysroot should be mounted on {{ device_name }}"
+      block:
+        - assert:
+            that:
+              - result_sysroot_mount_point.stdout == "{{ device_name }}"
+            fail_msg: "/var does not mount on {{ device_name }}"
+            success_msg: "/var mounts on {{ device_name }}"
+      always:
+        - set_fact:
+            total_counter: "{{ total_counter | int + 1 }}"
+      rescue:
+        - name: failed count + 1
+          set_fact:
+            failed_counter: "{{ failed_counter | int + 1 }}"
+
+    # case: check /sysroot mount status
+    - name: check /sysroot mount status
+      shell: findmnt -r -o OPTIONS -n /sysroot | awk -F "," '{print $1}'
+      register: result_sysroot_mount_status
+
+    - name: /sysroot should be mount with rw permission
+      block:
+        - assert:
+            that:
+              - result_sysroot_mount_status.stdout == "rw"
+            fail_msg: "/sysroot is not mounted with rw permission"
+            success_msg: "/sysroot is mounted with rw permission"
+      always:
+        - set_fact:
+            total_counter: "{{ total_counter | int + 1 }}"
+      rescue:
+        - name: failed count + 1
+          set_fact:
+            failed_counter: "{{ failed_counter | int + 1 }}"
+
+    # case: check /var mount point
+    - name: check /var mount point
+      command: findmnt -r -o SOURCE -n /var
+      register: result_var_mount_point
+
+    - name: "/var should be mounted on {{ device_name }}[/ostree/deploy/{{ image_type }}/var]"
+      block:
+        - assert:
+            that:
+              - result_var_mount_point.stdout == "{{ device_name }}[/ostree/deploy/{{ image_type }}/var]"
+            fail_msg: "/var does not mount on {{ device_name }}[/ostree/deploy/{{ image_type }}/var]"
+            success_msg: "/var mounts on {{ device_name }}[/ostree/deploy/{{ image_type }}/var]"
+      always:
+        - set_fact:
+            total_counter: "{{ total_counter | int + 1 }}"
+      rescue:
+        - name: failed count + 1
+          set_fact:
+            failed_counter: "{{ failed_counter | int + 1 }}"
+
+    # case: check /var mount status
+    - name: check /var mount status
+      shell: findmnt -r -o OPTIONS -n /var | awk -F "," '{print $1}'
+      register: result_var_mount_status
+
+    - name: /var should be mount with rw permission
+      block:
+        - assert:
+            that:
+              - result_var_mount_status.stdout == "rw"
+            fail_msg: "/var is not mounted with rw permission"
+            success_msg: "/var is mounted with rw permission"
+      always:
+        - set_fact:
+            total_counter: "{{ total_counter | int + 1 }}"
+      rescue:
+        - name: failed count + 1
+          set_fact:
+            failed_counter: "{{ failed_counter | int + 1 }}"
+
+    # case: check /usr mount point
+    - name: check /usr mount point
+      command: findmnt -r -o SOURCE -n /usr
+      register: result_usr_mount_point
+
+    - name: "/usr should be mounted on {{ device_name }}[/ostree/deploy/{{ image_type }}/deploy/{{ deploy_commit }}.0/usr]"
+      block:
+        - assert:
+            that:
+              - result_usr_mount_point.stdout == "{{ device_name }}[/ostree/deploy/{{ image_type }}/deploy/{{ deploy_commit }}.0/usr]"
+            fail_msg: "/usr does not mount on {{ device_name }}[/ostree/deploy/{{ image_type }}/deploy/{{ deploy_commit }}.0/usr]"
+            success_msg: "/usr mounts on {{ device_name }}[/ostree/deploy/{{ image_type }}/deploy/{{ deploy_commit }}.0/usr]"
+      always:
+        - set_fact:
+            total_counter: "{{ total_counter | int + 1 }}"
+      rescue:
+        - name: failed count + 1
+          set_fact:
+            failed_counter: "{{ failed_counter | int + 1 }}"
+
+    # case: check /usr mount status
+    - name: check /usr mount status
+      shell: findmnt -r -o OPTIONS -n /usr | awk -F "," '{print $1}'
+      register: result_usr_mount_status
+
+    - name: /usr should be mount with rw permission
+      block:
+        - assert:
+            that:
+              - result_usr_mount_status.stdout == "ro"
+            fail_msg: "/usr is not mounted with ro permission"
+            success_msg: "/usr is mounted with ro permission"
+      always:
+        - set_fact:
+            total_counter: "{{ total_counter | int + 1 }}"
+      rescue:
+        - name: failed count + 1
