Merge pull request #1193 from stanislav-zaprudskiy/add_termination_grace_period_seconds

AWX: Add `termination_grace_period_seconds`

roles/installer/files/pre-stop/termination-env

@@ -0,0 +1,66 @@
+# file, which when exists, indicates that `master` script has successfully
+# completed pre-stop script execution
+marker_file="${PRE_STOP_MARKER_FILE:-/var/lib/pre-stop/.termination_marker}"
+
+# file which the running `master` script continuously updates (mtime) to
+# indicate it's still running. this file is then read by `watcher`s to
+# understand if they still have to wait for `termination_marker`
+heartbeat_file="${PRE_STOP_HEARTBEAT_FILE:-/var/lib/pre-stop/.heartbeat}"
+
+# file which:
+#   * `watcher`s create when they bail out because they didn't see the
+#   `heartbeat_file` to be updated within `$heartbeat_failed_threshold`;
+#   * `master` creates when its handler command fails;
+# when scripts see such file, they also give up
+bailout_file="${PRE_STOP_BAILOUT_FILE:-/var/lib/pre-stop/.bailout}"
+heartbeat_threshold="${PRE_STOP_HEARTBEAT_THRESHOLD:-60}"
+
+# where the scripts' stdout/stderr are streamed
+stdout="${PRE_STOP_STDOUT:-/proc/1/fd/1}"
+stderr="${PRE_STOP_STDERR:-/proc/1/fd/2}"
+
+# command the `master` script executes, which when successfully finishes,
+# causes the script to create the `marker_file`
+handler="${PRE_STOP_HANDLER:-bash -c \"PYTHONUNBUFFERED=x awx-manage disable_instance --wait --retry=inf\"}"
+
+log_prefix="${PRE_STOP_LOG_PREFIX:-preStop.exec}"
+[[ -n ${PRE_STOP_LOG_ROLE} ]] && log_prefix="${log_prefix}] [$PRE_STOP_LOG_ROLE"
+
+# interval at which `watcher`s check for `marker_file` presence
+recheck_sleep="${PRE_STOP_RECHECK_SLEEP:-1}"
+# interval at which `watcher`s report into $stdout that they are still watching
+report_every="${PRE_STOP_REPORT_EVERY:-30}"
+
+function log {
+  printf "[%s] $1\n" "$log_prefix" "${@:2}"
+}
+
+function parameters_string {
+  for param in "$@"; do
+    printf "%s=\"%s\"\n" "$param" "${!param}"
+  done | paste -s -d ' '
+}
+
+function check_bailout {
+  if [[ -f $bailout_file ]]; then
+    log "\"%s\" file has been detected, accepting bail out signal and failing the hook script" \
+      "$bailout_file"
+    exit 1
+  fi
+}
+
+function check_heartbeat {
+  if [[ -f $heartbeat_file ]]; then
+    delta=$(( $(date +%s) - $(stat -c %Y "$heartbeat_file") ))
+  else
+    delta=$(( $(date +%s) - $1 ))
+  fi
+
+  if [[ $delta -gt $heartbeat_threshold ]]; then
+    log "The heartbeat file hasn't been updated since %ss, which is above the threshold of %ds, assuming the master is not operating and failing the hook script" \
+      $delta
+      $heartbeat_threshold
+    touch "$bailout_file"
+    exit 1
+  fi
+}

roles/installer/files/pre-stop/termination-master

@@ -0,0 +1,50 @@
+#/usr/bin/env bash
+
+PRE_STOP_LOG_ROLE="${PRE_STOP_LOG_ROLE:-master}"
+source $(dirname "$0")/termination-env
+
+{
+
+log "The hook has started: %s" \
+  "$(parameters_string \
+    "marker_file" \
+    "heartbeat_file" \
+    "bailout_file" \
+    "handler" \
+  )"
+
+touch "$heartbeat_file"
+
+set -o pipefail
+eval "$handler" 2>&1 | while IFS= read -r line; do
+  # we check the files here and break early, but overall script termination
+  # happens later - as we need to distinguish between files detection and
+  # command failure, while bash doesn't offer a simple way to do this here
+  # inside the loop (`exit` does not terminate the script)
+  [[ -f $bailout_file ]] && break
+  [[ -f $marker_file ]] && break
+
+  log "[handler] %s" "$line"
+  touch "$heartbeat_file"
+done
+ec=$?
+set +o pipefail
+
+# process various cases in specific order
+check_bailout
+
+if [[ -f $marker_file ]]; then
+  log "Done! The marker file has been detected, assuming some other instance of the script has run to completion"
+  exit 0
+elif [[ $ec -ne 0 ]]; then
+  log "The handler has failed with \"%d\" exit code, failing the hook script too" \
+    $ec
+  # signal others to bail out
+  touch "$bailout_file"
+  exit $ec
+else
+  log "Done! Generating the marker file allowing to proceed to termination"
+  touch "$marker_file"
+fi
+
+} > "$stdout" 2> "$stderr"

