Harden against bot floods and add post-provisioning service check

- Traefik rate-limits all routes (60 req/min standard, 300 for Nextcloud)
  and writes access logs to /var/log/traefik/access.log
- Fail2ban watches Traefik access logs and bans IPs after 10 x 403/404
  within 60 s for 24 h
- Nextcloud html directory created as www-data (UID 33) so Apache can
  process .htaccess and serve requests correctly
- scripts/check-services.sh verifies all seven endpoints return the
  expected HTTP status after provisioning
This commit is contained in:
ml
2026-06-28 08:02:45 +02:00
parent e197ba7370
commit b15754ba2f
17 changed files with 168 additions and 4 deletions
@@ -0,0 +1,6 @@
[Definition]
# Matches Traefik CLF access log lines where the response status is 403 or 404.
# Banning on these two codes catches vulnerability scanners (403) and path probers (404)
# without touching legitimate 401 auth challenges or 429 rate-limit responses.
failregex = ^<HOST> \S+ \S+ \[.*?\] ".*?" (?:403|404) .*$
ignoreregex =
@@ -0,0 +1,8 @@
[traefik]
enabled = true
port = http,https
filter = traefik
logpath = /var/log/traefik/access.log
maxretry = 10
findtime = 60
bantime = 86400
+5
View File
@@ -3,3 +3,8 @@
ansible.builtin.service:
name: ssh
state: restarted
- name: Restart fail2ban
ansible.builtin.service:
name: fail2ban
state: restarted
+20
View File
@@ -126,6 +126,26 @@
# --- Fail2ban ---
- name: Deploy fail2ban filter for Traefik
ansible.builtin.copy:
src: filter.d/traefik.conf
dest: /etc/fail2ban/filter.d/traefik.conf
owner: root
group: root
mode: "0644"
notify: Restart fail2ban
tags: base
- name: Deploy fail2ban jail for Traefik
ansible.builtin.copy:
src: jail.d/traefik.conf
dest: /etc/fail2ban/jail.d/traefik.conf
owner: root
group: root
mode: "0644"
notify: Restart fail2ban
tags: base
- name: Enable and start fail2ban
ansible.builtin.service:
name: fail2ban
@@ -35,6 +35,7 @@ services:
- "traefik.http.routers.gitea.entrypoints=websecure"
- "traefik.http.routers.gitea.tls.certresolver=letsencrypt"
- "traefik.http.services.gitea.loadbalancer.server.port=3000"
- "traefik.http.routers.gitea.middlewares=rate-limit@docker"
depends_on:
gitea-db:
condition: service_healthy
@@ -14,6 +14,7 @@ services:
- "traefik.http.routers.k8s.entrypoints=websecure"
- "traefik.http.routers.k8s.tls.certresolver=letsencrypt"
- "traefik.http.services.k8s.loadbalancer.server.port=80"
- "traefik.http.routers.k8s.middlewares=rate-limit@docker"
networks:
traefik_public:
@@ -49,6 +49,7 @@ services:
- "traefik.http.routers.mail.entrypoints=websecure"
- "traefik.http.routers.mail.tls.certresolver=letsencrypt"
- "traefik.http.services.mail.loadbalancer.server.port=80"
- "traefik.http.routers.mail.middlewares=rate-limit@docker"
depends_on:
mail-db:
condition: service_healthy
+11 -1
View File
@@ -8,10 +8,20 @@
mode: "0755"
loop:
- "{{ nextcloud_data_dir }}"
- "{{ nextcloud_data_dir }}/html"
- "{{ nextcloud_data_dir }}/redis"
tags: nextcloud
# www-data inside the container is UID 33. The html directory must be owned by
# this user so Apache can traverse it and process .htaccess files.
- name: Create Nextcloud html directory (www-data UID 33)
ansible.builtin.file:
path: "{{ nextcloud_data_dir }}/html"
state: directory
owner: "33"
group: "33"
mode: "0755"
tags: nextcloud
- name: Create Nextcloud database directory (postgres UID 999)
ansible.builtin.file:
path: "{{ nextcloud_data_dir }}/db"
@@ -68,7 +68,7 @@ services:
- "traefik.http.routers.nextcloud.entrypoints=websecure"
- "traefik.http.routers.nextcloud.tls.certresolver=letsencrypt"
- "traefik.http.services.nextcloud.loadbalancer.server.port=80"
- "traefik.http.routers.nextcloud.middlewares=nextcloud-wellknown,nextcloud-headers"
- "traefik.http.routers.nextcloud.middlewares=nextcloud-wellknown,nextcloud-headers,rate-limit-lax@docker"
# Redirect .well-known CalDAV/CardDAV to the proper Nextcloud endpoint
- "traefik.http.middlewares.nextcloud-wellknown.redirectregex.regex=^https://([^/]*)/.well-known/(card|cal)dav"
- "traefik.http.middlewares.nextcloud-wellknown.redirectregex.replacement=https://$$1/remote.php/dav/"
@@ -20,7 +20,7 @@ services:
- "traefik.http.routers.registry.entrypoints=websecure"
- "traefik.http.routers.registry.tls.certresolver=letsencrypt"
- "traefik.http.services.registry.loadbalancer.server.port=5000"
- "traefik.http.routers.registry.middlewares=registry-buffering"
- "traefik.http.routers.registry.middlewares=registry-buffering,rate-limit@docker"
# Remove body size limit so large image layers can be pushed
- "traefik.http.middlewares.registry-buffering.buffering.maxRequestBodyBytes=0"
@@ -47,6 +47,7 @@ services:
- "traefik.http.routers.sso.entrypoints=websecure"
- "traefik.http.routers.sso.tls.certresolver=letsencrypt"
- "traefik.http.services.sso.loadbalancer.server.port=8080"
- "traefik.http.routers.sso.middlewares=rate-limit@docker"
healthcheck:
test: ["CMD", "bash", "-c", "exec 3<>/dev/tcp/localhost/9000"]
interval: 30s
+9
View File
@@ -1,4 +1,13 @@
---
- name: Create Traefik log directory
ansible.builtin.file:
path: /var/log/traefik
state: directory
owner: root
group: root
mode: "0755"
tags: traefik
- name: Create Traefik data directory
ansible.builtin.file:
path: "{{ traefik_data_dir }}"
@@ -11,6 +11,7 @@ services:
- /var/run/docker.sock:/var/run/docker.sock:ro
- {{ traefik_data_dir }}/traefik.yml:/traefik.yml:ro
- {{ traefik_data_dir }}/acme.json:/acme.json
- /var/log/traefik:/var/log/traefik
networks:
- traefik_public
labels:
@@ -22,6 +23,15 @@ services:
- "traefik.http.routers.dashboard.service=api@internal"
- "traefik.http.routers.dashboard.middlewares=dashboard-auth"
- "traefik.http.middlewares.dashboard-auth.basicauth.users={{ traefik_dashboard_users }}"
# Rate-limit middlewares — referenced by services as rate-limit@docker / rate-limit-lax@docker
# Standard: 60 req/min per IP, burst 20 — protects all services from bot floods
- "traefik.http.middlewares.rate-limit.rateLimit.average=60"
- "traefik.http.middlewares.rate-limit.rateLimit.period=1m"
- "traefik.http.middlewares.rate-limit.rateLimit.burst=20"
# Lax: 300 req/min per IP, burst 100 — for Nextcloud sync clients making many small requests
- "traefik.http.middlewares.rate-limit-lax.rateLimit.average=300"
- "traefik.http.middlewares.rate-limit-lax.rateLimit.period=1m"
- "traefik.http.middlewares.rate-limit-lax.rateLimit.burst=100"
networks:
traefik_public:
@@ -9,7 +9,9 @@ api:
log:
level: INFO
accessLog: {}
accessLog:
filePath: /var/log/traefik/access.log
bufferingSize: 100
entryPoints:
web: