X-Git-Url: https://git.arvados.org/arvados.git/blobdiff_plain/e45ba3bbf5aa851bdf8de7612eb6b587b615ba21..b3243354b3f123f78c4f1d172455c4866e5e5477:/tools/salt-install/provision.sh

diff --git a/tools/salt-install/provision.sh b/tools/salt-install/provision.sh
index 7faf5c2fb5..f90386652b 100755
--- a/tools/salt-install/provision.sh
+++ b/tools/salt-install/provision.sh
@@ -1,4 +1,4 @@
-#!/bin/bash -x
+#!/bin/bash
 
 # Copyright (C) The Arvados Authors. All rights reserved.
 #
@@ -21,7 +21,6 @@ usage() {
   echo >&2
   echo >&2 "${0} options:"
   echo >&2 "  -d, --debug                                 Run salt installation in debug mode"
-  echo >&2 "  -p <N>, --ssl-port <N>                      SSL port to use for the web applications"
   echo >&2 "  -c <local.params>, --config <local.params>  Path to the local.params config file"
   echo >&2 "  -t, --test                                  Test installation running a CWL workflow"
   echo >&2 "  -r, --roles                                 List of Arvados roles to apply to the host, comma separated"
@@ -30,8 +29,10 @@ usage() {
   echo >&2 "                                                controller"
   echo >&2 "                                                dispatcher"
   echo >&2 "                                                keepproxy"
+  echo >&2 "                                                keepbalance"
   echo >&2 "                                                keepstore"
   echo >&2 "                                                keepweb"
+  echo >&2 "                                                monitoring"
   echo >&2 "                                                shell"
   echo >&2 "                                                webshell"
   echo >&2 "                                                websocket"
@@ -39,17 +40,36 @@ usage() {
   echo >&2 "                                                workbench2"
   echo >&2 "                                              Defaults to applying them all"
   echo >&2 "  -h, --help                                  Display this help and exit"
+  echo >&2 "  --dump-config <dest_dir>                    Dumps the pillars and states to a directory"
+  echo >&2 "                                              This parameter does not perform any installation at all. It's"
+  echo >&2 "                                              intended to give you a parsed set of configuration files so"
+  echo >&2 "                                              you can inspect them or use them in you Saltstack infrastructure."
+  echo >&2 "                                              It"
+  echo >&2 "                                                - parses the pillar and states templates,"
+  echo >&2 "                                                - downloads the helper formulas with their desired versions,"
+  echo >&2 "                                                - prepares the 'top.sls' files both for pillars and states"
+  echo >&2 "                                                  for the selected role(s)"
+  echo >&2 "                                                - writes the resulting files into <dest_dir>"
   echo >&2 "  -v, --vagrant                               Run in vagrant and use the /vagrant shared dir"
+  echo >&2 "  --development                               Run in dev mode, using snakeoil certs"
   echo >&2
 }
 
 arguments() {
   # NOTE: This requires GNU getopt (part of the util-linux package on Debian-based distros).
+  if ! which getopt > /dev/null; then
+    echo >&2 "GNU getopt is required to run this script. Please install it and re-reun it"
+    exit 1
+  fi
+
   TEMP=$(getopt -o c:dhp:r:tv \
-    --long config:,debug,help,ssl-port:,roles:,test,vagrant \
+    --long config:,debug,development,dump-config:,help,roles:,test,vagrant \
     -n "${0}" -- "${@}")
 
-  if [ ${?} != 0 ] ; then echo "GNU getopt missing? Use -h for help"; exit 1 ; fi
+  if [ ${?} != 0 ];
+    then echo "Please check the parameters you entered and re-run again"
+    exit 1
+  fi
   # Note the quotes around `$TEMP': they are essential!
   eval set -- "$TEMP"
 
@@ -62,16 +82,34 @@ arguments() {
       -d | --debug)
         LOG_LEVEL="debug"
         shift
+        set -x
         ;;
-      -p | --ssl-port)
-        CONTROLLER_EXT_SSL_PORT=${2}
+      --dump-config)
+        if [[ ${2} = /* ]]; then
+          DUMP_SALT_CONFIG_DIR=${2}
+        else
+          DUMP_SALT_CONFIG_DIR=${PWD}/${2}
+        fi
+        ## states
+        S_DIR="${DUMP_SALT_CONFIG_DIR}/salt"
+        ## formulas
+        F_DIR="${DUMP_SALT_CONFIG_DIR}/formulas"
+        ## pillars
+        P_DIR="${DUMP_SALT_CONFIG_DIR}/pillars"
+        ## tests
+        T_DIR="${DUMP_SALT_CONFIG_DIR}/tests"
+        DUMP_CONFIG="yes"
         shift 2
         ;;
+      --development)
+        DEV_MODE="yes"
+        shift 1
+        ;;
       -r | --roles)
         for i in ${2//,/ }
           do
             # Verify the role exists
-            if [[ ! "database,api,controller,keepstore,websocket,keepweb,workbench2,webshell,keepproxy,shell,workbench,dispatcher" == *"$i"* ]]; then
+            if [[ ! "database,api,controller,keepstore,websocket,keepweb,workbench2,webshell,keepbalance,keepproxy,shell,workbench,dispatcher,monitoring" == *"$i"* ]]; then
               echo "The role '${i}' is not a valid role"
               usage
               exit 1
@@ -100,18 +138,41 @@ arguments() {
   done
 }
 
+copy_custom_cert() {
+  cert_dir=${1}
+  cert_name=${2}
+
+  mkdir -p --mode=0700 /srv/salt/certs
+
+  if [ -f ${cert_dir}/${cert_name}.crt ]; then
+    install --mode=0600 ${cert_dir}/${cert_name}.crt /srv/salt/certs/arvados-${cert_name}.pem
+  else
+    echo "${cert_dir}/${cert_name}.crt does not exist. Exiting"
+    exit 1
+  fi
+  if [ -f ${cert_dir}/${cert_name}.key ]; then
+    install --mode=0600 ${cert_dir}/${cert_name}.key /srv/salt/certs/arvados-${cert_name}.key
+  else
+    echo "${cert_dir}/${cert_name}.key does not exist. Exiting"
+    exit 1
+  fi
+}
+
+DEV_MODE="no"
 CONFIG_FILE="${SCRIPT_DIR}/local.params"
 CONFIG_DIR="local_config_dir"
+DUMP_CONFIG="no"
 LOG_LEVEL="info"
 CONTROLLER_EXT_SSL_PORT=443
 TESTS_DIR="tests"
 
+NGINX_INSTALL_SOURCE="install_from_repo"
+
 CLUSTER=""
 DOMAIN=""
 
 # Hostnames/IPs used for single-host deploys
-HOSTNAME_EXT=""
-HOSTNAME_INT="127.0.1.1"
+IP_INT="127.0.1.1"
 
 # Initial user setup
 INITIAL_USER=""
@@ -127,80 +188,145 @@ WEBSOCKET_EXT_SSL_PORT=8002
 WORKBENCH1_EXT_SSL_PORT=443
 WORKBENCH2_EXT_SSL_PORT=3001
 
+SSL_MODE="self-signed"
+USE_LETSENCRYPT_ROUTE53="no"
+CUSTOM_CERTS_DIR="${SCRIPT_DIR}/local_config_dir/certs"
+
+GRAFANA_DASHBOARDS_DIR="${SCRIPT_DIR}/local_config_dir/dashboards"
+
+## These are ARVADOS-related parameters
 # For a stable release, change RELEASE "production" and VERSION to the
 # package version (including the iteration, e.g. X.Y.Z-1) of the
 # release.
+# The "local.params.example.*" files already set "RELEASE=production"
+# to deploy  production-ready packages
 RELEASE="development"
 VERSION="latest"
 
-# The arvados-formula version.  For a stable release, this should be a
+# These are arvados-formula-related parameters
+# An arvados-formula tag. For a stable release, this should be a
 # branch name (e.g. X.Y-dev) or tag for the release.
-ARVADOS_TAG="main"
+# ARVADOS_TAG="2.2.0"
+# BRANCH="main"
+
+# We pin the salt version to avoid potential incompatibilities when a new
+# stable version is released.
+SALT_VERSION="3004"
 