roles/installer/files/pre-stop/termination-waiter

@@ -0,0 +1,33 @@
+#/usr/bin/env bash
+
+PRE_STOP_LOG_ROLE="${PRE_STOP_LOG_ROLE:-waiter}"
+source $(dirname "$0")/termination-env
+
+{
+
+log "The hook has started: %s" \
+  "$(parameters_string \
+    "marker_file" \
+    "heartbeat_file" \
+    "bailout_file" \
+    "recheck_sleep" \
+    "report_every" \
+  )"
+
+n=0
+checks_started=$(date +%s)
+
+while ! [[ -f $marker_file ]]; do
+  check_bailout
+  check_heartbeat $checks_started
+
+  if [[ $(($n % $report_every)) -eq 0 ]]; then
+    log "Waiting for the marker file to be accessible..."
+  fi
+  n=$(($n + 1))
+  sleep $recheck_sleep
+done
+
+log "The marker file found, exiting to proceed to termination"
+
+} > "$stdout" 2> "$stderr"

roles/installer/tasks/install.yml

@@ -39,17 +39,17 @@
 - name: Load LDAP CAcert certificate
   include_tasks: load_ldap_cacert_secret.yml
   when:
-    - ldap_cacert_secret != ''
+  - ldap_cacert_secret != ''
 
 - name: Load ldap bind password
   include_tasks: load_ldap_password_secret.yml
   when:
-    - ldap_password_secret != ''
+  - ldap_password_secret != ''
 
 - name: Load bundle certificate authority certificate
   include_tasks: load_bundle_cacert_secret.yml
   when:
-    - bundle_cacert_secret != ''
+  - bundle_cacert_secret != ''
 
 - name: Include admin password configuration tasks
   include_tasks: admin_password_configuration.yml
@@ -66,8 +66,8 @@
 - name: Load Route TLS certificate
   include_tasks: load_route_tls_secret.yml
   when:
-    - ingress_type | lower == 'route'
-    - route_tls_secret != ''
+  - ingress_type | lower == 'route'
+  - route_tls_secret != ''
 
 - name: Include resources configuration tasks
   include_tasks: resources_configuration.yml
@@ -91,8 +91,8 @@
       bash -c "awx-manage migrate --noinput"
   register: migrate_result
   when:
-    - database_check is defined
-    - (database_check.stdout|trim) != '0'
+  - database_check is defined
+  - (database_check.stdout|trim) != '0'
 
 - name: Initialize Django
   include_tasks: initialize_django.yml

roles/installer/tasks/resources_configuration.yml

@@ -13,9 +13,17 @@
       - status.phase=Running
   register: tower_pod
 
+- name: Set the resource pod as a variable.
+  set_fact:
+    tower_pod: >-
+      {{ tower_pod['resources']
+         | rejectattr('metadata.deletionTimestamp', 'defined')
+         | sort(attribute='metadata.creationTimestamp')
+         | first | default({}) }}
+
 - name: Set the resource pod name as a variable.
   set_fact:
-    tower_pod_name: "{{ tower_pod['resources'][0]['metadata']['name'] | default('') }}"
+    tower_pod_name: "{{ tower_pod['metadata']['name'] | default('') }}"
 
 - name: Set user provided control plane ee image
   set_fact:
@@ -32,13 +40,13 @@
     kind: Secret
     namespace: '{{ ansible_operator_meta.namespace }}'
     name: '{{ ansible_operator_meta.name }}-receptor-ca'
-  register: _receptor_ca
+  register: receptor_ca
   no_log: "{{ no_log }}"
 