+          set_fact:
+            failed_counter: "{{ failed_counter | int + 1 }}"
+
+    - name: get the first 10 chars in commit hash
+      set_fact:
+        commit_log: "{{ deploy_commit[:11] }}"
+
+    # case: check wget installed after upgrade
+    - name: check installed package
+      shell: rpm -qa | sort
+      register: result_packages
+
+    - name: check wget installed
+      block:
+        - assert:
+            that:
+              - "'wget' in result_packages.stdout"
+            fail_msg: "wget not installed, ostree upgrade might be failed"
+            success_msg: "wget installed in ostree upgrade"
+      always:
+        - set_fact:
+            total_counter: "{{ total_counter | int + 1 }}"
+      rescue:
+        - name: failed count + 1
+          set_fact:
+            failed_counter: "{{ failed_counter | int + 1 }}"
+      when: checking_stage == "2"
+
+    - name: save installed package to log file
+      copy:
+        content: "{{ result_packages.stdout }}"
+        dest: "{{ workspace }}/{{ commit_log }}.installed.ostree.log"
+      delegate_to: localhost
+
+    # case: check ostree-remount mount log
+    - name: check ostree-remount mount log
+      command: journalctl -u ostree-remount
+      register: result_remount_jounalctl
+
+    - name: ostree-remount should remount /var and /sysroot
+      block:
+        - assert:
+            that:
+              - "'/sysroot' in result_remount_jounalctl.stdout"
+              - "'/var' in result_remount_jounalctl.stdout"
+            fail_msg: "/sysroot or /var are not remounted by ostree-remount"
+            success_msg: "/sysroot and /var are remount"
+      always:
+        - set_fact:
+            total_counter: "{{ total_counter | int + 1 }}"
+      rescue:
+        - name: failed count + 1
+          set_fact:
+            failed_counter: "{{ failed_counter | int + 1 }}"
+
+    # case: check dmesg error and failed log
+    - name: check dmesg output
+      command: dmesg
+      register: result_dmesg
+
+    - name: save dmesg output to log file
+      copy:
+        content: "{{ result_dmesg.stdout }}"
+        dest: "{{ workspace }}/{{ commit_log }}.dmesg.ostree.log"
+      delegate_to: localhost
+
+    - name: check dmesg error and fail log
+      shell: dmesg --notime | grep -i "error\|fail" || true
+      register: result_dmesg_error
+
+    # case: check running container with podman
+    - name: run ubi8 image
+      command: podman run registry.access.redhat.com/ubi8/ubi-minimal:latest cat /etc/redhat-release
+      register: podman_result
+      become: yes
+      ignore_errors: yes  # due to https://bugzilla.redhat.com/show_bug.cgi?id=1903983
+
+    - name: run container test
+      block:
+        - assert:
+            that:
+              - podman_result is succeeded
+              - "'Red Hat Enterprise Linux release' in podman_result.stdout"
+            fail_msg: "failed run container with podman"
+            success_msg: "running container with podman successed"
+      always:
+        - set_fact:
+            total_counter: "{{ total_counter | int + 1 }}"
+      rescue:
+        - name: failed count + 1
+          set_fact:
+            failed_counter: "{{ failed_counter | int + 1 }}"
+
+    # case: check dnf package and it should not be installed
+    # https://github.com/osbuild/osbuild-composer/blob/master/internal/distro/rhel8/distro.go#L642
+    - name: dnf should not be installed
+      block:
+        - name: dnf should not be installed
+          shell: rpm -qa | grep dnf || echo -n PASS
+          register: result_dnf
+
+        - assert:
+            that:
+              - result_dnf.stdout == "PASS"
+            fail_msg: "dnf is installed"
+            success_msg: "No dnf installed"
+      always:
+        - set_fact:
+            total_counter: "{{ total_counter | int + 1 }}"
+      rescue:
+        - name: failed count + 1
+          set_fact:
+            failed_counter: "{{ failed_counter | int + 1 }}"
+      when: ansible_facts['distribution'] != 'RedHat' and ansible_facts ['distribution_version'] != '8.5'
+
+    # case: check installed greenboot packages
+    # https://github.com/osbuild/osbuild-composer/blob/master/internal/distro/rhel8/distro.go#L634
+    - name: greenboot and it's related packages should be installed
+      block:
+        - name: greenboot and it's related packages should be installed
+          shell: rpm -qa | grep greenboot
+          register: result_greenboot_packages
+
+        - assert:
+            that:
+              - "'greenboot-0' in result_greenboot_packages.stdout"
+              - "'greenboot-grub2' in result_greenboot_packages.stdout"
+              - "'greenboot-rpm-ostree-grub2' in result_greenboot_packages.stdout"
+              - "'greenboot-reboot' in result_greenboot_packages.stdout"
+              - "'greenboot-status' in result_greenboot_packages.stdout"
+            fail_msg: "Some of greenboot and its related packages are not installed"
+            success_msg: "All greenboot and its related packages are installed"
+      always:
+        - set_fact:
+            total_counter: "{{ total_counter | int + 1 }}"
+      rescue:
+        - name: failed count + 1
+          set_fact:
+            failed_counter: "{{ failed_counter | int + 1 }}"
+
+    # case: check greenboot* services
+    - name: a list of greenboot* service should be enabled
+      block:
+        - name: a list of greenboot* service should be enabled
+          command: systemctl is-enabled greenboot-grub2-set-counter greenboot-grub2-set-success greenboot-healthcheck greenboot-rpm-ostree-grub2-check-fallback greenboot-status greenboot-task-runner redboot-auto-reboot redboot-task-runner
+          register: result_greenboot_service
+
+        - assert:
+            that:
+              - result_greenboot_service.stdout == 'enabled\nenabled\nenabled\nenabled\nenabled\nenabled\nenabled\nenabled'
+            fail_msg: "Some of greenboot* services are not enabled"
+            success_msg: "All greenboot* services are enabled"
+      always:
+        - set_fact:
+            total_counter: "{{ total_counter | int + 1 }}"
+      rescue:
+        - name: failed count + 1
+          set_fact:
+            failed_counter: "{{ failed_counter | int + 1 }}"
+
+    # case: check greenboot* services log
+    - name: all greenboot* service should run without error
+      block:
+        - name: all greenboot* service should run without error
+          command: journalctl -b -0 -u boot-complete.target -u greenboot -u greenboot-healthcheck -u greenboot-rpm-ostree-grub2-check-fallback -u greenboot-grub2-set-counter -u greenboot-grub2-set-success -u greenboot-status -u redboot -u redboot-auto-reboot -u redboot.target
+          register: result_greenboot_log
+
+        - assert:
+            that:
+              - "'Script \\'00_required_scripts_start.sh\\' SUCCESS' in result_greenboot_log.stdout"
+              - "'Script \\'00_wanted_scripts_start.sh\\' SUCCESS' in result_greenboot_log.stdout"
+              - "'greenboot Health Checks Runner' in result_greenboot_log.stdout"
+              - "'Reached target Boot Completion Check' in result_greenboot_log.stdout"
+              - "'Mark boot as successful in grubenv' in result_greenboot_log.stdout"
+              - "'Boot Status is GREEN - Health Check SUCCESS' in result_greenboot_log.stdout"
+              - "'greenboot MotD Generator' in result_greenboot_log.stdout"
+            fail_msg: "Some errors happened in service boot"
+            success_msg: "All greenboot services booted success"
+
+      always:
+        - set_fact:
+            total_counter: "{{ total_counter | int + 1 }}"
+      rescue:
+        - name: failed count + 1
+          set_fact:
+            failed_counter: "{{ failed_counter | int + 1 }}"
+
+    # case: check grubenv variables
+    - name: grubenv variables should contain boot_success=1
+      block:
+        - name: grubenv variables should contain boot_success=1
+          command: grub2-editenv list
+          register: result_grubenv
+          become: yes
+
+        - assert:
+            that:
+              - "'boot_success=1' in result_grubenv.stdout"
+            fail_msg: "Not found boot_success=1"
+            success_msg: "Found boot_success=1"
+      always:
+        - set_fact:
+            total_counter: "{{ total_counter | int + 1 }}"
+      rescue:
+        - name: failed count + 1
+          set_fact:
+            failed_counter: "{{ failed_counter | int + 1 }}"
+
+    # case: check rollback function if boot error found
+    - name: install sanely failing health check unit to test red boot status behavior
+      block:
+        - name: install sanely failing health check unit to test red boot status behavior
+          command: rpm-ostree install http://file-server-virt-qe-3rd.apps.ocp4.prod.psi.redhat.com/greenboot-failing-unit-1.0-1.el8.noarch.rpm
+          become: yes
+
+        - name: reboot to deploy new ostree commit
+          reboot:
+          become: yes
+
+        - name: waits until instance is reachable
+          wait_for:
+            host: "{{ ansible_all_ipv4_addresses[0] }}"
+            port: 22
+            search_regex: OpenSSH
+            delay: 10
+          register: result_rollback
+
+        - assert:
+            that:
+              - result_rollback is succeeded
+            fail_msg: "Rollback failed"
+            success_msg: "Rollback success"
+      always:
+        - set_fact:
+            total_counter: "{{ total_counter | int + 1 }}"
+      rescue:
+        - name: failed count + 1
+          set_fact:
+            failed_counter: "{{ failed_counter | int + 1 }}"
+
+    # case: check ostree commit after rollback
+    - name: check ostree commit after rollback
+      block:
+        - name: check ostree commit after rollback
+          command: rpm-ostree status --json
+          register: result_commit
+
+        - name: make a json result
+          set_fact:
+            deploy_commit: "{{ result_commit.stdout | from_json | json_query('deployments[0].checksum') }}"
+
+        - assert:
+            that:
+              - deploy_commit == ostree_commit
+            fail_msg: "Not rollback to last commit"
+            success_msg: "Rollback success"
+      always:
+        - set_fact:
+            total_counter: "{{ total_counter | int + 1 }}"
+      rescue:
+        - name: failed count + 1
+          set_fact:
+            failed_counter: "{{ failed_counter | int + 1 }}"
+      when: result_rollback is succeeded
+
+    # case: check greenboot* services log again
+    - name: fallback log should be found here
+      block:
+        - name: fallback log should be found here
+          command: journalctl -b -0 -u boot-complete.target -u greenboot -u greenboot-healthcheck -u greenboot-rpm-ostree-grub2-check-fallback -u greenboot-grub2-set-counter -u greenboot-grub2-set-success -u greenboot-status -u redboot -u redboot-auto-reboot -u redboot.target
+          register: result_greenboot_log
+
+        - assert:
+            that:
+              - "'FALLBACK BOOT DETECTED! Default rpm-ostree deployment has been rolled back' in result_greenboot_log.stdout"
+              - "'Script \\'00_required_scripts_start.sh\\' SUCCESS' in result_greenboot_log.stdout"
+              - "'Script \\'00_wanted_scripts_start.sh\\' SUCCESS' in result_greenboot_log.stdout"
+              - "'greenboot Health Checks Runner' in result_greenboot_log.stdout"
+              - "'Reached target Boot Completion Check' in result_greenboot_log.stdout"
+              - "'Mark boot as successful in grubenv' in result_greenboot_log.stdout"
+              - "'Boot Status is GREEN - Health Check SUCCESS' in result_greenboot_log.stdout"
+              - "'greenboot MotD Generator' in result_greenboot_log.stdout"
+            fail_msg: "Fallback log not found"
+            success_msg: "Found fallback log"
+
+      always:
+        - set_fact:
+            total_counter: "{{ total_counter | int + 1 }}"
+      rescue:
+        - name: failed count + 1
+          set_fact:
+            failed_counter: "{{ failed_counter | int + 1 }}"
+      when: result_rollback is succeeded
+
+    # case: check grubenv variables again
+    - name: grubenv variables should contain boot_success=1
+      block:
+        - name: grubenv variables should contain boot_success=1
+          command: grub2-editenv list
+          register: result_grubenv
+          become: yes
+
+        - assert:
+            that:
+              - "'boot_success=1' in result_grubenv.stdout"
+            fail_msg: "Not found boot_success=1"
+            success_msg: "Found boot_success=1"
+      always:
+        - set_fact:
+            total_counter: "{{ total_counter | int + 1 }}"
+      rescue:
+        - name: failed count + 1
+          set_fact:
+            failed_counter: "{{ failed_counter | int + 1 }}"
+      when: result_rollback is succeeded
+
+    - assert:
+        that:
+          - failed_counter == "0"
+        fail_msg: "Run {{ total_counter }} tests, but {{ failed_counter }} of them failed"
+        success_msg: "Totally {{ total_counter }} test passed"