 # Other formula versions we depend on
-POSTGRES_TAG="v0.41.6"
-NGINX_TAG="temp-fix-missing-statements-in-pillar"
-DOCKER_TAG="v1.0.0"
+#POSTGRES_TAG="v0.44.0"
+#POSTGRES_URL="https://github.com/saltstack-formulas/postgres-formula.git"
+POSTGRES_TAG="0.45.0-bugfix327"
+POSTGRES_URL="https://github.com/arvados/postgres-formula.git"
+NGINX_TAG="v2.8.1"
+DOCKER_TAG="v2.4.2"
 LOCALE_TAG="v0.3.4"
 LETSENCRYPT_TAG="v2.1.0"
+LOGROTATE_TAG="v0.14.0"
+PROMETHEUS_TAG="v5.6.5"
+GRAFANA_TAG="v3.1.3"
 
 # Salt's dir
+DUMP_SALT_CONFIG_DIR=""
 ## states
 S_DIR="/srv/salt"
 ## formulas
 F_DIR="/srv/formulas"
-##pillars
+## pillars
 P_DIR="/srv/pillars"
+## tests
+T_DIR="/tmp/cluster_tests"
 
 arguments ${@}
 
+declare -A NODES
+
 if [ -s ${CONFIG_FILE} ]; then
   source ${CONFIG_FILE}
 else
-  echo >&2 "Please create a '${CONFIG_FILE}' file with initial values, as described in"
+  echo >&2 "You don't seem to have a config file with initial values."
+  echo >&2 "Please create a '${CONFIG_FILE}' file as described in"
   echo >&2 "  * https://doc.arvados.org/install/salt-single-host.html#single_host, or"
   echo >&2 "  * https://doc.arvados.org/install/salt-multi-host.html#multi_host_multi_hostnames"
   exit 1
 fi
 
 if [ ! -d ${CONFIG_DIR} ]; then
-  echo >&2 "Please create a '${CONFIG_DIR}' with initial values, as described in"
+  echo >&2 "You don't seem to have a config directory with pillars and states."
+  echo >&2 "Please create a '${CONFIG_DIR}' directory (as configured in your '${CONFIG_FILE}'). Please see"
   echo >&2 "  * https://doc.arvados.org/install/salt-single-host.html#single_host, or"
   echo >&2 "  * https://doc.arvados.org/install/salt-multi-host.html#multi_host_multi_hostnames"
   exit 1
 fi
 
-if grep -q 'fixme_or_this_wont_work' ${CONFIG_FILE} ; then
+if grep -rni 'fixme' ${CONFIG_FILE} ${CONFIG_DIR} ; then
   echo >&2 "The config file ${CONFIG_FILE} has some parameters that need to be modified."
   echo >&2 "Please, fix them and re-run the provision script."
   exit 1
 fi
 
-if ! grep -E '^[[:alnum:]]{5}$' <<<${CLUSTER} ; then
-  echo >&2 "ERROR: <CLUSTER> must be exactly 5 alphanumeric characters long"
+if ! grep -qE '^[[:alnum:]]{5}$' <<<${CLUSTER} ; then
+  echo >&2 "ERROR: <CLUSTER> must be exactly 5 lowercase alphanumeric characters long"
   echo >&2 "Fix the cluster name in the 'local.params' file and re-run the provision script"
   exit 1
 fi
 
 # Only used in single_host/single_name deploys
-if [ "x${HOSTNAME_EXT}" = "x" ] ; then
-  HOSTNAME_EXT="${CLUSTER}.${DOMAIN}"
+if [ ! -z "${HOSTNAME_EXT}" ] ; then
+  # We need to add some extra control vars to manage a single certificate vs. multiple
+  USE_SINGLE_HOSTNAME="yes"
+  # Make sure that the value configured as IP_INT is a real IP on the system.
+  # If we don't error out early here when there is a mismatch, the formula will
+  # fail with hard to interpret nginx errors later on.
+  ip addr list |grep "${IP_INT}/" >/dev/null
+  if [[ $? -ne 0 ]]; then
+    echo "Unable to find the IP_INT address '${IP_INT}' on the system, please correct the value in local.params. Exiting..."
+    exit 1
+  fi
+else
+  USE_SINGLE_HOSTNAME="no"
+  # We set this variable, anyway, so sed lines do not fail and we don't need to add more
+  # conditionals
+  HOSTNAME_EXT="${DOMAIN}"
 fi
 
-apt-get update
-apt-get install -y curl git jq
-
-if which salt-call; then
-  echo "Salt already installed"
+if [ "${DUMP_CONFIG}" = "yes" ]; then
+  echo "The provision installer will just dump a config under ${DUMP_SALT_CONFIG_DIR} and exit"
 else
-  curl -L https://bootstrap.saltstack.com -o /tmp/bootstrap_salt.sh
-  sh /tmp/bootstrap_salt.sh -XdfP -x python3
-  /bin/systemctl stop salt-minion.service
-  /bin/systemctl disable salt-minion.service
-fi
+  # Install a few dependency packages
+  # First, let's figure out the OS we're working on
+  OS_ID=$(grep ^ID= /etc/os-release |cut -f 2 -d=  |cut -f 2 -d \")
+  echo "Detected distro: ${OS_ID}"
+
+  case ${OS_ID} in
+    "centos")
+      echo "WARNING! Disabling SELinux, see https://dev.arvados.org/issues/18019"
+      sed -i 's/SELINUX=enforcing/SELINUX=permissive/g' /etc/sysconfig/selinux
+      setenforce permissive
+      yum install -y  curl git jq
+      ;;
+    "debian"|"ubuntu")
+      # Wait 2 minutes for any apt locks to clear
+      # This option is supported from apt 1.9.1 and ignored in older apt versions.
+      # Cf. https://blog.sinjakli.co.uk/2021/10/25/waiting-for-apt-locks-without-the-hacky-bash-scripts/
+      DEBIAN_FRONTEND=noninteractive apt -o DPkg::Lock::Timeout=120 update
+      DEBIAN_FRONTEND=noninteractive apt install -y curl git jq
+      ;;
+  esac
+
+  if which salt-call; then
+    echo "Salt already installed"
+  else
+    curl -L https://bootstrap.saltstack.com -o /tmp/bootstrap_salt.sh
+    sh /tmp/bootstrap_salt.sh -XdfP -x python3 stable ${SALT_VERSION}
+    /bin/systemctl stop salt-minion.service
+    /bin/systemctl disable salt-minion.service
+  fi
+
+  # Set salt to masterless mode
+  cat > /etc/salt/minion << EOFSM
+failhard: "True"
 
-# Set salt to masterless mode
-cat > /etc/salt/minion << EOFSM
 file_client: local
 file_roots:
   base:
@@ -211,38 +337,78 @@ pillar_roots:
   base:
     - ${P_DIR}
 EOFSM
+fi
 
-mkdir -p ${S_DIR} ${F_DIR} ${P_DIR}
+mkdir -p ${S_DIR} ${F_DIR} ${P_DIR} ${T_DIR}
 