 - name: Migrate Receptor CA Secret
   when:
-    - _receptor_ca['resources'] | default([]) | length
-    - _receptor_ca['resources'][0]['type'] != "kubernetes.io/tls"
+    - receptor_ca['resources'] | default([]) | length
+    - receptor_ca['resources'][0]['type'] != "kubernetes.io/tls"
   block:
     - name: Delete old Receptor CA Secret
       k8s:
@@ -53,7 +61,7 @@
       register: _receptor_ca_key_file
     - name: Copy Receptor CA key from old secret to tempfile
       copy:
-        content: "{{ _receptor_ca['resources'][0]['data']['receptor-ca.key'] | b64decode }}"
+        content: "{{ receptor_ca['resources'][0]['data']['receptor-ca.key'] | b64decode }}"
         dest: "{{ _receptor_ca_key_file.path }}"
       no_log: "{{ no_log }}"
     - name: Create tempfile for receptor-ca.crt
@@ -63,7 +71,7 @@
       register: _receptor_ca_crt_file
     - name: Copy Receptor CA cert from old secret to tempfile
       copy:
-        content: "{{ _receptor_ca['resources'][0]['data']['receptor-ca.crt'] | b64decode }}"
+        content: "{{ receptor_ca['resources'][0]['data']['receptor-ca.crt'] | b64decode }}"
         dest: "{{ _receptor_ca_crt_file.path }}"
       no_log: "{{ no_log }}"
     - name: Create New Receptor CA secret
@@ -71,6 +79,17 @@
         apply: true
         definition: "{{ lookup('template', 'secrets/receptor_ca_secret.yaml.j2') }}"
       no_log: "{{ no_log }}"
+    - name: Read New Receptor CA Secret
+      k8s_info:
+        kind: Secret
+        namespace: '{{ ansible_operator_meta.namespace }}'
+        name: '{{ ansible_operator_meta.name }}-receptor-ca'
+      register: _receptor_ca
+      no_log: "{{ no_log }}"
+    - name: Set receptor_ca variable
+      set_fact:
+        receptor_ca: '{{ _receptor_ca }}'
+      no_log: "{{ no_log }}"
     - name: Remove tempfiles
       file:
         path: "{{ item }}"
@@ -106,6 +125,17 @@
         apply: true
         definition: "{{ lookup('template', 'secrets/receptor_ca_secret.yaml.j2') }}"
       no_log: "{{ no_log }}"
+    - name: Read Receptor CA secret
+      k8s_info:
+        kind: Secret
+        namespace: '{{ ansible_operator_meta.namespace }}'
+        name: '{{ ansible_operator_meta.name }}-receptor-ca'
+      register: _receptor_ca
+      no_log: "{{ no_log }}"
+    - name: Set receptor_ca variable
+      set_fact:
+        receptor_ca: '{{ _receptor_ca }}'
+      no_log: "{{ no_log }}"
     - name: Remove tempfiles
       file:
         path: "{{ item }}"
@@ -113,14 +143,14 @@
       loop:
         - "{{ _receptor_ca_key_file.path }}"
         - "{{ _receptor_ca_crt_file.path }}"
-  when: not _receptor_ca['resources'] | default([]) | length
+  when: not receptor_ca['resources'] | default([]) | length
 
 - name: Check for Receptor work signing Secret
   k8s_info:
     kind: Secret
     namespace: '{{ ansible_operator_meta.namespace }}'
     name: '{{ ansible_operator_meta.name }}-receptor-work-signing'
-  register: _receptor_work_signing
+  register: receptor_work_signing
   no_log: "{{ no_log }}"
 
 - name: Generate Receptor work signing RSA key pair
@@ -151,6 +181,17 @@
         apply: true
         definition: "{{ lookup('template', 'secrets/receptor_work_signing_secret.yaml.j2') }}"
       no_log: "{{ no_log }}"
+    - name: Read Receptor work signing Secret
+      k8s_info:
+        kind: Secret
+        namespace: '{{ ansible_operator_meta.namespace }}'
+        name: '{{ ansible_operator_meta.name }}-receptor-work-signing'
+      register: _receptor_work_signing
+      no_log: "{{ no_log }}"
+    - name: Set receptor_work_signing variable
+      set_fact:
+        receptor_work_signing: '{{ _receptor_work_signing }}'
+      no_log: "{{ no_log }}"
     - name: Remove tempfiles
       file:
         path: "{{ item }}"
@@ -158,16 +199,16 @@
       loop:
         - "{{ _receptor_work_signing_private_key_file.path }}"
         - "{{ _receptor_work_signing_public_key_file.path }}"
-  when: not _receptor_work_signing['resources'] | default([]) | length
+  when: not receptor_work_signing['resources'] | default([]) | length
 
