PMM-14170: Add Valkey setup and configuration scripts (#154)

* PMM-14170: Add Valkey setup and configuration scripts

* PMM-14170 Fixes peter review

* PMM-14170 Fix peter comments

* PMM-14170 Fix comments

* PMM-14170 Fix loop for pmm client

* PMM-14170 Variables setup

* PMM-14170 Fix OL-10 Password

* PMM-14170 fix loop

* test

* test loop

* test loop

* PMM-14170 Review of Peter

* PMM-14170 Update cleanup.sh

* PMM-14170 Review from Vasyl

* PMM-14170 Fix random number

Author: travagliad

pmm_qa/pmm-framework.py

@@ -789,6 +789,31 @@ def setup_dockerclients(db_type, db_version=None, db_config=None, args=None):
     # Call the function to run the setup_docker_client_images script
     execute_shell_scripts(shell_scripts, shell_scripts_path, env_vars, args)
 
+def setup_valkey(db_type, db_version=None, db_config=None, args=None):
+
+    # Check if PMM server is running
+    container_name = get_running_container_name()
+    if container_name is None and args.pmm_server_ip is None:
+        print(f"Check if PMM Server is Up and Running..Exiting")
+        exit()
+
+    # Gather Version details
+    valkey_version = os.getenv('VALKEY_VERSION') or db_version or database_configs[db_type]["versions"][-1]
+
+    # Define environment variables for playbook
+    env_vars = {
+        'PMM_SERVER_IP': args.pmm_server_ip or container_name or '127.0.0.1',
+        'VALKEY_VERSION': valkey_version,
+        'CLIENT_VERSION': get_value('CLIENT_VERSION', db_type, args, db_config),
+        'ADMIN_PASSWORD': os.getenv('ADMIN_PASSWORD') or args.pmm_server_password or 'admin',
+        'PMM_QA_GIT_BRANCH': os.getenv('PMM_QA_GIT_BRANCH') or 'v3'
+    }
+
+    # Ansible playbook filename
+    playbook_filename = 'valkey/valkey.yml'
+
+    # Call the function to run the Ansible playbook
+    run_ansible_playbook(playbook_filename, env_vars, args)
 
 # Set up databases based on arguments received
 def setup_database(db_type, db_version=None, db_config=None, args=None):
@@ -835,6 +860,8 @@ def setup_database(db_type, db_version=None, db_config=None, args=None):
         setup_ssl_mlaunch(db_type, db_version, db_config, args)
     elif db_type == 'BUCKET':
         setup_bucket(db_type, db_version, db_config, args)
+    elif db_type == 'VALKEY':
+        setup_valkey(db_type, db_version, db_config, args)
     else:
         print(f"Database type {db_type} is not recognised, Exiting...")
         exit(1)

pmm_qa/scripts/database_options.py

@@ -76,5 +76,9 @@
     },
     "BUCKET": {
         "configurations": {"BUCKET_NAMES": 'bcp'}
+    },
+    "VALKEY": {
+        "versions": ["7", "8"],
+        "configurations": {"CLIENT_VERSION": "3-dev-latest", "SETUP_TYPE": "", "TARBALL": ""}
     }
 }

pmm_qa/valkey/cleanup.sh

@@ -0,0 +1,15 @@
+#!/bin/bash -e
+
+docker exec -it pmm-server pmm-admin remove valkey valkey-primary-svc || :
+docker exec -it pmm-server pmm-admin remove valkey valkey-replica1-svc || :
+docker exec -it pmm-server pmm-admin remove valkey valkey-replica2-svc || :
+docker exec -it pmm-server pmm-admin remove valkey sentinel1-svc || :
+docker exec -it pmm-server pmm-admin remove valkey sentinel2-svc || :
+docker exec -it pmm-server pmm-admin remove valkey sentinel3-svc || :
+
+docker rm -vf valkey-primary valkey-replica-1 valkey-replica-2 || :
+docker rm -vf sentinel-1 sentinel-2 sentinel-3 || :
+
+docker volume rm -f valkey-primary-data valkey-replica-1-data valkey-replica-2-data || :
+
+rm -rf "$HOME/valkey"

pmm_qa/valkey/sentinel.conf.j2

@@ -0,0 +1,22 @@
+# sentinel.conf
+bind 0.0.0.0
+
+port 26379
+
+# Monitor the master
+sentinel monitor valkey-primary valkey-primary 6379 {{ sentinel_quorum }}
+sentinel auth-user valkey-primary default
+sentinel auth-pass valkey-primary "{{ valkey_password }}"
+sentinel resolve-hostnames yes
+
+# Failover timeouts
+sentinel down-after-milliseconds valkey-primary 5000
+sentinel failover-timeout valkey-primary 10000
+sentinel parallel-syncs valkey-primary 1
+
+# Security
+protected-mode no
+
+# Logging
+loglevel notice
+logfile ""

pmm_qa/valkey/valkey-primary.conf.j2

@@ -0,0 +1,27 @@
+# Basic Valkey configuration for primary
+bind 0.0.0.0
+port 6379
+
+requirepass "{{ valkey_password }}"
+masterauth "{{ valkey_password }}"
+
+# Persistence
+save 900 1
+save 300 10
+save 60 10000
+
+# Replication
+replica-serve-stale-data yes
+replica-read-only yes
+repl-diskless-sync no
+repl-diskless-sync-delay 5
+
+# Security
+protected-mode no
+
+# Logging
+loglevel notice
+logfile ""
+
+# Memory management
+maxmemory-policy allkeys-lru

pmm_qa/valkey/valkey-replica.conf.j2

@@ -0,0 +1,28 @@
+# Basic Valkey configuration for replica
+bind 0.0.0.0
+port 6379
+
+requirepass "{{ valkey_password }}"
+masterauth "{{ valkey_password }}"
+
+# Replication
+replicaof valkey-primary 6379
+replica-serve-stale-data yes
+replica-read-only yes
+repl-diskless-sync no
+repl-diskless-sync-delay 5
+
+# Persistence
+save 900 1
+save 300 10
+save 60 10000
+
+# Security
+protected-mode no
+
+# Logging
+loglevel notice
+logfile ""
+
+# Memory management
+maxmemory-policy allkeys-lru

pmm_qa/valkey/valkey.yml

@@ -0,0 +1,220 @@
+---
+- name: Deploy Valkey Cluster with Sentinel
+  hosts: localhost
+  gather_facts: false
+  vars:
+    pmm_server_ip: "{{ lookup('vars', 'extra_pmm_server_ip', default=lookup('env','PMM_SERVER_IP') | default('127.0.0.1', true) ) }}"
+    metrics_mode: "{{ lookup('env', 'metrics_mode') }}"
+    client_version: "{{ lookup('vars', 'extra_client_version', default=lookup('env','CLIENT_VERSION') | default('3-dev-latest', true) ) }}"
+    admin_password: "{{ lookup('vars', 'extra_admin_password', default=lookup('env','ADMIN_PASSWORD') | default('admin', true) ) }}"
+    valkey_version: "{{ lookup('env', 'VALKEY_VERSION') | default('7', true) }}"
+    valkey_image: "valkey/valkey:{{ valkey_version }}"
+    valkey_network_name: "pmm-qa"
+    valkey_password: "VKvl41568AsE"
+    valkey_data_dir: "{{ lookup('env', 'HOME') }}/valkey/data"
+    valkey_config_dir: "{{ lookup('env', 'HOME') }}/valkey/config"
+    valkey_primary_port: 6379
+    valkey_replica_count: 2
+    valkey_replica_start_port: 6380
+    sentinel_count: 3
+    sentinel_start_port: 26379
+    sentinel_quorum: 2
+
+    pmm_server_name: "pmm-server"
+
+  tasks:
+    - name: Set Random Number Fact
+      set_fact:
+        random_number: "{{ (10000 | random) | int }}"
+
+    - name: Create Docker network
+      community.docker.docker_network:
+        name: "{{ valkey_network_name }}"
+        driver: bridge
+        state: present
+
+    - name: Create config directory
+      file:
+        path: "{{ valkey_config_dir }}"
+        state: directory
+        mode: '0755'
+
+    - name: Create a config directory per Sentinel
+      file:
+        path: "{{ valkey_config_dir }}/sentinel-{{ item }}"
+        state: directory
+        mode: '0755'
+      loop: "{{ range(1, sentinel_count + 1) | list }}"
+
+    - name: Create data directory
+      file:
+        path: "{{ valkey_data_dir }}"
+        state: directory
+        mode: '0755'
+
+    - name: Generate Valkey primary configuration
+      template:
+        src: valkey-primary.conf.j2
+        dest: "{{ valkey_config_dir }}/valkey-primary.conf"
+        mode: '0644'
+
+    - name: Generate Valkey replica configurations
+      template:
+        src: valkey-replica.conf.j2
+        dest: "{{ valkey_config_dir }}/valkey-replica-{{ item }}.conf"
+        mode: '0644'
+      loop: "{{ range(1, valkey_replica_count + 1) | list }}"
+
+    - name: Generate Sentinel configurations
+      template:
+        src: sentinel.conf.j2
+        dest: "{{ valkey_config_dir }}/sentinel-{{ item }}/sentinel.conf"
+        mode: '0664'
+      loop: "{{ range(1, sentinel_count + 1) | list }}"
+
+    - name: Create Docker volume for primary data
+      community.docker.docker_volume:
+        name: "valkey-primary-data"
+        state: present
+
+    - name: Create Docker volumes for replica data
+      community.docker.docker_volume:
+        name: "valkey-replica-{{ item }}-data"
+        state: present
+      loop: "{{ range(1, valkey_replica_count + 1) | list }}"
+
+    - name: Start Valkey primary container
+      community.docker.docker_container:
+        name: "valkey-primary"
+        image: "{{ valkey_image }}"
+        state: started
+        restart_policy: unless-stopped
+        networks:
+          - name: "{{ valkey_network_name }}"
+        ports:
+          - "{{ valkey_primary_port }}:6379"
+        volumes:
+          - "valkey-primary-data:/data"
+          - "{{ valkey_config_dir }}/valkey-primary.conf:/usr/local/etc/valkey/valkey.conf:ro"
+        command: ["valkey-server", "/usr/local/etc/valkey/valkey.conf"]
+        healthcheck:
+          test: ["CMD", "valkey-cli", "-a", "{{ valkey_password }}", "ping"]
+          interval: 10s
+          timeout: 5s
+          retries: 5
+
+    - name: Wait for the primary to be ready
+      wait_for:
+        host: localhost
+        port: "{{ valkey_primary_port }}"
+        timeout: 30
+        delay: 1
+
+    - name: Start Valkey replica containers
+      community.docker.docker_container:
+        name: "valkey-replica-{{ item }}"
+        image: "{{ valkey_image }}"
+        state: started
+        restart_policy: unless-stopped
+        networks:
+          - name: "{{ valkey_network_name }}"
+        ports:
+          - "{{ valkey_replica_start_port + item - 1 }}:6379"
+        volumes:
+          - "valkey-replica-{{ item }}-data:/data"
+          - "{{ valkey_config_dir }}/valkey-replica-{{ item }}.conf:/usr/local/etc/valkey/valkey.conf:ro"
+        command: ["valkey-server", "/usr/local/etc/valkey/valkey.conf"]
+        healthcheck:
+          test: ["CMD", "valkey-cli", "-a", "{{ valkey_password }}", "ping"]
+          interval: 10s
+          timeout: 5s
+          retries: 5
+      loop: "{{ range(1, valkey_replica_count + 1) | list }}"
+
+    - name: Wait for replicas to be ready
+      wait_for:
+        host: localhost
+        port: "{{ valkey_replica_start_port + item - 1 }}"
+        timeout: 30
+        delay: 1
+      loop: "{{ range(1, valkey_replica_count + 1) | list }}"
+
+    - name: Start Sentinel containers
+      community.docker.docker_container:
+        name: "sentinel-{{ item }}"
+        image: "{{ valkey_image }}"
+        state: started
+        restart_policy: unless-stopped
+        networks:
+          - name: "{{ valkey_network_name }}"
+        ports:
+          - "{{ sentinel_start_port + item - 1 }}:26379"
+        volumes:
+          - "{{ valkey_config_dir }}/sentinel-{{ item }}:/usr/local/etc/valkey"
+        command: ["valkey-sentinel", "/usr/local/etc/valkey/sentinel.conf"]
+        healthcheck:
+          test: ["CMD", "valkey-cli", "-p", "{{ sentinel_start_port }}", "ping"]
+          interval: 10s
+          timeout: 5s
+          retries: 5
+      loop: "{{ range(1, sentinel_count + 1) | list }}"
+
+    - name: Wait for Sentinels to be ready
+      wait_for:
+        host: localhost
+        port: "{{ sentinel_start_port + item - 1 }}"
+        timeout: 30
+        delay: 1
+      loop: "{{ range(1, sentinel_count + 1) | list }}"
+
+    - name: Verify cluster status
+      community.docker.docker_container_exec:
+        container: "valkey-primary"
+        command: valkey-cli -a "{{ valkey_password }}" info replication
+      register: cluster_status
+
+    - name: Display cluster status
+      debug:
+        msg: "{{ cluster_status.stdout_lines }}"
+
+    - name: Run Sentinel status command
+      community.docker.docker_container_exec:
+        container: "sentinel-1"
+        command: valkey-cli -p {{ sentinel_start_port }} sentinel masters
+      register: sentinel_status
+
+    - name: Display Sentinel status
+      debug:
+        msg: "{{ sentinel_status.stdout_lines }}"
+
+    - name: Install PMM Client in each container
+      ansible.builtin.include_tasks: ../tasks/install_pmm_client.yml
+      loop: >-
+          {{ ['valkey-primary']
+            + (range(1, valkey_replica_count + 1) | map('string') | map('regex_replace', '^(.*)$', 'valkey-replica-\1') | list)
+            + (range(1, sentinel_count + 1) | map('string') | map('regex_replace', '^(.*)$', 'sentinel-\1') | list)
+          }}
+      loop_control:
+        loop_var: current_container_name
+      vars:
+        container_name: "{{ current_container_name }}"
+
+    - name: Add the primary to monitoring
+      community.docker.docker_container_exec:
+        container: "valkey-primary"
+        command: pmm-admin add valkey --cluster=valkey-cluster --replication-set=valkey-repl --environment=valkey-test --username=default --password="{{ valkey_password }}" --service-name=valkey-primary-svc-{{ random_number }} --host=valkey-primary --port=6379 --custom-labels='role=primary'
+      ignore_errors: yes
+
+    - name: Add the replicas to monitoring
+      community.docker.docker_container_exec:
+        container: "valkey-replica-{{ item }}"
+        command: pmm-admin add valkey --cluster=valkey-cluster --replication-set=valkey-repl --environment=valkey-test --username=default --password="{{ valkey_password }}" --service-name=valkey-replica{{ item }}-svc-{{ random_number }} --host=valkey-replica-{{ item }} --port=6379 --custom-labels='role=replica'
+      loop: "{{ range(1, valkey_replica_count + 1) | list }}"
+      ignore_errors: yes
+
+    - name: Add Sentinels to monitoring
+      community.docker.docker_container_exec:
+        container: "sentinel-{{ item }}"
+        command: pmm-admin add valkey --cluster=valkey-cluster --environment=valkey-test --username=default --password="{{ valkey_password }}" --service-name=sentinel{{ item }}-svc-{{ random_number }} --host=sentinel-{{ item }} --port={{ sentinel_start_port }} --custom-labels='role=sentinel'
+      loop: "{{ range(1, sentinel_count + 1) | list }}"
+      ignore_errors: yes