 # Get the formula and dependencies
 cd ${F_DIR} || exit 1
-git clone --branch "${ARVADOS_TAG}"     https://git.arvados.org/arvados-formula.git
-git clone --branch "${DOCKER_TAG}"      https://github.com/saltstack-formulas/docker-formula.git
-git clone --branch "${LOCALE_TAG}"      https://github.com/saltstack-formulas/locale-formula.git
-# git clone --branch "${NGINX_TAG}"       https://github.com/saltstack-formulas/nginx-formula.git
-git clone --branch "${NGINX_TAG}"       https://github.com/netmanagers/nginx-formula.git
-git clone --branch "${POSTGRES_TAG}"    https://github.com/saltstack-formulas/postgres-formula.git
-git clone --branch "${LETSENCRYPT_TAG}" https://github.com/saltstack-formulas/letsencrypt-formula.git
+echo "Cloning formulas"
+test -d docker && ( cd docker && git fetch ) \
+  || git clone --quiet https://github.com/saltstack-formulas/docker-formula.git ${F_DIR}/docker
+( cd docker && git checkout --quiet tags/"${DOCKER_TAG}" )
+
+echo "...locale"
+test -d locale && ( cd locale && git fetch ) \
+  || git clone --quiet https://github.com/saltstack-formulas/locale-formula.git ${F_DIR}/locale
+( cd locale && git checkout --quiet tags/"${LOCALE_TAG}" )
+
+echo "...nginx"
+test -d nginx && ( cd nginx && git fetch ) \
+  || git clone --quiet https://github.com/saltstack-formulas/nginx-formula.git ${F_DIR}/nginx
+( cd nginx && git checkout --quiet tags/"${NGINX_TAG}" )
+
+echo "...postgres"
+test -d postgres && ( cd postgres && git fetch ) \
+  || git clone --quiet ${POSTGRES_URL} ${F_DIR}/postgres
+( cd postgres && git checkout --quiet tags/"${POSTGRES_TAG}" )
+
+echo "...prometheus"
+test -d prometheus && ( cd prometheus && git fetch ) \
+  || git clone --quiet https://github.com/saltstack-formulas/prometheus-formula.git ${F_DIR}/prometheus
+( cd prometheus && git checkout --quiet tags/"${PROMETHEUS_TAG}" )
+
+echo "...grafana"
+test -d grafana && ( cd grafana && git fetch ) \
+  || git clone --quiet https://github.com/saltstack-formulas/grafana-formula.git ${F_DIR}/grafana
+( cd grafana && git checkout --quiet "${GRAFANA_TAG}" )
+
+echo "...letsencrypt"
+test -d letsencrypt && ( cd letsencrypt && git fetch ) \
+  || git clone --quiet https://github.com/saltstack-formulas/letsencrypt-formula.git ${F_DIR}/letsencrypt
+( cd letsencrypt && git checkout --quiet tags/"${LETSENCRYPT_TAG}" )
+
+echo "...logrotate"
+test -d logrotate && ( cd logrotate && git fetch ) \
+  || git clone --quiet https://github.com/saltstack-formulas/logrotate-formula.git ${F_DIR}/logrotate
+( cd logrotate && git checkout --quiet tags/"${LOGROTATE_TAG}" )
+
+echo "...arvados"
+test -d arvados || git clone --quiet https://git.arvados.org/arvados-formula.git ${F_DIR}/arvados
 
 # If we want to try a specific branch of the formula
 if [ "x${BRANCH}" != "x" ]; then
-  cd ${F_DIR}/arvados-formula || exit 1
-  git checkout -t origin/"${BRANCH}" -b "${BRANCH}"
-  cd -
+  ( cd ${F_DIR}/arvados && git checkout --quiet -t origin/"${BRANCH}" -b "${BRANCH}" )
+elif [ "x${ARVADOS_TAG}" != "x" ]; then
+  ( cd ${F_DIR}/arvados && git checkout --quiet tags/"${ARVADOS_TAG}" -b "${ARVADOS_TAG}" )
 fi
 
 if [ "x${VAGRANT}" = "xyes" ]; then
   EXTRA_STATES_DIR="/home/vagrant/${CONFIG_DIR}/states"
   SOURCE_PILLARS_DIR="/home/vagrant/${CONFIG_DIR}/pillars"
+  SOURCE_TOFS_DIR="/home/vagrant/${CONFIG_DIR}/tofs"
   SOURCE_TESTS_DIR="/home/vagrant/${TESTS_DIR}"
 else
   EXTRA_STATES_DIR="${SCRIPT_DIR}/${CONFIG_DIR}/states"
   SOURCE_PILLARS_DIR="${SCRIPT_DIR}/${CONFIG_DIR}/pillars"
+  SOURCE_TOFS_DIR="${SCRIPT_DIR}/${CONFIG_DIR}/tofs"
   SOURCE_TESTS_DIR="${SCRIPT_DIR}/${TESTS_DIR}"
 fi
 
 SOURCE_STATES_DIR="${EXTRA_STATES_DIR}"
 
+echo "Writing pillars and states"
+
 # Replace variables (cluster,  domain, etc) in the pillars, states and tests
 # to ease deployment for newcomers
 if [ ! -d "${SOURCE_PILLARS_DIR}" ]; then
@@ -256,7 +422,7 @@ for f in $(ls "${SOURCE_PILLARS_DIR}"/*); do
        s#__CLUSTER__#${CLUSTER}#g;
        s#__DOMAIN__#${DOMAIN}#g;
        s#__HOSTNAME_EXT__#${HOSTNAME_EXT}#g;
-       s#__HOSTNAME_INT__#${HOSTNAME_INT}#g;
+       s#__IP_INT__#${IP_INT}#g;
        s#__INITIAL_USER_EMAIL__#${INITIAL_USER_EMAIL}#g;
        s#__INITIAL_USER_PASSWORD__#${INITIAL_USER_PASSWORD}#g;
        s#__INITIAL_USER__#${INITIAL_USER}#g;
@@ -279,40 +445,57 @@ for f in $(ls "${SOURCE_PILLARS_DIR}"/*); do
        s#__WEBSOCKET_INT_IP__#${WEBSOCKET_INT_IP}#g;
        s#__KEEP_INT_IP__#${KEEP_INT_IP}#g;
        s#__KEEPSTORE0_INT_IP__#${KEEPSTORE0_INT_IP}#g;
-       s#__KEEPSTORE1_INT_IP__#${KEEPSTORE1_INT_IP}#g;
        s#__KEEPWEB_INT_IP__#${KEEPWEB_INT_IP}#g;
        s#__WEBSHELL_INT_IP__#${WEBSHELL_INT_IP}#g;
        s#__SHELL_INT_IP__#${SHELL_INT_IP}#g;
        s#__WORKBENCH1_INT_IP__#${WORKBENCH1_INT_IP}#g;
        s#__WORKBENCH2_INT_IP__#${WORKBENCH2_INT_IP}#g;
        s#__DATABASE_INT_IP__#${DATABASE_INT_IP}#g;
-       s#__WORKBENCH_SECRET_KEY__#${WORKBENCH_SECRET_KEY}#g" \
+       s#__WORKBENCH_SECRET_KEY__#${WORKBENCH_SECRET_KEY}#g;
+       s#__SSL_KEY_ENCRYPTED__#${SSL_KEY_ENCRYPTED}#g;
+       s#__SSL_KEY_AWS_REGION__#${SSL_KEY_AWS_REGION}#g;
+       s#__SSL_KEY_AWS_SECRET_NAME__#${SSL_KEY_AWS_SECRET_NAME}#g;
+       s#__CONTROLLER_NGINX_WORKERS__#${CONTROLLER_NGINX_WORKERS}#g;
+       s#__CONTROLLER_MAX_CONCURRENT_REQUESTS__#${CONTROLLER_MAX_CONCURRENT_REQUESTS}#g;
+       s#__MONITORING_USERNAME__#${MONITORING_USERNAME}#g;
+       s#__MONITORING_EMAIL__#${MONITORING_EMAIL}#g;
+       s#__MONITORING_PASSWORD__#${MONITORING_PASSWORD}#g" \
   "${f}" > "${P_DIR}"/$(basename "${f}")
 done
 
-if [ "x${TEST}" = "xyes" ] && [ ! -d "${SOURCE_TESTS_DIR}" ]; then
-  echo "You requested to run tests, but ${SOURCE_TESTS_DIR} does not exist or is not a directory. Exiting."
-  exit 1
+if [ ! -d "${SOURCE_TESTS_DIR}" ]; then
+  echo "WARNING: The tests directory was not copied to \"${SOURCE_TESTS_DIR}\"."
+  if [ "x${TEST}" = "xyes" ]; then
+    echo "WARNING: Disabling tests for this installation."
+  fi
+  TEST="no"
+else
+  mkdir -p ${T_DIR}
+  # Replace cluster and domain name in the test files
+  for f in $(ls "${SOURCE_TESTS_DIR}"/*); do
+    FILTERS="s#__CLUSTER__#${CLUSTER}#g;
+         s#__CONTROLLER_EXT_SSL_PORT__#${CONTROLLER_EXT_SSL_PORT}#g;
+         s#__DOMAIN__#${DOMAIN}#g;
+         s#__IP_INT__#${IP_INT}#g;
+         s#__INITIAL_USER_EMAIL__#${INITIAL_USER_EMAIL}#g;
+         s#__INITIAL_USER_PASSWORD__#${INITIAL_USER_PASSWORD}#g
+         s#__INITIAL_USER__#${INITIAL_USER}#g;
+         s#__DATABASE_PASSWORD__#${DATABASE_PASSWORD}#g;
+         s#__SYSTEM_ROOT_TOKEN__#${SYSTEM_ROOT_TOKEN}#g"
+    if [ "$USE_SINGLE_HOSTNAME" = "yes" ]; then
+      FILTERS="s#__CLUSTER__.__DOMAIN__#${HOSTNAME_EXT}#g;
+         $FILTERS"
+    fi
+    sed "$FILTERS" \
+      "${f}" > ${T_DIR}/$(basename "${f}")
+  done
+  chmod 755 ${T_DIR}/run-test.sh
 fi
-mkdir -p /tmp/cluster_tests
-# Replace cluster and domain name in the test files
-for f in $(ls "${SOURCE_TESTS_DIR}"/*); do
-  sed "s#__CLUSTER__#${CLUSTER}#g;
-       s#__CONTROLLER_EXT_SSL_PORT__#${CONTROLLER_EXT_SSL_PORT}#g;
-       s#__DOMAIN__#${DOMAIN}#g;
-       s#__HOSTNAME_INT__#${HOSTNAME_INT}#g;
-       s#__INITIAL_USER_EMAIL__#${INITIAL_USER_EMAIL}#g;
-       s#__INITIAL_USER_PASSWORD__#${INITIAL_USER_PASSWORD}#g
-       s#__INITIAL_USER__#${INITIAL_USER}#g;
-       s#__DATABASE_PASSWORD__#${DATABASE_PASSWORD}#g;
-       s#__SYSTEM_ROOT_TOKEN__#${SYSTEM_ROOT_TOKEN}#g" \
-  "${f}" > "/tmp/cluster_tests"/$(basename "${f}")
-done
-chmod 755 /tmp/cluster_tests/run-test.sh
 
 # Replace helper state files that differ from the formula's examples
 if [ -d "${SOURCE_STATES_DIR}" ]; then
   mkdir -p "${F_DIR}"/extra/extra
+  rm -f "${F_DIR}"/extra/extra/*
 
   for f in $(ls "${SOURCE_STATES_DIR}"/*); do
     sed "s#__ANONYMOUS_USER_TOKEN__#${ANONYMOUS_USER_TOKEN}#g;
@@ -321,7 +504,7 @@ if [ -d "${SOURCE_STATES_DIR}" ]; then
          s#__CONTROLLER_EXT_SSL_PORT__#${CONTROLLER_EXT_SSL_PORT}#g;
          s#__DOMAIN__#${DOMAIN}#g;
          s#__HOSTNAME_EXT__#${HOSTNAME_EXT}#g;
-         s#__HOSTNAME_INT__#${HOSTNAME_INT}#g;
+         s#__IP_INT__#${IP_INT}#g;
          s#__INITIAL_USER_EMAIL__#${INITIAL_USER_EMAIL}#g;
          s#__INITIAL_USER_PASSWORD__#${INITIAL_USER_PASSWORD}#g;
          s#__INITIAL_USER__#${INITIAL_USER}#g;
@@ -337,17 +520,25 @@ if [ -d "${SOURCE_STATES_DIR}" ]; then
          s#__WEBSOCKET_INT_IP__#${WEBSOCKET_INT_IP}#g;
          s#__KEEP_INT_IP__#${KEEP_INT_IP}#g;
          s#__KEEPSTORE0_INT_IP__#${KEEPSTORE0_INT_IP}#g;
-         s#__KEEPSTORE1_INT_IP__#${KEEPSTORE1_INT_IP}#g;
          s#__KEEPWEB_INT_IP__#${KEEPWEB_INT_IP}#g;
          s#__WEBSHELL_INT_IP__#${WEBSHELL_INT_IP}#g;
          s#__WORKBENCH1_INT_IP__#${WORKBENCH1_INT_IP}#g;
          s#__WORKBENCH2_INT_IP__#${WORKBENCH2_INT_IP}#g;
          s#__DATABASE_INT_IP__#${DATABASE_INT_IP}#g;
          s#__WEBSHELL_EXT_SSL_PORT__#${WEBSHELL_EXT_SSL_PORT}#g;
+         s#__SHELL_INT_IP__#${SHELL_INT_IP}#g;
          s#__WEBSOCKET_EXT_SSL_PORT__#${WEBSOCKET_EXT_SSL_PORT}#g;
          s#__WORKBENCH1_EXT_SSL_PORT__#${WORKBENCH1_EXT_SSL_PORT}#g;
          s#__WORKBENCH2_EXT_SSL_PORT__#${WORKBENCH2_EXT_SSL_PORT}#g;
-         s#__WORKBENCH_SECRET_KEY__#${WORKBENCH_SECRET_KEY}#g" \
+         s#__WORKBENCH_SECRET_KEY__#${WORKBENCH_SECRET_KEY}#g;
+         s#__SSL_KEY_ENCRYPTED__#${SSL_KEY_ENCRYPTED}#g;
+         s#__SSL_KEY_AWS_REGION__#${SSL_KEY_AWS_REGION}#g;
+         s#__SSL_KEY_AWS_SECRET_NAME__#${SSL_KEY_AWS_SECRET_NAME}#g;
+         s#__CONTROLLER_NGINX_WORKERS__#${CONTROLLER_NGINX_WORKERS}#g;
+         s#__CONTROLLER_MAX_CONCURRENT_REQUESTS__#${CONTROLLER_MAX_CONCURRENT_REQUESTS}#g;
+         s#__MONITORING_USERNAME__#${MONITORING_USERNAME}#g;
+         s#__MONITORING_EMAIL__#${MONITORING_EMAIL}#g;
+         s#__MONITORING_PASSWORD__#${MONITORING_PASSWORD}#g" \
     "${f}" > "${F_DIR}/extra/extra"/$(basename "${f}")
   done
 fi
@@ -356,6 +547,12 @@ fi
 # As we need to separate both states and pillars in case we want specific
 # roles, we iterate on both at the same time
 
+# Formula template overrides (TOFS)
+# See: https://template-formula.readthedocs.io/en/latest/TOFS_pattern.html#template-override
+if [ -d ${SOURCE_TOFS_DIR} ]; then
+  find ${SOURCE_TOFS_DIR} -mindepth 1 -maxdepth 1 -type d -exec cp -r "{}" ${S_DIR} \;
+fi
+
 # States
 cat > ${S_DIR}/top.sls << EOFTSLS
 base:
@@ -373,9 +570,21 @@ EOFPSLS
 
 # States, extra states
 if [ -d "${F_DIR}"/extra/extra ]; then
-  for f in $(ls "${F_DIR}"/extra/extra/*.sls); do
+  SKIP_SNAKE_OIL="snakeoil_certs"
+
+  if [[ "$DEV_MODE" = "yes" || "${SSL_MODE}" == "self-signed" ]] ; then
+    # In dev mode, we create some snake oil certs that we'll
+    # use as CUSTOM_CERTS, so we don't skip the states file.
+    # Same when using self-signed certificates.
+    SKIP_SNAKE_OIL="dont_add_snakeoil_certs"
+  fi
+  for f in $(ls "${F_DIR}"/extra/extra/*.sls | egrep -v "${SKIP_SNAKE_OIL}|shell_"); do
   echo "    - extra.$(basename ${f} | sed 's/.sls$//g')" >> ${S_DIR}/top.sls
   done
+  # Use byo or self-signed certificates
+  if [ "${SSL_MODE}" != "lets-encrypt" ]; then
+    mkdir -p "${F_DIR}"/extra/extra/files
+  fi
 fi
 
 # If we want specific roles for a node, just add the desired states
@@ -383,20 +592,38 @@ fi
 if [ -z "${ROLES}" ]; then
   # States
   echo "    - nginx.passenger" >> ${S_DIR}/top.sls
-  # Currently, only available on config_examples/multi_host/aws
-  if [ "x${USE_LETSENCRYPT}" = "xyes" ]; then
-    if [ "x${USE_LETSENCRYPT_IAM_USER}" = "xyes" ]; then
-      grep -q "aws_credentials" ${S_DIR}/top.sls || echo "    - aws_credentials" >> ${S_DIR}/top.sls
+  if [ "${SSL_MODE}" = "lets-encrypt" ]; then
+    if [ "${USE_LETSENCRYPT_ROUTE53}" = "yes" ]; then
+      grep -q "aws_credentials" ${S_DIR}/top.sls || echo "    - extra.aws_credentials" >> ${S_DIR}/top.sls
+    fi
+    grep -q "letsencrypt" ${S_DIR}/top.sls || echo "    - letsencrypt" >> ${S_DIR}/top.sls
+  else
+    mkdir -p --mode=0700 /srv/salt/certs
+    if [ "${SSL_MODE}" = "bring-your-own" ]; then
+      # Copy certs to formula extra/files
+      install --mode=0600 ${CUSTOM_CERTS_DIR}/* /srv/salt/certs/
+      # We add the custom_certs state
+      grep -q "custom_certs" ${S_DIR}/top.sls || echo "    - extra.custom_certs" >> ${S_DIR}/top.sls
+      if [ "${SSL_KEY_ENCRYPTED}" = "yes" ]; then
+        grep -q "ssl_key_encrypted" ${S_DIR}/top.sls || echo "    - extra.ssl_key_encrypted" >> ${S_DIR}/top.sls
+      fi
     fi
-    grep -q "letsencrypt"     ${S_DIR}/top.sls || echo "    - letsencrypt" >> ${S_DIR}/top.sls
+    # In self-signed mode, the certificate files will be created and put in the
+    # destination directory by the snakeoil_certs.sls state file
   fi
+
   echo "    - postgres" >> ${S_DIR}/top.sls
+  echo "    - logrotate" >> ${S_DIR}/top.sls
   echo "    - docker.software" >> ${S_DIR}/top.sls
   echo "    - arvados" >> ${S_DIR}/top.sls
+  echo "    - extra.shell_sudo_passwordless" >> ${S_DIR}/top.sls
+  echo "    - extra.shell_cron_add_login_sync" >> ${S_DIR}/top.sls
+  echo "    - extra.passenger_rvm" >> ${S_DIR}/top.sls
 
   # Pillars
   echo "    - docker" >> ${P_DIR}/top.sls
   echo "    - nginx_api_configuration" >> ${P_DIR}/top.sls
+  echo "    - logrotate_api" >> ${P_DIR}/top.sls
   echo "    - nginx_controller_configuration" >> ${P_DIR}/top.sls
   echo "    - nginx_keepproxy_configuration" >> ${P_DIR}/top.sls
   echo "    - nginx_keepweb_configuration" >> ${P_DIR}/top.sls
@@ -405,88 +632,301 @@ if [ -z "${ROLES}" ]; then
   echo "    - nginx_webshell_configuration" >> ${P_DIR}/top.sls
   echo "    - nginx_workbench2_configuration" >> ${P_DIR}/top.sls
   echo "    - nginx_workbench_configuration" >> ${P_DIR}/top.sls
+  echo "    - logrotate_wb1" >> ${P_DIR}/top.sls
   echo "    - postgresql" >> ${P_DIR}/top.sls
-  # Currently, only available on config_examples/multi_host/aws
-  if [ "x${USE_LETSENCRYPT}" = "xyes" ]; then
-    if [ "x${USE_LETSENCRYPT_IAM_USER}" = "xyes" ]; then
+
+  # We need to tweak the Nginx's pillar depending whether we want plan nginx or nginx+passenger
+  NGINX_INSTALL_SOURCE="install_from_phusionpassenger"
+  sed -i "s/__NGINX_INSTALL_SOURCE__/${NGINX_INSTALL_SOURCE}/g" ${P_DIR}/nginx_passenger.sls
+
+  if [ "${SSL_MODE}" = "lets-encrypt" ]; then
+    if [ "${USE_LETSENCRYPT_ROUTE53}" = "yes" ]; then
       grep -q "aws_credentials" ${P_DIR}/top.sls || echo "    - aws_credentials" >> ${P_DIR}/top.sls
     fi
-    grep -q "letsencrypt"     ${P_DIR}/top.sls || echo "    - letsencrypt" >> ${P_DIR}/top.sls
+    grep -q "letsencrypt" ${P_DIR}/top.sls || echo "    - letsencrypt" >> ${P_DIR}/top.sls
+
+    hosts=("controller" "websocket" "workbench" "workbench2" "webshell" "keepproxy")
+    if [ ${USE_SINGLE_HOSTNAME} = "no" ]; then
+      hosts+=("download" "collections")
+    else
+      hosts+=("keepweb")
+    fi
+
+    for c in "${hosts[@]}"; do
+      # Are we in a single-host-single-hostname env?
+      if [ "${USE_SINGLE_HOSTNAME}" = "yes" ]; then
+        # Are we in a single-host-single-hostname env?
+        CERT_NAME=${HOSTNAME_EXT}
+      else
+        # We are in a multiple-hostnames env
+        CERT_NAME=${c}.${DOMAIN}
+      fi
+
+      # As the pillar differs whether we use LE or custom certs, we need to do a final edition on them
+      sed -i "s/__CERT_REQUIRES__/cmd: create-initial-cert-${CERT_NAME}*/g;
+              s#__CERT_PEM__#/etc/letsencrypt/live/${CERT_NAME}/fullchain.pem#g;
+              s#__CERT_KEY__#/etc/letsencrypt/live/${CERT_NAME}/privkey.pem#g" \
+      ${P_DIR}/nginx_${c}_configuration.sls
+    done
+  else
+    # Use custom certs (either dev mode or prod)
+    grep -q "extra_custom_certs" ${P_DIR}/top.sls || echo "    - extra_custom_certs" >> ${P_DIR}/top.sls
+    # And add the certs in the custom_certs pillar
+    echo "extra_custom_certs_dir: /srv/salt/certs" > ${P_DIR}/extra_custom_certs.sls
+    echo "extra_custom_certs:" >> ${P_DIR}/extra_custom_certs.sls
+
+    for c in controller websocket workbench workbench2 webshell keepweb keepproxy; do
+      # Are we in a single-host-single-hostname env?
+      if [ "${USE_SINGLE_HOSTNAME}" = "yes" ]; then
+        # Are we in a single-host-single-hostname env?
+        CERT_NAME=${HOSTNAME_EXT}
+      else
+        # We are in a multiple-hostnames env
+        CERT_NAME=${c}
+      fi
+
+      if [[ "$SSL_MODE" == "bring-your-own" ]]; then
+        copy_custom_cert ${CUSTOM_CERTS_DIR} ${CERT_NAME}
+      fi
+
+      grep -q ${CERT_NAME} ${P_DIR}/extra_custom_certs.sls || echo "  - ${CERT_NAME}" >> ${P_DIR}/extra_custom_certs.sls
+
+      # As the pillar differs whether we use LE or custom certs, we need to do a final edition on them
+      sed -i "s/__CERT_REQUIRES__/file: extra_custom_certs_file_copy_arvados-${CERT_NAME}.pem/g;
+              s#__CERT_PEM__#/etc/nginx/ssl/arvados-${CERT_NAME}.pem#g;
+              s#__CERT_KEY__#/etc/nginx/ssl/arvados-${CERT_NAME}.key#g" \
+      ${P_DIR}/nginx_${c}_configuration.sls
+    done
   fi
 else
   # If we add individual roles, make sure we add the repo first
   echo "    - arvados.repo" >> ${S_DIR}/top.sls
+  # We add the extra_custom_certs state
+  grep -q "extra.custom_certs"    ${S_DIR}/top.sls || echo "    - extra.custom_certs" >> ${S_DIR}/top.sls
+  if [ "${SSL_KEY_ENCRYPTED}" = "yes" ]; then
+    grep -q "ssl_key_encrypted" ${S_DIR}/top.sls || echo "    - extra.ssl_key_encrypted" >> ${S_DIR}/top.sls
+  fi
+
+  # And we add the basic part for the certs pillar
+  if [ "${SSL_MODE}" != "lets-encrypt" ]; then
+    # And add the certs in the custom_certs pillar
+    echo "extra_custom_certs_dir: /srv/salt/certs" > ${P_DIR}/extra_custom_certs.sls
+    echo "extra_custom_certs:" >> ${P_DIR}/extra_custom_certs.sls
+    grep -q "extra_custom_certs" ${P_DIR}/top.sls || echo "    - extra_custom_certs" >> ${P_DIR}/top.sls
+  fi
+
+  # Prometheus state on all nodes due to the node exporter below
+  grep -q "\- prometheus$" ${S_DIR}/top.sls || echo "    - prometheus" >> ${S_DIR}/top.sls
+  # Prometheus node exporter pillar
+  grep -q "prometheus_node_exporter" ${P_DIR}/top.sls || echo "    - prometheus_node_exporter" >> ${P_DIR}/top.sls
+
   for R in ${ROLES}; do
     case "${R}" in
       "database")
         # States
-        echo "    - postgres" >> ${S_DIR}/top.sls
+        grep -q "\- postgres$" ${S_DIR}/top.sls || echo "    - postgres" >> ${S_DIR}/top.sls
+        grep -q "extra.prometheus_pg_exporter" ${S_DIR}/top.sls || echo "    - extra.prometheus_pg_exporter" >> ${S_DIR}/top.sls
         # Pillars
-        echo '    - postgresql' >> ${P_DIR}/top.sls
+        grep -q "postgresql" ${P_DIR}/top.sls || echo "    - postgresql" >> ${P_DIR}/top.sls
+        grep -q "prometheus_pg_exporter" ${P_DIR}/top.sls || echo "    - prometheus_pg_exporter" >> ${P_DIR}/top.sls
+      ;;
+      "monitoring")
+        ### Support files ###
+        GRAFANA_DASHBOARDS_DEST_DIR=/srv/salt/dashboards
+        mkdir -p "${GRAFANA_DASHBOARDS_DEST_DIR}"
+        rm -f "${GRAFANA_DASHBOARDS_DEST_DIR}"/*
+        # "ArvadosPromDataSource" is the hardcoded UID for Prometheus' datasource
+        # in Grafana.
+        for f in $(ls "${GRAFANA_DASHBOARDS_DIR}"/*.json); do
+          sed 's#${DS_PROMETHEUS}#ArvadosPromDataSource#g' \
+          "${f}" > "${GRAFANA_DASHBOARDS_DEST_DIR}"/$(basename "${f}")
+        done
+
+        ### States ###
+        grep -q "\- nginx$" ${S_DIR}/top.sls || echo "    - nginx" >> ${S_DIR}/top.sls
+        grep -q "extra.nginx_prometheus_configuration" ${S_DIR}/top.sls || echo "    - extra.nginx_prometheus_configuration" >> ${S_DIR}/top.sls
+
+        grep -q "\- grafana$" ${S_DIR}/top.sls || echo "    - grafana" >> ${S_DIR}/top.sls
+        grep -q "extra.grafana_datasource" ${S_DIR}/top.sls || echo "    - extra.grafana_datasource" >> ${S_DIR}/top.sls
+        grep -q "extra.grafana_dashboards" ${S_DIR}/top.sls || echo "    - extra.grafana_dashboards" >> ${S_DIR}/top.sls
+        grep -q "extra.grafana_admin_user" ${S_DIR}/top.sls || echo "    - extra.grafana_admin_user" >> ${S_DIR}/top.sls
+
+        if [ "${SSL_MODE}" = "lets-encrypt" ]; then
+          grep -q "letsencrypt"     ${S_DIR}/top.sls || echo "    - letsencrypt" >> ${S_DIR}/top.sls
+          if [ "x${USE_LETSENCRYPT_ROUTE53}" = "xyes" ]; then
+            grep -q "aws_credentials" ${S_DIR}/top.sls || echo "    - aws_credentials" >> ${S_DIR}/top.sls
+          fi
+        elif [ "${SSL_MODE}" = "bring-your-own" ]; then
+          for SVC in grafana prometheus; do
+            copy_custom_cert ${CUSTOM_CERTS_DIR} ${SVC}
+          done
+        fi
+        ### Pillars ###
+        grep -q "prometheus_server" ${P_DIR}/top.sls || echo "    - prometheus_server" >> ${P_DIR}/top.sls
+        grep -q "grafana" ${P_DIR}/top.sls || echo "    - grafana" >> ${P_DIR}/top.sls
+        for SVC in grafana prometheus; do
+          grep -q "nginx_${SVC}_configuration" ${P_DIR}/top.sls || echo "    - nginx_${SVC}_configuration" >> ${P_DIR}/top.sls
+        done
+        if [ "${SSL_MODE}" = "lets-encrypt" ]; then
+          grep -q "letsencrypt"     ${P_DIR}/top.sls || echo "    - letsencrypt" >> ${P_DIR}/top.sls
+          for SVC in grafana prometheus; do
+            grep -q "letsencrypt_${SVC}_configuration" ${P_DIR}/top.sls || echo "    - letsencrypt_${SVC}_configuration" >> ${P_DIR}/top.sls
+            sed -i "s/__CERT_REQUIRES__/cmd: create-initial-cert-${SVC}.${DOMAIN}*/g;
+                    s#__CERT_PEM__#/etc/letsencrypt/live/${SVC}.${DOMAIN}/fullchain.pem#g;
+                    s#__CERT_KEY__#/etc/letsencrypt/live/${SVC}.${DOMAIN}/privkey.pem#g" \
+            ${P_DIR}/nginx_${SVC}_configuration.sls
+          done
+          if [ "${USE_LETSENCRYPT_ROUTE53}" = "yes" ]; then
+            grep -q "aws_credentials" ${P_DIR}/top.sls || echo "    - aws_credentials" >> ${P_DIR}/top.sls
+          fi
+        elif [ "${SSL_MODE}" = "bring-your-own" ]; then
+          grep -q "ssl_key_encrypted" ${P_DIR}/top.sls || echo "    - ssl_key_encrypted" >> ${P_DIR}/top.sls
+          for SVC in grafana prometheus; do
+            sed -i "s/__CERT_REQUIRES__/file: extra_custom_certs_file_copy_arvados-${SVC}.pem/g;
+                    s#__CERT_PEM__#/etc/nginx/ssl/arvados-${SVC}.pem#g;
+                    s#__CERT_KEY__#/etc/nginx/ssl/arvados-${SVC}.key#g" \
+              ${P_DIR}/nginx_${SVC}_configuration.sls
+            grep -q ${SVC} ${P_DIR}/extra_custom_certs.sls || echo "  - ${SVC}" >> ${P_DIR}/extra_custom_certs.sls
+          done
+        fi
       ;;
       "api")
         # States
-        # FIXME: https://dev.arvados.org/issues/17352
-        grep -q "postgres.client" ${S_DIR}/top.sls || echo "    - postgres.client" >> ${S_DIR}/top.sls
-        grep -q "nginx.passenger" ${S_DIR}/top.sls || echo "    - nginx.passenger" >> ${S_DIR}/top.sls
+        grep -q "    - logrotate" ${S_DIR}/top.sls || echo "    - logrotate" >> ${S_DIR}/top.sls
+        if grep -q "    - nginx.*$" ${S_DIR}/top.sls; then
+          sed -i s/"^    - nginx.*$"/"    - nginx.passenger"/g ${S_DIR}/top.sls
+        else
+          echo "    - nginx.passenger" >> ${S_DIR}/top.sls
+        fi
+        echo "    - extra.passenger_rvm" >> ${S_DIR}/top.sls
         ### If we don't install and run LE before arvados-api-server, it fails and breaks everything
-        ### after it so we add this here, as we are, after all, sharing the host for api and controller
-        # Currently, only available on config_examples/multi_host/aws
-        if [ "x${USE_LETSENCRYPT}" = "xyes" ]; then
-          if [ "x${USE_LETSENCRYPT_IAM_USER}" = "xyes" ]; then
+        ### after it. So we add this here as we are, after all, sharing the host for api and controller
+        if [ "${SSL_MODE}" = "lets-encrypt" ]; then
+          if [ "${USE_LETSENCRYPT_ROUTE53}" = "yes" ]; then
             grep -q "aws_credentials" ${S_DIR}/top.sls || echo "    - aws_credentials" >> ${S_DIR}/top.sls
           fi
-          grep -q "letsencrypt"     ${S_DIR}/top.sls || echo "    - letsencrypt" >> ${S_DIR}/top.sls
+          grep -q "letsencrypt" ${S_DIR}/top.sls || echo "    - letsencrypt" >> ${S_DIR}/top.sls
+        else
+          # Use custom certs
+          if [ "${SSL_MODE}" = "bring-your-own" ]; then
+            copy_custom_cert ${CUSTOM_CERTS_DIR} controller
+          fi
+          grep -q controller ${P_DIR}/extra_custom_certs.sls || echo "  - controller" >> ${P_DIR}/extra_custom_certs.sls
         fi
         grep -q "arvados.${R}" ${S_DIR}/top.sls    || echo "    - arvados.${R}" >> ${S_DIR}/top.sls
         # Pillars
+        grep -q "logrotate_api" ${P_DIR}/top.sls            || echo "    - logrotate_api" >> ${P_DIR}/top.sls
         grep -q "aws_credentials" ${P_DIR}/top.sls          || echo "    - aws_credentials" >> ${P_DIR}/top.sls
-        grep -q "docker" ${P_DIR}/top.sls                   || echo "    - docker" >> ${P_DIR}/top.sls
         grep -q "postgresql" ${P_DIR}/top.sls               || echo "    - postgresql" >> ${P_DIR}/top.sls
         grep -q "nginx_passenger" ${P_DIR}/top.sls          || echo "    - nginx_passenger" >> ${P_DIR}/top.sls
         grep -q "nginx_${R}_configuration" ${P_DIR}/top.sls || echo "    - nginx_${R}_configuration" >> ${P_DIR}/top.sls
+
+        # We need to tweak the Nginx's pillar depending whether we want plain nginx or nginx+passenger
+        NGINX_INSTALL_SOURCE="install_from_phusionpassenger"
+        sed -i "s/__NGINX_INSTALL_SOURCE__/${NGINX_INSTALL_SOURCE}/g" ${P_DIR}/nginx_passenger.sls
       ;;
       "controller" | "websocket" | "workbench" | "workbench2" | "webshell" | "keepweb" | "keepproxy")
         # States
-        grep -q "nginx.passenger" ${S_DIR}/top.sls || echo "    - nginx.passenger" >> ${S_DIR}/top.sls
-        # Currently, only available on config_examples/multi_host/aws
-        if [ "x${USE_LETSENCRYPT}" = "xyes" ]; then
-          if [ "x${USE_LETSENCRYPT_IAM_USER}" = "xyes" ]; then
+        if [ "${R}" = "workbench" ]; then
+          grep -q "    - logrotate" ${S_DIR}/top.sls || echo "    - logrotate" >> ${S_DIR}/top.sls
+          NGINX_INSTALL_SOURCE="install_from_phusionpassenger"
+          if grep -q "    - nginx$" ${S_DIR}/top.sls; then
+            sed -i s/"^    - nginx.*$"/"    - nginx.passenger"/g ${S_DIR}/top.sls
+          else
+            echo "    - nginx.passenger" >> ${S_DIR}/top.sls
+          fi
+        else
+          grep -q "\- nginx$" ${S_DIR}/top.sls || echo "    - nginx" >> ${S_DIR}/top.sls
+        fi
+        if [ "${SSL_MODE}" = "lets-encrypt" ]; then
+          if [ "x${USE_LETSENCRYPT_ROUTE53}" = "xyes" ]; then
             grep -q "aws_credentials" ${S_DIR}/top.sls || echo "    - aws_credentials" >> ${S_DIR}/top.sls
           fi
           grep -q "letsencrypt"     ${S_DIR}/top.sls || echo "    - letsencrypt" >> ${S_DIR}/top.sls
+        else
+          # Use custom certs, special case for keepweb
+          if [ ${R} = "keepweb" ]; then
+            if [ "${SSL_MODE}" = "bring-your-own" ]; then
+              copy_custom_cert ${CUSTOM_CERTS_DIR} download
+              copy_custom_cert ${CUSTOM_CERTS_DIR} collections
+            fi
+          else
+            if [ "${SSL_MODE}" = "bring-your-own" ]; then
+              copy_custom_cert ${CUSTOM_CERTS_DIR} ${R}
+            fi
+          fi
         fi
         # webshell role is just a nginx vhost, so it has no state
         if [ "${R}" != "webshell" ]; then
-          grep -q "arvados.${R}" ${S_DIR}/top.sls    || echo "    - arvados.${R}" >> ${S_DIR}/top.sls
+          grep -q "arvados.${R}" ${S_DIR}/top.sls || echo "    - arvados.${R}" >> ${S_DIR}/top.sls
         fi
         # Pillars
+        if [ "${R}" = "workbench" ]; then
+          grep -q "logrotate_wb1" ${P_DIR}/top.sls || echo "    - logrotate_wb1" >> ${P_DIR}/top.sls
+        fi
         grep -q "nginx_passenger" ${P_DIR}/top.sls          || echo "    - nginx_passenger" >> ${P_DIR}/top.sls
         grep -q "nginx_${R}_configuration" ${P_DIR}/top.sls || echo "    - nginx_${R}_configuration" >> ${P_DIR}/top.sls
-        # Currently, only available on config_examples/multi_host/aws
-        if [ "x${USE_LETSENCRYPT}" = "xyes" ]; then
-          if [ "x${USE_LETSENCRYPT_IAM_USER}" = "xyes" ]; then
+        # Special case for keepweb
+        if [ ${R} = "keepweb" ]; then
+          grep -q "nginx_download_configuration" ${P_DIR}/top.sls || echo "    - nginx_download_configuration" >> ${P_DIR}/top.sls
+          grep -q "nginx_collections_configuration" ${P_DIR}/top.sls || echo "    - nginx_collections_configuration" >> ${P_DIR}/top.sls
+        fi
+
+        if [ "${SSL_MODE}" = "lets-encrypt" ]; then
+          if [ "${USE_LETSENCRYPT_ROUTE53}" = "yes" ]; then
             grep -q "aws_credentials" ${P_DIR}/top.sls || echo "    - aws_credentials" >> ${P_DIR}/top.sls
           fi
           grep -q "letsencrypt"     ${P_DIR}/top.sls || echo "    - letsencrypt" >> ${P_DIR}/top.sls
           grep -q "letsencrypt_${R}_configuration" ${P_DIR}/top.sls || echo "    - letsencrypt_${R}_configuration" >> ${P_DIR}/top.sls
+
+          # As the pillar differ whether we use LE or custom certs, we need to do a final edition on them
+          # Special case for keepweb
+          if [ ${R} = "keepweb" ]; then
+            for kwsub in download collections; do
+              sed -i "s/__CERT_REQUIRES__/cmd: create-initial-cert-${kwsub}.${DOMAIN}*/g;
+                      s#__CERT_PEM__#/etc/letsencrypt/live/${kwsub}.${DOMAIN}/fullchain.pem#g;
+                      s#__CERT_KEY__#/etc/letsencrypt/live/${kwsub}.${DOMAIN}/privkey.pem#g" \
+              ${P_DIR}/nginx_${kwsub}_configuration.sls
+            done
+          else
+            sed -i "s/__CERT_REQUIRES__/cmd: create-initial-cert-${R}.${DOMAIN}*/g;
+                    s#__CERT_PEM__#/etc/letsencrypt/live/${R}.${DOMAIN}/fullchain.pem#g;
+                    s#__CERT_KEY__#/etc/letsencrypt/live/${R}.${DOMAIN}/privkey.pem#g" \
+            ${P_DIR}/nginx_${R}_configuration.sls
+          fi
+        else
+          grep -q "ssl_key_encrypted" ${P_DIR}/top.sls || echo "    - ssl_key_encrypted" >> ${P_DIR}/top.sls
+          # As the pillar differ whether we use LE or custom certs, we need to do a final edition on them
+          # Special case for keepweb
+          if [ ${R} = "keepweb" ]; then
+            for kwsub in download collections; do
+              sed -i "s/__CERT_REQUIRES__/file: extra_custom_certs_file_copy_arvados-${kwsub}.pem/g;
+                      s#__CERT_PEM__#/etc/nginx/ssl/arvados-${kwsub}.pem#g;
+                      s#__CERT_KEY__#/etc/nginx/ssl/arvados-${kwsub}.key#g" \
+              ${P_DIR}/nginx_${kwsub}_configuration.sls
+              grep -q ${kwsub} ${P_DIR}/extra_custom_certs.sls || echo "  - ${kwsub}" >> ${P_DIR}/extra_custom_certs.sls
+            done
+          else
+            sed -i "s/__CERT_REQUIRES__/file: extra_custom_certs_file_copy_arvados-${R}.pem/g;
+                    s#__CERT_PEM__#/etc/nginx/ssl/arvados-${R}.pem#g;
+                    s#__CERT_KEY__#/etc/nginx/ssl/arvados-${R}.key#g" \
+            ${P_DIR}/nginx_${R}_configuration.sls
+            grep -q ${R} ${P_DIR}/extra_custom_certs.sls || echo "  - ${R}" >> ${P_DIR}/extra_custom_certs.sls
+          fi
         fi
+        # We need to tweak the Nginx's pillar depending whether we want plain nginx or nginx+passenger
+        sed -i "s/__NGINX_INSTALL_SOURCE__/${NGINX_INSTALL_SOURCE}/g" ${P_DIR}/nginx_passenger.sls
       ;;
       "shell")
         # States
+        echo "    - extra.shell_sudo_passwordless" >> ${S_DIR}/top.sls
+        echo "    - extra.shell_cron_add_login_sync" >> ${S_DIR}/top.sls
         grep -q "docker" ${S_DIR}/top.sls       || echo "    - docker.software" >> ${S_DIR}/top.sls
         grep -q "arvados.${R}" ${S_DIR}/top.sls || echo "    - arvados.${R}" >> ${S_DIR}/top.sls
         # Pillars
-        grep -q "" ${P_DIR}/top.sls                             || echo "    - docker" >> ${P_DIR}/top.sls
+        grep -q "docker" ${P_DIR}/top.sls       || echo "    - docker" >> ${P_DIR}/top.sls
       ;;
-      "dispatcher")
-        # States
-        grep -q "docker" ${S_DIR}/top.sls       || echo "    - docker.software" >> ${S_DIR}/top.sls
-        grep -q "arvados.${R}" ${S_DIR}/top.sls || echo "    - arvados.${R}" >> ${S_DIR}/top.sls
-        # Pillars
-        # ATM, no specific pillar needed
-      ;;
-      "keepstore")
+      "dispatcher" | "keepbalance" | "keepstore")
         # States
         grep -q "arvados.${R}" ${S_DIR}/top.sls || echo "    - arvados.${R}" >> ${S_DIR}/top.sls
         # Pillars
@@ -500,48 +940,43 @@ else
   done
 fi
 
-# FIXME! #16992 Temporary fix for psql call in arvados-api-server
-if [ -e /root/.psqlrc ]; then
-  if ! ( grep 'pset pager off' /root/.psqlrc ); then
-    RESTORE_PSQL="yes"
-    cp /root/.psqlrc /root/.psqlrc.provision.backup
-  fi
-else
-  DELETE_PSQL="yes"
+if [ "${DUMP_CONFIG}" = "yes" ]; then
+  # We won't run the rest of the script because we're just dumping the config
+  exit 0
 fi
 
-echo '\pset pager off' >> /root/.psqlrc
-# END FIXME! #16992 Temporary fix for psql call in arvados-api-server
-
 # Now run the install
-salt-call --local state.apply -l ${LOG_LEVEL}
+salt-call --state-output=mixed --local state.apply -l ${LOG_LEVEL}
 
-# FIXME! #16992 Temporary fix for psql call in arvados-api-server
-if [ "x${DELETE_PSQL}" = "xyes" ]; then
-  echo "Removing .psql file"
-  rm /root/.psqlrc
+# Finally, make sure that /etc/hosts is not overwritten on reboot
+if [ -d /etc/cloud/cloud.cfg.d ]; then
+  # TODO: will this work on CentOS?
+  sed -i 's/^manage_etc_hosts: true/#manage_etc_hosts: true/g' /etc/cloud/cloud.cfg.d/*
 fi
 
-if [ "x${RESTORE_PSQL}" = "xyes" ]; then
-  echo "Restoring .psql file"
-  mv -v /root/.psqlrc.provision.backup /root/.psqlrc
+# Leave a copy of the Arvados CA so the user can copy it where it's required
+if [ "${SSL_MODE}" = "self-signed" ]; then
+  echo "Copying the Arvados CA certificate '${DOMAIN}-arvados-snakeoil-ca.crt' to the installer dir, so you can import it"
+  if [ "x${VAGRANT}" = "xyes" ]; then
+    cp /etc/ssl/certs/arvados-snakeoil-ca.pem /vagrant/${DOMAIN}-arvados-snakeoil-ca.pem
+  else
+    cp /etc/ssl/certs/arvados-snakeoil-ca.pem ${SCRIPT_DIR}/${DOMAIN}-arvados-snakeoil-ca.crt
+  fi
 fi
-# END FIXME! #16992 Temporary fix for psql call in arvados-api-server
 
-# Leave a copy of the Arvados CA so the user can copy it where it's required
-echo "Copying the Arvados CA certificate to the installer dir, so you can import it"
-# If running in a vagrant VM, also add default user to docker group
 if [ "x${VAGRANT}" = "xyes" ]; then
-  cp /etc/ssl/certs/arvados-snakeoil-ca.pem /vagrant/${CLUSTER}.${DOMAIN}-arvados-snakeoil-ca.pem
-
-  echo "Adding the vagrant user to the docker group"
-  usermod -a -G docker vagrant
-else
-  cp /etc/ssl/certs/arvados-snakeoil-ca.pem ${SCRIPT_DIR}/${CLUSTER}.${DOMAIN}-arvados-snakeoil-ca.pem
+    # If running in a vagrant VM, also add default user to docker group
+    echo "Adding the vagrant user to the docker group"
+    usermod -a -G docker vagrant
 fi
 
 # Test that the installation finished correctly
 if [ "x${TEST}" = "xyes" ]; then
-  cd /tmp/cluster_tests
-  ./run-test.sh
+  cd ${T_DIR}
+  # If we use RVM, we need to run this with it, or most ruby commands will fail
+  RVM_EXEC=""
+  if [ -x /usr/local/rvm/bin/rvm-exec ]; then
+    RVM_EXEC="/usr/local/rvm/bin/rvm-exec"
+  fi
+  ${RVM_EXEC} ./run-test.sh
 fi