 - name: Apply Resources
   k8s:
     apply: yes
     definition: "{{ lookup('template', item + '.yaml.j2') }}"
     wait: yes
-  register: tower_resources_result
   loop:
     - 'configmaps/config'
+    - 'configmaps/pre_stop_scripts'
     - 'secrets/app_credentials'
     - 'rbac/service_account'
     - 'storage/persistent'
@@ -210,21 +251,10 @@
     apply: yes
     definition: "{{ lookup('template', 'deployments/deployment.yaml.j2') }}"
     wait: yes
+    wait_timeout: "{{ 120 * replicas or 120 }}"
   register: this_deployment_result
 
 - block:
-    - name: Delete pod to reload a resource configuration
-      k8s:
-        api_version: v1
-        state: absent
-        kind: Pod
-        namespace: '{{ ansible_operator_meta.namespace }}'
-        name: '{{ tower_pod_name }}'
-        wait: yes
-      when:
-        - tower_resources_result.changed
-        - tower_pod_name | length
-
     - name: Get the new resource pod information after updating resource.
       k8s_info:
         kind: Pod
@@ -236,17 +266,20 @@
         field_selectors:
           - status.phase=Running
       register: _new_pod
-      until:
-        - _new_pod['resources'] | length
-        - _new_pod['resources'][0]['metadata']['name'] != tower_pod_name
-      delay: 5
-      retries: 60
+
+    - name: Update new resource pod as a variable.
+      set_fact:
+        tower_pod: >-
+          {{ _new_pod['resources']
+             | rejectattr('metadata.deletionTimestamp', 'defined')
+             | sort(attribute='metadata.creationTimestamp')
+             | last | default({}) }}
 
     - name: Update new resource pod name as a variable.
       set_fact:
-        tower_pod_name: '{{ _new_pod["resources"][0]["metadata"]["name"] }}'
+        tower_pod_name: '{{ tower_pod["metadata"]["name"] | default("")}}'
   when:
-    - tower_resources_result.changed or this_deployment_result.changed
+    - this_deployment_result.changed
 
 - name: Verify the resource pod name is populated.
   assert:

roles/installer/tasks/secret_key_configuration.yml

@@ -40,10 +40,10 @@
 
 - name: Set secret key secret
   set_fact:
-    __secret_key_secret: '{{ _generated_secret_key["resources"] | default([]) | length | ternary(_generated_secret_key, _secret_key_secret)  }}'
+    secret_key: '{{ _generated_secret_key["resources"] | default([]) | length | ternary(_generated_secret_key, _secret_key_secret)  }}'
   no_log: "{{ no_log }}"
 
 - name: Store secret key secret name
   set_fact:
-    secret_key_secret_name: "{{ __secret_key_secret['resources'][0]['metadata']['name'] }}"
+    secret_key_secret_name: "{{ secret_key['resources'][0]['metadata']['name'] }}"
   no_log: "{{ no_log }}"

roles/installer/templates/configmaps/pre_stop_scripts.yaml.j2

@@ -0,0 +1,16 @@
+{% if termination_grace_period_seconds is defined %}
+apiVersion: v1
+kind: ConfigMap
+metadata:
+  name: '{{ ansible_operator_meta.name }}-{{ deployment_type }}-pre-stop-scripts'
+  namespace: '{{ ansible_operator_meta.namespace }}'
+  labels:
+    {{ lookup("template", "../common/templates/labels/common.yaml.j2") | indent(width=4) | trim }}
+data:
+  termination-master: |
+    {{ lookup("file", "files/pre-stop/termination-master") | indent(width=4) | trim }}
+  termination-waiter: |
+    {{ lookup("file", "files/pre-stop/termination-waiter") | indent(width=4) | trim }}
+  termination-env: |
+    {{ lookup("file", "files/pre-stop/termination-env") | indent(width=4) | trim }}
+{% endif %}

