UDP scans very slow

Scanning / scan configuration
1

Hello,

I’m testing openvas as an alternative to our Nessus vulnerability scanner, so i have created all our subnets as targets and configured scans for each of them. total is 88 scans.

I automated launching of the scans to run 1 at a time, first few scans take approximately 2-3 hours, then they start becoming slower, till eventually each scan starts taking a day or more. i don’t see any errors in the logs, server has enough resources (load average: 0.10) free mem 30Gi, no swap used.

scan configs are “full and fast”, hosts alive test is set to “TCP-ACK Service Ping” port list is set to “all iana assigned tcp and udp”

I don’t believe this is normal behavior, can anyone shed some light on how to improve performance?

below is my docker-compose:

$ cat docker-compose.yml
name: greenbone-community-edition

services:
  vulnerability-tests:
    image: registry.community.greenbone.net/community/vulnerability-tests
    environment:
      FEED_RELEASE: "24.10"
    volumes:
      - vt_data_vol:/mnt

  notus-data:
    image: registry.community.greenbone.net/community/notus-data
    volumes:
      - notus_data_vol:/mnt

  scap-data:
    image: registry.community.greenbone.net/community/scap-data
    volumes:
      - scap_data_vol:/mnt

  cert-bund-data:
    image: registry.community.greenbone.net/community/cert-bund-data
    volumes:
      - cert_data_vol:/mnt

  dfn-cert-data:
    image: registry.community.greenbone.net/community/dfn-cert-data
    volumes:
      - cert_data_vol:/mnt
    depends_on:
      - cert-bund-data

  data-objects:
    image: registry.community.greenbone.net/community/data-objects
    environment:
      FEED_RELEASE: "24.10"
    volumes:
      - data_objects_vol:/mnt

  report-formats:
    image: registry.community.greenbone.net/community/report-formats
    environment:
      FEED_RELEASE: "24.10"
    volumes:
      - data_objects_vol:/mnt
    depends_on:
      - data-objects

  gpg-data:
    image: registry.community.greenbone.net/community/gpg-data
    volumes:
      - gpg_data_vol:/mnt

  redis-server:
    image: registry.community.greenbone.net/community/redis-server
    restart: on-failure
    volumes:
      - redis_socket_vol:/run/redis/

  pg-gvm:
    image: registry.community.greenbone.net/community/pg-gvm:stable
    restart: on-failure
    volumes:
      - psql_data_vol:/var/lib/postgresql
      - psql_socket_vol:/var/run/postgresql

  gvmd:
    image: registry.community.greenbone.net/community/gvmd:stable
    restart: on-failure
    volumes:
      - gvmd_data_vol:/var/lib/gvm
      - scap_data_vol:/var/lib/gvm/scap-data/
      - cert_data_vol:/var/lib/gvm/cert-data
      - data_objects_vol:/var/lib/gvm/data-objects/gvmd
      - vt_data_vol:/var/lib/openvas/plugins
      - psql_data_vol:/var/lib/postgresql
      - gvmd_socket_vol:/run/gvmd
      - ospd_openvas_socket_vol:/run/ospd
      - psql_socket_vol:/var/run/postgresql
    depends_on:
      pg-gvm:
        condition: service_started
      scap-data:
        condition: service_completed_successfully
      cert-bund-data:
        condition: service_completed_successfully
      dfn-cert-data:
        condition: service_completed_successfully
      data-objects:
        condition: service_completed_successfully
      report-formats:
        condition: service_completed_successfully

  gsa:
    image: registry.community.greenbone.net/community/gsa:stable
    restart: on-failure
    ports:
      - 9392:9392
    volumes:
      - gvmd_socket_vol:/run/gvmd
    depends_on:
      - gvmd
  # Sets log level of openvas to the set LOG_LEVEL within the env
  # and changes log output to /var/log/openvas instead /var/log/gvm
  # to reduce likelyhood of unwanted log interferences
  configure-openvas:
    image: registry.community.greenbone.net/community/openvas-scanner:stable
    volumes:
      - openvas_data_vol:/mnt
      - openvas_log_data_vol:/var/log/openvas
    command:
      - /bin/sh
      - -c
      - |
        printf "table_driven_lsc = yes\nopenvasd_server = http://openvasd:80\n" > /mnt/openvas.conf
        sed "s/127/128/" /etc/openvas/openvas_log.conf | sed 's/gvm/openvas/' > /mnt/openvas_log.conf
        chmod 644 /mnt/openvas.conf
        chmod 644 /mnt/openvas_log.conf
        touch /var/log/openvas/openvas.log
        chmod 666 /var/log/openvas/openvas.log

  # shows logs of openvas
  openvas:
    image: registry.community.greenbone.net/community/openvas-scanner:stable
    restart: on-failure
    volumes:
      - openvas_data_vol:/etc/openvas
      - openvas_log_data_vol:/var/log/openvas
    command:
      - /bin/sh
      - -c
      - |
        cat /etc/openvas/openvas.conf
        tail -f /var/log/openvas/openvas.log
    depends_on:
      configure-openvas:
        condition: service_completed_successfully

  openvasd:
    image: registry.community.greenbone.net/community/openvas-scanner:stable
    restart: on-failure
    environment:
      # `service_notus` is set to disable everything but notus,
      # if you want to utilize openvasd directly, remove `OPENVASD_MODE`
      OPENVASD_MODE: service_notus
      GNUPGHOME: /etc/openvas/gnupg
      LISTENING: 0.0.0.0:80
    volumes:
      - openvas_data_vol:/etc/openvas
      - openvas_log_data_vol:/var/log/openvas
      - gpg_data_vol:/etc/openvas/gnupg
      - notus_data_vol:/var/lib/notus
    # enable port forwarding when you want to use the http api from your host machine
    # ports:
    #   - 127.0.0.1:3000:80
    depends_on:
      vulnerability-tests:
        condition: service_completed_successfully
      configure-openvas:
        condition: service_completed_successfully
      gpg-data:
        condition: service_completed_successfully
    networks:
      default:
        aliases:
          - openvasd

  ospd-openvas:
    image: registry.community.greenbone.net/community/ospd-openvas:stable
    restart: on-failure
    hostname: ospd-openvas.local
    cap_add:
      - NET_ADMIN # for capturing packages in promiscuous mode
      - NET_RAW # for raw sockets e.g. used for the boreas alive detection
    security_opt:
      - seccomp=unconfined
      - apparmor=unconfined
    command:
      [
        "ospd-openvas",
        "-f",
        "--config",
        "/etc/gvm/ospd-openvas.conf",
        "--notus-feed-dir",
        "/var/lib/notus/advisories",
        "-m",
        "666",
      ]
    volumes:
      - gpg_data_vol:/etc/openvas/gnupg
      - vt_data_vol:/var/lib/openvas/plugins
      - notus_data_vol:/var/lib/notus
      - ospd_openvas_socket_vol:/run/ospd
      - redis_socket_vol:/run/redis/
      - openvas_data_vol:/etc/openvas/
      - openvas_log_data_vol:/var/log/openvas
    depends_on:
      redis-server:
        condition: service_started
      gpg-data:
        condition: service_completed_successfully
      vulnerability-tests:
        condition: service_completed_successfully
      configure-openvas:
        condition: service_completed_successfully

  gvm-tools:
    image: registry.community.greenbone.net/community/gvm-tools
    volumes:
      - gvmd_socket_vol:/run/gvmd
      - ospd_openvas_socket_vol:/run/ospd
    depends_on:
      - gvmd
      - ospd-openvas

volumes:
  gpg_data_vol:
  scap_data_vol:
  cert_data_vol:
  data_objects_vol:
  gvmd_data_vol:
  psql_data_vol:
  vt_data_vol:
  notus_data_vol:
  psql_socket_vol:
  gvmd_socket_vol:
  ospd_openvas_socket_vol:
  redis_socket_vol:
  openvas_data_vol:
  openvas_log_data_vol:
2

@Mbannout OpenVAS’s UDP scans are slow. It’s best to scan only TCP ports and maybe at most the top 50 UDP ports.

Eero

3

If you have many UDP ports, try to improve the timeout. UDP ports with long timeout will delay scans. Be sure no firewall is between the target and the scanner.

4

Fixed sentence, the tool is irrelevant and UDP scans are generally slow independent on which tool is used.

Further existing reading:

1 Like
5

@Mbannout Each org has a list of accepted UDP ports, you can list add them to your scanner. and if you are unaware of those ports you can perform port scanner on the targeted VLANs and configure your scanner to focus them.

P.S. UDP scan is always slow no matter which scanner you are using.

6

Thank you all for the replies, my problem is not regular slowness. the first 10-15 scans are somewhat fast, but after some time the scans stop completely or become extremely slow. i tweaked the ulimits and memory overcommit and started my testing again. so far, it’s going fine. i’ll update again when the scans complete.