Added soft-serve

Added graphana/prometheus (not working yet)

Signed-off-by: TuDatTr <tuan-dat.tran@tudattr.dev>

README.md

@@ -138,8 +138,18 @@ used ipv4:
   interface: wg_tunnel01
 
 ## troubleshooting
-when problems with docker: `docker system prune -a`
+### Docker networking problem
+`docker system prune -a`
+### Time problems (NTP service: n/a)
+systemctl status systemd-timesyncd.service
+when not available
+sudo apt install systemd-timesyncd/stable
+### Syncthing inotify
+echo "fs.inotify.max_user_watches=204800" | sudo tee -a /etc/sysctl.conf
+https://forum.cloudron.io/topic/7163/how-to-increase-inotify-limit-for-syncthing/2
 
 ## Todo
 - Role to load customization/configurations from backup to servers
 - split docker containers to different composes or tag
+- syncthing and grafana config on hosts
+- grafana/prometheus

aya01.yml

@@ -9,5 +9,3 @@
         - backup
     - role: power_management
     - role: docker
-      tags:
-        - reload_compose

group_vars/all/vars.yml

@@ -36,6 +36,7 @@ common_packages:
   - curl
   - tree
   - rsync
+  - systemd-timesyncd
   
 #
 # Docker
@@ -65,6 +66,11 @@ grafana_data: "{{docker_data_dir}}/grafana/"
 grafana_log: "{{docker_dir}}/grafana/logs/"
 grafana_config: "{{docker_dir}}/grafana/config/"
 
+prometheus_data: "{{docker_data_dir}}/prometheus/"
+prometheus_config: "{{docker_dir}}/prometheus/config"
+
+softserve_data: "{{docker_dir}}/softserve/data"
+
 #
 # pi
 #

roles/backblaze/tasks/backup.yml

@@ -11,7 +11,7 @@
   loop: "{{ backblaze_paths | dict2items | subelements('value') }}"
   become: true
 
-- name: Shut down docker
+- name: Restart docker
   shell:
-    cmd: "docker compose down --remove-orphans"
+    cmd: "docker compose up -d"
     chdir: "{{ docker_compose_dir }}"

roles/docker/tasks/aya01_compose.yml

@@ -10,6 +10,7 @@
   file:
     path: "{{ item }}"
     owner: 911
+    group: 911
     mode: '700'
     state: directory
   loop:
@@ -20,6 +21,7 @@
   file:
     path: "{{ item }}"
     owner: 911
+    group: 911
     mode: '755'
     state: directory
   loop:
@@ -29,23 +31,51 @@
 - name: Create syncthing directory
   file:
     path: "{{ item }}"
-    owner: 1000
+    owner: "{{ puid }}"
+    group: "{{ pgid }}"
     mode: '755'
     state: directory
   loop:
     - "{{ syncthing_data }}"
   become: true
 
+- name: Resolve inotify error for syncthing
+  template:
+    src: "templates/aya01/syncthing/syncthing.conf"
+    dest: "/etc/sysctl.d/syncthing.conf"
+    mode: "660"
+  become: true
+
 - name: Create grafana data directory
   file:
     path: "{{ item }}"
-    owner: 1000
+    owner: "{{ puid }}"
+    group: "{{ pgid }}"
     mode: '755'
     state: directory
   loop:
     - "{{ grafana_data }}"
     - "{{ grafana_log }}"
-    - "{{ grafana_config}}"
+    - "{{ grafana_config }}"
+  become: true
+
+- name: Copy grafana config
+  template:
+    owner: "{{ puid }}"
+    src: "templates/aya01/grafana/etc-grafana/grafana.ini"
+    dest: "{{ grafana_config }}/grafana.ini"
+    mode: '660'
+  become: true
+
+- name: Create soft-serve directory
+  file:
+    path: "{{ item }}"
+    owner: "{{ puid }}"
+    group: "{{ pgid }}"
+    mode: '755'
+    state: directory
+  loop:
+    - "{{ softserve_data }}"
   become: true
 
 # Todo, check if docker compose is running

roles/docker/templates/aya01/compose.yaml

@@ -16,7 +16,6 @@ services:
       - "MAX_LOG_SIZE_BYTES=1000000"
       - "MAX_LOG_NUMBER=20"
       - "TZ=Europe/Berlin"
-
   zoneminder:
     image: ghcr.io/zoneminder-containers/zoneminder-base:latest
     restart: always
@@ -45,6 +44,7 @@ services:
       - "MAX_LOG_SIZE_BYTES=1000000"
       - "MAX_LOG_NUMBER=20"
       - "TZ=Europe/Berlin"
+
   pihole:
     container_name: pihole
     image: pihole/pihole:latest
@@ -76,6 +76,7 @@ services:
       - "traefik.http.routers.pihole.rule=Host(`pihole.{{local_domain}}`)"
       - "traefik.http.routers.pihole.entrypoints=web"
       - "traefik.http.services.pihole.loadbalancer.server.port=8089"
+
   syncthing:
     image: syncthing/syncthing
     container_name: syncthing
@@ -83,8 +84,8 @@ services:
     networks:
       - net
     environment:
-      - PUID=1000
+      - "PUID={{ puid }}"
-      - PGID=1000
+      - "PGID={{ pgid }}"
     volumes:
       - "{{syncthing_data}}:/var/syncthing"
     ports:
@@ -93,23 +94,29 @@ services:
       - 22000:22000/udp # QUIC file transfers
       - 21027:21027/udp # Receive local discovery broadcasts
     restart: unless-stopped
+
   grafana:
     image: grafana/grafana-oss
     container_name: grafana
     hostname: grafana
+    user: "{{ puid }}:{{ pgid }}"
     networks:
       - net
     environment:
-      - "PUID={{ puid }}"
-      - "PGID={{ pgid }}"
       - "GF_LOG_MODE=console file"
     volumes:
       - "{{ grafana_data }}:/var/lib/grafana/"
       - "{{ grafana_log }}:/var/log/grafana/"
-      - "{{ grafana_config }}:/etc/grafana/"
     ports:
       - 3000:3000
-
+  soft-serve:
+    image: charmcli/soft-serve:latest
+    container_name: soft-serve
+    volumes:
+      - "{{ softserve_data }}:/soft-serve"
+    ports:
+      - 23231:23231
+    restart: unless-stopped
 
 networks:
   zoneminder:

roles/docker/templates/aya01/prometheus/prometheus.yml

@@ -0,0 +1,54 @@
+# Sample config for Prometheus.
+
+global:
+  scrape_interval:     15s # Set the scrape interval to every 15 seconds. Default is every 1 minute.
+  evaluation_interval: 15s # Evaluate rules every 15 seconds. The default is every 1 minute.
+  # scrape_timeout is set to the global default (10s).
+
+  # Attach these labels to any time series or alerts when communicating with
+  # external systems (federation, remote storage, Alertmanager).
+  external_labels:
+      monitor: 'Mikrotik'
+
+# Alertmanager configuration
+alerting:
+  alertmanagers:
+  - static_configs:
+    - targets: ['localhost:9093']
+
+# Load rules once and periodically evaluate them according to the global 'evaluation_interval'.
+rule_files:
+  # - "first_rules.yml"
+  # - "second_rules.yml"
+
+# A scrape configuration containing exactly one endpoint to scrape:
+# Here it's Prometheus itself.
+scrape_configs:
+  # The job name is added as a label `job=<job_name>` to any timeseries scraped from this config.
+  - job_name: 'prometheus'
+
+    # Override the global default and scrape targets from this job every 5 seconds.
+    scrape_interval: 10s
+    scrape_timeout: 10s
+    tls_config:
+      insecure_skip_verify: true
+    # metrics_path defaults to '/metrics'
+    # scheme defaults to 'http'.
+
+    #static_configs:
+      #- targets: ['localhost:9090']
+
+  - job_name: Mikrotik
+    static_configs:
+      - targets:
+        - {{ mikrotik_ip }} # mikrotik_ip
+    metrics_path: /snmp
+    params:
+      module: [mikrotik]
+    relabel_configs:
+      - source_labels: [__address__]
+        target_label: __param_target
+      - source_labels: [__param_target]
+        target_label: instance
+      - target_label: __address__
+        replacement: mk_snmp_exporter:9116  # The SNMP exporter's real hostname:port.

roles/docker/templates/aya01/syncthing/syncthing.conf

@@ -0,0 +1 @@
+fs.inotify.max_user_watches=204800