roles/installer/templates/deployments/deployment.yaml.j2

@@ -20,8 +20,26 @@ spec:
       labels:
         {{ lookup("template", "../common/templates/labels/common.yaml.j2")  | indent(width=8) | trim }}
         {{ lookup("template", "../common/templates/labels/version.yaml.j2") | indent(width=8) | trim }}
-{% if annotations %}
       annotations:
+{% for template in [
+    "configmaps/config",
+    "configmaps/pre_stop_scripts",
+    "secrets/app_credentials",
+    "storage/persistent",
+  ] %}
+        checksum-{{ template | replace('/', '-') }}: "{{ lookup('template', template + '.yaml.j2') | md5 }}"
+{% endfor %}
+{% for secret in [
+    "bundle_cacert",
+    "route_tls",
+    "ldap_cacert",
+    "secret_key",
+    "receptor_ca",
+    "receptor_work_signing",
+  ] %}
+        checksum-secret-{{ secret }}: "{{ lookup('ansible.builtin.vars', secret, default='')["resources"][0]["data"] | default('') | md5 }}"
+{% endfor %}
+{% if annotations %}
         {{ annotations | indent(width=8) }}
 {% endif %}
     spec:
@@ -122,6 +140,23 @@ spec:
               mountPath: "/var/run/redis"
             - name: "{{ ansible_operator_meta.name }}-redis-data"
               mountPath: "/data"
+{% if termination_grace_period_seconds is defined %}
+            - name: pre-stop-data
+              mountPath: /var/lib/pre-stop
+            - name: pre-stop-scripts
+              mountPath: /var/lib/pre-stop/scripts
+          lifecycle:
+            preStop:
+              exec:
+                command:
+                - bash
+                - -c
+                # redis image doesn't support writing to `/proc/1/fd/*`
+                - >
+                  PRE_STOP_STDOUT=/dev/stdout
+                  PRE_STOP_STDERR=/dev/stderr
+                  /var/lib/pre-stop/scripts/termination-waiter
+{% endif %}
           resources: {{ redis_resource_requirements }}
         - image: '{{ _image }}'
           name: '{{ ansible_operator_meta.name }}-web'
@@ -291,6 +326,18 @@ spec:
 {% endif %}
 {% if task_extra_volume_mounts -%}
             {{ task_extra_volume_mounts | indent(width=12, first=True) }}
+{% endif %}
+{% if termination_grace_period_seconds is defined %}
+            - name: pre-stop-data
+              mountPath: /var/lib/pre-stop
+            - name: pre-stop-scripts
+              mountPath: /var/lib/pre-stop/scripts
+          lifecycle:
+            preStop:
+              exec:
+                command:
+                - bash
+                - /var/lib/pre-stop/scripts/termination-master
 {% endif %}
           env:
             - name: SUPERVISOR_WEB_CONFIG_PATH
@@ -360,6 +407,18 @@ spec:
               mountPath: "/var/lib/awx/projects"
 {% if ee_extra_volume_mounts -%}
             {{ ee_extra_volume_mounts | indent(width=12, first=True) }}
+{% endif %}
+{% if termination_grace_period_seconds is defined %}
+            - name: pre-stop-data
+              mountPath: /var/lib/pre-stop
+            - name: pre-stop-scripts
+              mountPath: /var/lib/pre-stop/scripts
+          lifecycle:
+            preStop:
+              exec:
+                command:
+                - bash
+                - /var/lib/pre-stop/scripts/termination-waiter
 {% endif %}
           env:
 {% if development_mode | bool %}
@@ -395,6 +454,9 @@ spec:
 {% if security_context_settings|length %}
         {{ security_context_settings | to_nice_yaml | indent(8) }}
 {% endif %}
+{% endif %}
+{% if termination_grace_period_seconds is defined %}
+      terminationGracePeriodSeconds: {{ termination_grace_period_seconds }}
 {% endif %}
       volumes:
 {% if bundle_ca_crt %}
@@ -424,6 +486,14 @@ spec:
             items:
               - key: ldap-ca.crt
                 path: 'ldap-ca.crt'
+{% endif %}
+{% if termination_grace_period_seconds is defined %}
+        - name: pre-stop-data
+          emptyDir: {}
+        - name: pre-stop-scripts
+          configMap:
+            name: '{{ ansible_operator_meta.name }}-{{ deployment_type }}-pre-stop-scripts'
+            defaultMode: 0775
 {% endif %}
         - name: "{{ ansible_operator_meta.name }}-application-credentials"
           secret: