[BUG] Empty timeout Settings leading to failing nginx check

[OLED1]

What happened?

Since one of the last updates I am getting this errors:

Failover Message:
bunker-bunkerweb-1:5000 - nginx: configuration file /etc/nginx/nginx.conf test failed

Further checks of the scheduler container revelealed:

[2026-04-17 06:49:42 +0000] [API.CALLER] [35] [❌] - Error while sending API request to http://bunker-bunkerweb-1:5000/reload?test=yes : status = 500, msg = config check failed: nginx: [emerg] invalid number of arguments in "client_body_timeout" directive in /etc/nginx/http.conf:29
nginx: configuration file /etc/nginx/nginx.conf test failed

These are this settings which I did not override anywhere in my config:

# timeouts
client_body_timeout ;
client_header_timeout ;
keepalive_timeout ;
send_timeout ;

They are all empty and causing problems. What could be the problem?

How to reproduce?

Don't know.

Configuration file(s) (yaml or .env)

Relevant log output

BunkerWeb version

1.6.9

What integration are you using?

Docker

Linux distribution (if applicable)

Fedora release 43 (Forty Three)

Removed private data

• I have removed all private data from the configuration file and the logs

Code of Conduct

• I agree to follow this project's Code of Conduct

[TheophileDiot]

Hi @OLED1, this is maybe a Fedora 43 only issue. I'll investigate thank you

[OLED1]

@TheophileDiot I need to note that I deployed the docker stack, but the host OS is Fedora 43.

[TheophileDiot]

Hi @OLED1, thanks for the report. I can see exactly where the empty string reaches the template, but I need a little more state from your environment to confirm the origin. Could you run these on your instance and share the output?

1. DB state for the 4 timeouts (replace <bw-db with your MariaDB container name):

docker exec <bw-dbmariadb -ubunkerweb -p<your-passworddb -e \
  "SELECT setting_id, suffix, method, value, LENGTH(value) AS len \
   FROM bw_global_values \
   WHERE setting_id IN ('CLIENT_BODY_TIMEOUT','CLIENT_HEADER_TIMEOUT','KEEPALIVE_TIMEOUT','SEND_TIMEOUT');"

2. Full compose + .env (you can redact passwords, keys and hostnames — I only need to see the env keys and the keys with empty values).

3. Scheduler logs around the failure — roughly 60 lines before the [API.CALLER] error you pasted:

docker compose logs bw-scheduler --since 10m --no-color scheduler.log

4. Version history: which BunkerWeb version did you upgrade from to 1.6.9, and did the error appear immediately on the first scheduler run after the upgrade or only later?

Also, a quick yes/no: is IGNORE_REGEX_CHECK=yes set anywhere in your env, and did you ever edit these four timeout fields in the Web UI General/Misc settings page?

[OLED1]

Hi @TheophileDiot, of course. Sorry for the late response.

1.:

MariaDB [(none)]> show databases;
+--------------------+
| Database           |
+--------------------+
| db                 |
| information_schema |
+--------------------+
2 rows in set (0.000 sec)

MariaDB [(none)]> use db;
Reading table information for completion of table and column names
You can turn off this feature to get a quicker startup with -A

Database changed
MariaDB [db]> SELECT setting_id, suffix, method, value, LENGTH(value) AS len \
    ->    FROM bw_global_values \
    ->    WHERE setting_id IN ('CLIENT_BODY_TIMEOUT','CLIENT_HEADER_TIMEOUT','KEEPALIVE_TIMEOUT','SEND_TIMEOUT');
Empty set (0.000 sec)

2.: compose:

x-ui-env: &bw-ui-env
  # Wir verankern die Umgebungsvariablen, um Duplikate zu vermeiden
  AUTOCONF_MODE: "yes"
  DATABASE_URI: "mariadb+pymysql://bunkerweb:changeme@bw-db:3306/db" # Denken Sie daran, ein stärkeres Passwort für die Datenbank festzulegen

services:
  bunkerweb:
    image: bunkerity/bunkerweb:1.6.9
    ports:
      - "80:8080/tcp"
      - "443:8443/tcp"
      - "443:8443/udp" # Für QUIC / HTTP3-Unterstützung
    labels:
      - "bunkerweb.INSTANCE=yes" # Wir setzen das Instanz-Label, damit die Autoconf die Instanz erkennen kann
    environment:
      AUTOCONF_MODE: "yes"
      API_WHITELIST_IP: "127.0.0.0/8 10.20.30.0/24"
      USE_IPV6: "yes"
    restart: "unless-stopped"
    networks:
      - bw-universe
      - bw-services

  bw-scheduler:
    image: bunkerity/bunkerweb-scheduler:1.6.9
    environment:
      <<: *bw-ui-env
      BUNKERWEB_INSTANCES: ""
      SERVER_NAME: ""
      API_WHITELIST_IP: "127.0.0.0/8 10.20.30.0/24"
      MULTISITE: "yes"
      UI_HOST: "http://bw-ui:7000" # Ändern Sie dies bei Bedarf
      USE_REDIS: "yes"
      REDIS_HOST: "redis"
      USE_IPV6: "yes"
    volumes:
      - bw-storage:/data # Dies wird verwendet, um den Cache und andere Daten wie die Backups zu persistieren
    restart: "unless-stopped"
    networks:
      - bw-universe
      - bw-db

  bw-autoconf:
    image: bunkerity/bunkerweb-autoconf:1.6.9
    depends_on:
      - bw-docker
    environment:
      <<: *bw-ui-env
      DOCKER_HOST: "tcp://bw-docker:2375"
    restart: "unless-stopped"
    networks:
      - bw-universe
      - bw-docker
      - bw-db

  bw-docker:
    image: tecnativa/docker-socket-proxy:nightly
    volumes:
      - /var/run/docker.sock:/var/run/docker.sock:ro
    environment:
      CONTAINERS: "1"
      LOG_LEVEL: "warning"
    networks:
      - bw-docker

  bw-ui:
    image: bunkerity/bunkerweb-ui:1.6.9
    environment:
      <<: *bw-ui-env
      TOTP_ENCRYPTION_KEYS: "mysecret" # Denken Sie daran, einen stärkeren geheimen Schlüssel festzulegen (siehe Abschnitt Voraussetzungen)
    restart: "unless-stopped"
    networks:
      - bw-universe
      - bw-db

  bw-db:
    image: mariadb:11
    # Wir setzen die maximal zulässige Paketgröße, um Probleme mit großen Abfragen zu vermeiden
    command: --max-allowed-packet=67108864
    environment:
      MYSQL_RANDOM_ROOT_PASSWORD: "yes"
      MYSQL_DATABASE: "db"
      MYSQL_USER: "bunkerweb"
      MYSQL_PASSWORD: "changeme" # Denken Sie daran, ein stärkeres Passwort für die Datenbank festzulegen
    volumes:
      - bw-data:/var/lib/mysql
    restart: "unless-stopped"
    networks:
      - bw-db

  redis: # Redis-Dienst für die Persistenz von Berichten/Sperren/Statistiken
    image: redis:7-alpine
    command: >
      redis-server
      --maxmemory 256mb
      --maxmemory-policy allkeys-lru
      --save 60 1000
      --appendonly yes
    volumes:
      - redis-data:/data
    restart: "unless-stopped"
    networks:
      - bw-universe

#  crowdsec:
#    image: crowdsecurity/crowdsec
#    restart: always
#    ports:
#      - 127.0.0.1:8080:8080
#    environment:
#      COLLECTIONS: "crowdsecurity/nginx"
#      GID: "${GID:-1000}"
#    volumes:
#      - ./crowdsec/acquis.yaml:/etc/crowdsec/
#      - crowdsec-logs:/var/log/nginx
#      - crowdsec-db:/var/lib/crowdsec/data/
#      - crowdsec-config:/etc/crowdsec/
#    networks:
#      - bw-universe

volumes:
  bw-data:
  bw-storage:
  redis-data:
  crowdsec-logs:
  crowdsec-db:
  crowdsec-config:

networks:
  bw-universe:
    name: bw-universe
    enable_ipv6: true
    ipam:
      driver: default
      config:
        - subnet: 10.20.30.0/24
        - subnet: fd00:14:38::/48
  bw-services:
    name: bw-services
    enable_ipv6: true
    ipam:
      config:
        - subnet: fd00:13:37::/48
          gateway: fd00:13:37::1
  bw-docker:
    name: bw-docker
  bw-db:
    name: bw-db

I am having no .env File.

3. Logs:

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
[2026-04-25 14:08:03 +0000] [LETS-ENCRYPT.RENEW] [36] [ℹ️ ] - Successfully saved Let's Encrypt data to db cache
[2026-04-25 14:08:03 +0000] [SCHEDULER] [36] [ℹ️ ] - Updating /var/cache/bunkerweb permissions...
[2026-04-25 14:08:03 +0000] [SCHEDULER] [36] [ℹ️ ] - Successfully added job run for the job 'certbot-renew'
[2026-04-25 14:08:03 +0000] [SCHEDULER] [36] [ℹ️ ] - All jobs in run_once() were successful
[2026-04-25 14:08:03 +0000] [SCHEDULER] [36] [ℹ️ ] - First start: regenerating config after once-jobs to pick up files created by jobs (e.g. api-server-cert.pem)
[2026-04-25 14:08:04 +0000] [GENERATOR] [117] [ℹ️ ] - Generator started ...
[2026-04-25 14:08:04 +0000] [GENERATOR] [117] [ℹ️ ] - Settings : /usr/share/bunkerweb/settings.json
[2026-04-25 14:08:04 +0000] [GENERATOR] [117] [ℹ️ ] - Templates : /usr/share/bunkerweb/confs
[2026-04-25 14:08:04 +0000] [GENERATOR] [117] [ℹ️ ] - Core : /usr/share/bunkerweb/core
[2026-04-25 14:08:04 +0000] [GENERATOR] [117] [ℹ️ ] - Plugins : /etc/bunkerweb/plugins
[2026-04-25 14:08:04 +0000] [GENERATOR] [117] [ℹ️ ] - Pro plugins : /etc/bunkerweb/pro/plugins
[2026-04-25 14:08:04 +0000] [GENERATOR] [117] [ℹ️ ] - Output : /etc/nginx
[2026-04-25 14:08:04 +0000] [GENERATOR] [117] [ℹ️ ] - Target : /etc/nginx
[2026-04-25 14:08:04 +0000] [GENERATOR] [117] [ℹ️ ] - ✅ Database connection established
[2026-04-25 14:08:05 +0000] [GENERATOR] [117] [ℹ️ ] - Removing old files ...
[2026-04-25 14:08:05 +0000] [GENERATOR] [117] [ℹ️ ] - Rendering templates ...
[2026-04-25 14:08:05 +0000] [GENERATOR] [117] [ℹ️ ] - Generator successfully executed !
[2026-04-25 14:08:05 +0000] [SCHEDULER] [36] [ℹ️ ] - Sending /etc/nginx to all reachable BunkerWeb instances ...
[2026-04-25 14:08:05 +0000] [SCHEDULER] [36] [ℹ️ ] - Sending /var/cache/bunkerweb to all reachable BunkerWeb instances ...
[2026-04-25 14:08:06 +0000] [API.CALLER] [36] [ℹ️ ] - Successfully sent API request to http://bunker-bunkerweb-1:5000/confs
[2026-04-25 14:08:06 +0000] [SCHEDULER] [36] [ℹ️ ] - Successfully sent /etc/nginx folder to reachable BunkerWeb instances
[2026-04-25 14:08:07 +0000] [API.CALLER] [36] [ℹ️ ] - Successfully sent API request to http://bunker-bunkerweb-1:5000/cache
[2026-04-25 14:08:07 +0000] [SCHEDULER] [36] [ℹ️ ] - Successfully sent /var/cache/bunkerweb folder to reachable BunkerWeb instances
[2026-04-25 14:08:07 +0000] [API.CALLER] [36] [❌] - Error while sending API request to http://bunker-bunkerweb-1:5000/reload?test=yes : status = 500, msg = config check failed: nginx: [emerg] invalid number of arguments in "client_body_timeout" directive in /etc/nginx/http.conf:29
nginx: configuration file /etc/nginx/nginx.conf test failed

[2026-04-25 14:08:07 +0000] [SCHEDULER] [36] [❌] - Error while reloading bunkerweb, failing over to last working configuration ...
[2026-04-25 14:08:07 +0000] [SCHEDULER] [36] [ℹ️ ] - Sending /var/tmp/bunkerweb/failover/config to all reachable BunkerWeb instances ...
[2026-04-25 14:08:07 +0000] [SCHEDULER] [36] [ℹ️ ] - Sending /var/tmp/bunkerweb/failover/custom_configs to all reachable BunkerWeb instances ...
[2026-04-25 14:08:07 +0000] [SCHEDULER] [36] [ℹ️ ] - Sending /var/tmp/bunkerweb/failover/cache to all reachable BunkerWeb instances ...
[2026-04-25 14:08:07 +0000] [API.CALLER] [36] [ℹ️ ] - Successfully sent API request to http://bunker-bunkerweb-1:5000/custom_configs
[2026-04-25 14:08:07 +0000] [SCHEDULER] [36] [ℹ️ ] - Successfully sent /var/tmp/bunkerweb/failover/custom_configs folder to reachable BunkerWeb instances
[2026-04-25 14:08:07 +0000] [API.CALLER] [36] [ℹ️ ] - Successfully sent API request to http://bunker-bunkerweb-1:5000/confs
[2026-04-25 14:08:07 +0000] [SCHEDULER] [36] [ℹ️ ] - Successfully sent /var/tmp/bunkerweb/failover/config folder to reachable BunkerWeb instances
[2026-04-25 14:08:08 +0000] [API.CALLER] [36] [ℹ️ ] - Successfully sent API request to http://bunker-bunkerweb-1:5000/cache
[2026-04-25 14:08:08 +0000] [SCHEDULER] [36] [ℹ️ ] - Successfully sent /var/tmp/bunkerweb/failover/cache folder to reachable BunkerWeb instances
[2026-04-25 14:08:08 +0000] [API.CALLER] [36] [ℹ️ ] - Successfully sent API request to http://bunker-bunkerweb-1:5000/reload?test=yes
[2026-04-25 14:08:08 +0000] [SCHEDULER] [36] [ℹ️ ] - Executing job scheduler ...
[2026-04-25 15:08:04 +0000] [SCHEDULER] [36] [ℹ️ ] - Executing job 'blacklist-download' from plugin 'blacklist'...
[2026-04-25 15:08:04 +0000] [BLACKLIST] [36] [⚠️ ] - Community blacklist ip:laurent-minne-fr-be-aggressive not found in predefined lists
[2026-04-25 15:08:04 +0000] [BLACKLIST] [36] [ℹ️ ] - ✅ Database connection established
[2026-04-25 15:08:04 +0000] [BLACKLIST] [36] [ℹ️ ] - Downloading blacklist data from https://www.dan.me.uk/torlist/?exit ...
[2026-04-25 15:08:04 +0000] [BLACKLIST] [36] [ℹ️ ] - Downloading blacklist data from https://raw.githubusercontent.com/mitchellkrogza/nginx-ultimate-bad-bot-blocker/master/_generator_lists/bad-user-agents.list ...
[2026-04-25 15:08:04 +0000] [BLACKLIST] [36] [ℹ️ ] - Downloading blacklist data from https://raw.githubusercontent.com/duggytuxy/Data-Shield_IPv4_Blocklist/refs/heads/main/prod_data-shield_ipv4_blocklist.txt ...
[2026-04-25 15:08:18 +0000] [BLACKLIST] [36] [ℹ️ ] - Recap for IP urls: Total Services: 13, Successful: 2, Skipped (cached): 0, Failed: 0, Total Lines: 86233
[2026-04-25 15:08:18 +0000] [BLACKLIST] [36] [ℹ️ ] - Recap for USER_AGENT urls: Total Services: 13, Successful: 1, Skipped (cached): 0, Failed: 0, Total Lines: 683
[2026-04-25 15:08:18 +0000] [SCHEDULER] [36] [ℹ️ ] - Executing job 'bunkernet-register' from plugin 'bunkernet'...
[2026-04-25 15:08:18 +0000] [SCHEDULER] [36] [ℹ️ ] - Successfully added job run for the job 'blacklist-download'
[2026-04-25 15:08:18 +0000] [BUNKERNET.REGISTER] [36] [ℹ️ ] - ✅ Database connection established
[2026-04-25 15:08:18 +0000] [BUNKERNET.REGISTER] [36] [ℹ️ ] - Already registered on BunkerNet API with instance id 5ee4b1e2-19f6-49d2-abce-103c36c52e2b
[2026-04-25 15:08:18 +0000] [SCHEDULER] [36] [ℹ️ ] - Executing job 'bunkernet-send' from plugin 'bunkernet'...
[2026-04-25 15:08:18 +0000] [SCHEDULER] [36] [ℹ️ ] - Successfully added job run for the job 'bunkernet-register'
[2026-04-25 15:08:18 +0000] [BUNKERNET.SEND] [36] [ℹ️ ] - ✅ Database connection established
[2026-04-25 15:08:18 +0000] [BUNKERNET.SEND] [36] [ℹ️ ] - Successfully retrieved BunkerNet ID from db cache
[2026-04-25 15:08:18 +0000] [API.CALLER] [36] [ℹ️ ] - Successfully sent API request to http://bunker-bunkerweb-1:5000/bunkernet/reports
[2026-04-25 15:08:18 +0000] [BUNKERNET.SEND] [36] [ℹ️ ] - Caching 17 reports...
[2026-04-25 15:08:18 +0000] [SCHEDULER] [36] [ℹ️ ] - Executing job 'dnsbl-download' from plugin 'dnsbl'...
[2026-04-25 15:08:18 +0000] [SCHEDULER] [36] [ℹ️ ] - Successfully added job run for the job 'bunkernet-send'
[2026-04-25 15:08:18 +0000] [DNSBL] [36] [ℹ️ ] - ✅ Database connection established
[2026-04-25 15:08:18 +0000] [DNSBL] [36] [⚠️ ] - No DNSBL ignore IP URL configured, nothing to do...
[2026-04-25 15:08:18 +0000] [SCHEDULER] [36] [ℹ️ ] - Executing job 'greylist-download' from plugin 'greylist'...
[2026-04-25 15:08:18 +0000] [GREYLIST] [36] [ℹ️ ] - Greylist is not activated, skipping downloads...
[2026-04-25 15:08:18 +0000] [SCHEDULER] [36] [ℹ️ ] - Executing job 'realip-download' from plugin 'realip'...
[2026-04-25 15:08:18 +0000] [SCHEDULER] [36] [ℹ️ ] - Successfully added job run for the job 'dnsbl-download'
[2026-04-25 15:08:18 +0000] [SCHEDULER] [36] [ℹ️ ] - Successfully added job run for the job 'greylist-download'
[2026-04-25 15:08:18 +0000] [REALIP] [36] [ℹ️ ] - ✅ Database connection established
[2026-04-25 15:08:18 +0000] [REALIP] [36] [⚠️ ] - No URL configured, nothing to do...
[2026-04-25 15:08:18 +0000] [SCHEDULER] [36] [ℹ️ ] - Executing job 'robots-txt-download' from plugin 'robotstxt'...
[2026-04-25 15:08:18 +0000] [SCHEDULER] [36] [ℹ️ ] - Successfully added job run for the job 'realip-download'
[2026-04-25 15:08:19 +0000] [ROBOTSTXT.DOWNLOAD] [36] [ℹ️ ] - ✅ Database connection established
[2026-04-25 15:08:19 +0000] [ROBOTSTXT.DOWNLOAD] [36] [⚠️ ] - No robots.txt URL is configured, nothing to do...
[2026-04-25 15:08:19 +0000] [SCHEDULER] [36] [ℹ️ ] - Executing job 'robots-txt-darkvisitors' from plugin 'robotstxt'...
[2026-04-25 15:08:19 +0000] [ROBOTSTXT.DARKVISITORS] [36] [ℹ️ ] - Robots.txt with DarkVisitors is not activated, skipping downloads...
[2026-04-25 15:08:19 +0000] [SCHEDULER] [36] [ℹ️ ] - Executing job 'whitelist-download' from plugin 'whitelist'...
[2026-04-25 15:08:19 +0000] [SCHEDULER] [36] [ℹ️ ] - Successfully added job run for the job 'robots-txt-download'
[2026-04-25 15:08:19 +0000] [SCHEDULER] [36] [ℹ️ ] - Successfully added job run for the job 'robots-txt-darkvisitors'
[2026-04-25 15:08:19 +0000] [WHITELIST] [36] [ℹ️ ] - ✅ Database connection established
[2026-04-25 15:08:19 +0000] [WHITELIST] [36] [⚠️ ] - No whitelist URL is configured, nothing to do...
[2026-04-25 15:08:19 +0000] [SCHEDULER] [36] [ℹ️ ] - Successfully added job run for the job 'whitelist-download'

4. Good querstion I can't answer because it was some weeks ago I updated the instance. Is there a commmand where I can check this?

grep "IGNORE_REGEX_CHECK" bunker-compose.yaml returned nothing.
I don't have a .env.

Sorry for the wall of text. Github uploads are recently failing for whatever reason ...

[OLED1]

@TheophileDiot We can also do a remote debugging if you like.

[TheophileDiot]

Thanks @OLED1, that pins down half of it.

The empty bw_global_values set is expected — it just confirms you never overrode the timeouts in the UI. The actual default lives in a separate table (bw_settings), and that's the one I need to see to know which of two upgrade-time bugs you hit. Could you run the following against the same DB?

SELECT id, plugin_id, `default`, LENGTH(`default`) AS len
FROM bw_settings
WHERE id IN ('CLIENT_BODY_TIMEOUT','CLIENT_HEADER_TIMEOUT','KEEPALIVE_TIMEOUT','SEND_TIMEOUT');

SELECT id, value FROM bw_metadata WHERE id IN ('version','integration');

Three things the first query can show, and what each means:

• Empty set → the rows were never inserted on the upgrade. The fix has to land in init_tables to make sure every settings.json key produces a row, and we'll also need a one-shot backfill for already-broken installs.
• Rows present but default is NULL or '' → the rows were inserted with the wrong payload. We'll backfill via migration and patch the change-detection branch in init_tables.
• Rows present with '10s'/'15s' → the catalogue itself is healthy and the empty render comes from somewhere downstream (env, plugin override, stale Jinja bytecode cache). We'll need a fresh trace.

The bw_metadata query just helps me confirm which version you upgraded from; you mentioned it was a few weeks ago and the migration window matters for narrowing the bug.