Need help with hostsblock?
Click the “chat” button below for chat support from the developer who created it, or find similar developers for support.

About the developer

204 Stars 30 Forks 211 Commits 15 Opened issues


an ad- and malware-blocking script for Linux

Services available


Need anything else?

Contributors list


An ad- and malware-blocking utility for POSIX systems


  1. Description: Features
  2. Installation: Dependencies, Arch Linux, Other POSIX
  3. Configuration: Edit
    , Enable Timer, Enable Postprocessing
  4. Usage: Configuring
    , Manual Usage, UrlCheck Usage (examples)
  5. FAQ
  6. News & Bugs: Upgrading to 0.999.8
  7. License


Hostsblock is a POSIX-compatible script designed to take advantage of the

file to provide system-wide blocking of internet advertisements, malicious domains, trackers, and other undesirable content.

To do so, it downloads a configurable set of blocklists and processes their entries into a single HOSTS file.

Hostsblock also provides a command-line utility that allows you to configure how individual websites and any other domains contained in that website are handled.


  • Enhanced security - Runs as an unprivileged user instead of root. New: Includes systemd service files that heavily sandbox the background process.

  • System-wide blocking - All non-proxied connections use the HOSTS file (Proxied connections can be modified to use the HOSTS file)

  • Compression-friendly - Can download and process zip- and 7zip-compressed files automatically. (Provided that

    are installed)
  • Non-interactive - Can be run as a periodic background job without needing user interaction.

  • Extensive configurability - Allows for custom deny & allow listing, redirection, ~~post-processing scripting~~ (now provided via systemd configuration), etc.

  • Bandwith-efficient - Only downloads blocklists that have been changed, using http compression when available.

  • Resource-efficient - Only processes blocklists when changes are registered.

  • High performance blocking - Only when using dns caching.

  • Redirection capability - Enchances security by combating DNS cache poisoning.

  • Extensive choice of blocklists included - Allowing the user to choose how much or how little is blocked/redirected.



  • curl
  • A POSIX environment (which should already be in place on most Linux, *BSD, and macOS environments, including the following commands:
    (e.g. bash or dash,
    , and

Optional dependencies for additional features

  • sudo to enable the user-friendly wrapper script (highly recommended)

Unarchivers to use archive blocklists instead of plain text:

  • unzip (for zip archives)
  • p7zip (for 7z archives) must include either
    , or

A DNS caching daemon to help speed up DNS resolutions:

If you use as your blocking redirect address (

), a pseudo-server that serves blank pages to remove boilerplate page and speed up page resolution on blocked domains:

Note that the default configuration gets no benefit from having a pseudo-server

Arch Linux

If you have yaourt installed:

yaourt -S hostsblock
yaourt -S hostsblock-git

Or use one of the AUR packages: hostsblock, hostsblock-git

Don't forget to enable and start the systemd timer by running this:

$ sudo systemctl enable --now hostsblock.timer

For Other POSIX Flavors and Distros

The Best and Easiest Way

Please check with your distribution to see if a package is available. If there is not, ask for it or contribute your own!

If you are a package maintainer, let me know so that I can post the instructions here.

The Easy Way

First download the archive here or with curl like so:

curl -O ""

Unzip the archive, e.g.


Execute the
script as root, which will guide you through installation.


By default, the configuration files are included in the

directory. Copy them over to
to customize your setup.


Most of the hostsblock configuration is done in the

. This file is commented really well, so please read through it before first use:

# CACHE DIRECTORY. Directory where blocklists will be downloaded and stored.

#cachedir="$HOME/cache" # DEFAULT

WORK DIRECTORY. Temporary directory where interim files will be unzipped and

# processed. This directory will be deleted after hostsblock completes.

#tmpdir="/tmp/hostsblock" # DEFAULT

FINAL HOSTSFILE. Final hosts file that combines together all downloaded blocklists.

#hostsfile="$HOME/hosts.block" # DEFAULT

REDIRECT URL. IP address to which blocked hosts will be redirect, either or This replaces any entries to and If you run a

pixelserver such as pixelserv or kwakd, it is advisable to use

#redirecturl="" # DEFAULT

HEAD FILE. File containing hosts file entries which you want at the beginning

of the resultant hosts file, e.g. for loopback devices and IPv6 entries. Use

your original /etc/hosts file here if you are writing your final blocklist to

/etc/hosts so as to preserve your loopback devices. Give hostshead="0" to

disable this feature. For those targeting /etc/hosts, it is advisable to copy

their old /etc/hosts file to this file so as to preserve existing entries.

#hostshead="0" # DEFAULT

DENYLISTED SUBDOMAINS. File containing specific subdomains to denylist which

may not be in the downloaded denylists. Be sure to provide not just the

domain, e.g. "", but also the specific subdomain a la

"" without quotations.

#denylist="$HOME/deny.list" # DEFAULT

ALLOWLIST. File containing the specific subdomains to allow through that may

be blocked by the downloaded blocklists. In this file, put a space in front of

a string in order to let through that specific site (without quotations), e.g.

"" will unblock "" but not

"". Leave no space in front of the entry to

unblock all subdomains that contain that string, e.g. "" will let

through "", "", "",

"", etc.


CONNECT_TIMEOUT. Parameter passed to curl. Determines how long to try to

connect to each blocklist url before giving up.

#connect_timeout=60 # DEFAULT

RETRY. Parameter passed to curl. Number of times to retry connecting to

each blocklist url before giving up.

#retry=0 # DEFAULT

MAX SIMULTANEOUS DOWNLOADS. Hostsblock can check and download files in parallel.

By default, it will attempt to check and download four files at a time.

#max_simultaneous_downloads=4 # DEFAULT

BLOCKLISTS FILE. File containing URLs of blocklists to be downloaded,

each on a separate line. Downloaded files may be either

plaintext, zip, or 7z files. Hostsblock will automatically

identify the file type.


REDIRECTLISTS FILE. File containing URLs of redirectlists to be downloaded,

each on a separate line. Downloaded files may be either

plaintext, zip, or 7z files. Hostsblock will automatically

identify the file type.

#redirectlists="" # DEFAULT, otherwise "$HOME/redirect.urls"

If you have any additional lists, please post a bug report to

Enable the systemd service

Don't forget to enable and start the systemd timer with:

$ sudo systemctl enable --now hostsblock.timer

Configure Postprocessing

Hostsblock does not write to

or manipulate any DNS caching daemons anymore. Instead, it will just compile a hosts-formatted file to
. To make this file actually do work, you have one of two options:

OPTION 1: Using a DNS Caching Daemon (Here: dnsmasq)

Using a DNS caching daemon like dnsmasq offers better performance.

To use hostsblock together with dnsmasq, configure dnsmasq as DNS caching daemon. Please refer to your distribution's manual. For ArchLinux read the following: Wiki section.

After that, add the following line to

(usually under
) so that dnsmasq will reference the file:

Enable and start

$ sudo systemctl enable --now hostsblock-dnsmasq-restart.path

This has systemd watch the target file

for changes and then restart
whenever they are found.

OPTION 2: Copy /var/lib/hostsblock/hosts.block to /etc/hosts

It is possible to have

with the generated file.


to make sure you don't remove the default system loopback address(es), e.g.:

Then put your necessary loopback entries in

. For example, you can copy over your existing
to this file:
$ sudo cp /etc/hosts /var/lib/hostsblock/hosts.head
$ sudo chown hostsblock:hostsblock /var/lib/hostsblock/hosts.head
$ sudo chmod 600 /var/lib/hostsblock/hosts.head

Enable and start

$ sudo systemctl enable --now hostsblock-hosts-clobber.path

This has systemd watch the target file

for changes and then copy


In its normal systemd-job configuration, hostsblock requires no interaction from the user aside from the steps above. If, however, you want to manually run the process, or to use the UrlCheck tool (

hostsblock -c URL
), you need to configure


Because hostsblock executes as a heavily sandboxed unpriviledged user (instead of root), you must configure

to allow other users to manually execute it.

To do so, edit

by typing
sudo visudo
and add the following line to the end:
%hostsblock ALL =   (hostsblock)    NOPASSWD:   /usr/lib/

Add any users you want to be able to manually execute or use the urlcheck mode to the

$ sudo gpasswd -a [MY USER NAME] hostsblock

The wrapper script installed in your PATH will automatically use sudo to execute the main script as the user


hostsblock [OPTION...] - download and combine HOSTS files

Without the

-c URL
option, hostsblock will check to see if its monitored blocklists have changed. If it detects changes (or if forced by the
flag), it will download the changed blocklist(s) and recompile the target HOSTS file.
Help Options:
  -h                            Show help options

Options: -f CONFIGFILE Specify an alternative configuration file -q Show only fatal errors -v Be verbose -d Be very verbose/debug -u Force hostsblock to update its target file

hostsblock [OPTION...] -c URL [COMMANDS...] - Manage how URL is handled

With the

-c URL
flag option, hostsblock can check and manipulate how it handles specific domains.

Note: The

symlink is now officially depreciated. Use
hostsblock -c

In addition to the above options, the following commands and subcommands can be used with

hostsblock -c URL
hostsblock -c URL (urlCheck) Commands:
  -s [-r -k]            State how hostblock modifies URL
  -b [-o -r]            Temporarily (un)block URL
  -e [-o -r -b]         Add/remove URL to/from denylist
  -a [-o -r -b]         Add/remove URL to/from allowlist
  -i [-o -r -k]         Interactively inspect URL

hostsblock -c URL Command Subcommands: -r COMMAND recurses to all domains on URL's page -k COMMAND recurses for all BLOCKED domains on page -o Perform opposite of COMMAND (e.g UNblock) -b With "-e", immediately block URL With "-a", immediately unblock URL

Note that the

subcommand turns a command into its opposite, e.g.
  • hostsblock -c URL -b -o
    unblocks URL
  • hostsblock -c URL -e -o
    removes URL from the denylist
  • hostsblock -c URL -a -o
    removes URL from the allowlist


Once you have configured

, you can execute the following as any user in the
See if "" is blocked, denylisted, allowlisted, or redirected by hostsblock:
$ hostsblock -c "" -s
Do the same thing for any of the sites referenced on this page:
$ hostsblock -c "" -s -r
Do the same thing for any of the sites referenced on this page that are presently blocked:
$ hostsblock -c "" -s -k
Block the domain containing "" (that is, ""):
$ hostsblock -c "" -b

Note that "blocking" (and "unblocking", i.e.

-b -o
) a domain only works until the next time hostsblock refreshes
, unless you use a blocklist that does include it. To permanently block this domain, use the denylist (
) command.
Permanently block (denylist) the domain containing "" (that is, ""):
$ hostsblock -c "" -e

Note that "denylisting" on its own will not block the target domain until hostblock refreshes. You can combine both "blocking" and "denylisting" in one command, however:

Permanently and immediately block the domain containing "" (that is, ""):
$ hostsblock -c "" -e -b
Temporarily unblock all blocked domains on "" (helpful if the page isn't working quite right):
$ hostsblock -c "" -b -o -k
Interactively scan through "", prompting you if you want the domains referenced therein to be blocked, denylisted, or allowlisted
$ hostsblock -c "" -i -r


  • Why isn't it working with Chrome/Chromium?

    • Because they bypass the system's DNS settings and use their own.

    To force them to use the system's DNS settings, refer to this question.

  • Hostsblock's systemd job fails with error "FAILED TO COMPILE BLOCK/REDIRECT ENTRIES FROM [...]" and leaves an empty
    • You may have a blank line with a single space in your allowlist. Hostsblock matches that line with the space in between the IP address and the domain name that every single line has, i.e. it matches every single would-be entry in your target file. Remove the empty line, and hostsblock will function as expected.

News & Bugs

Upgrading to 0.999.8

For existing hostsblock users, please note the following changes in version 0.999.8:

Changes in

Due to the shift to POSIX-shell compatibility, the list of blocklists to be downloaded cannot be held in

via the
parameter. Instead, this parameter contains the path to a file that contains the list of URLs, e.g.

The new

file is simply a newline separated list of URLs without quotations. Whitespace and text after # are ignored. An example
file could look like this: # General blocking meta-list

See the example

in the
directory for details.

No more postprocessing within script

Due to enhanced security and sandboxing, hostsblock no longer handles postprocessing on its own. Instead, users should use other systemd capabilities to replace the

postprocess() {}

Hostsblock comes with systemd service files that replicate the most common scenarios. See the directions above for instructions on how to enable them.

Changes with

is no longer as widely used as before. The main systemd service no longer requires it. You only need it if you want to use the
hostsblock -c URL
(urlcheck) utility. See the above directions for details.

Other Caveats

  • The
    symlink is depreciated. Please use
    hostsblock -c URL
  • In UrlCheck mode, large hosts files will generate large temporary cache files that will eat up a lot of temporary storage. If you have a machine with little RAM (<6GB) and want to block a lot of domains, consider changing your $tmpdir to an HDD- or SSD-backed filesystem instead of the default tmpfs under
  • UrlCheck mode will not be able to provide information on which blocklist blocked which domains anymore (annotation feature removed)
  • Hostsblock uses as default redirection IP address instead of theoretically offers better performance without the need of a pseudo-server.

Other Changes from 0.999.7 to 0.999.8

Systemd Job Improvements
  • Systemd service now heavily hardened and sandboxed for enhanced security
  • Fixed simultaneous download feature so that it actually does what it is supposed to
  • Added processing support for source blocklists that just list domain names to be blocked, e.g.
    instead of
  • Added support to read directly from zip and 7z files containing a single file without decompressing to a cache
  • Optimized filters used to process domains with improved throughput
  • If run with dash instead of bash, hostsblock has significant performance improvements
  • Removed annotation feature to reduce dependencies and overall processing demands
  • Vastly expanded list of potential blocklists (see
POSIX-Compatibility Improvements
  • Supports POSIX shells (dash, ash, zsh) instead of just bash
  • Removed GNU-specific utilities, relies only on POSIX options
  • Should now run on *BSD and macOS (and perhaps even Android and iOS!) if proper POSIX environments are installed. UNTESTED
UrlCheck Mode Improvements


Hostsblock is licensed under GNU GPL

We use cookies. If you continue to browse the site, you agree to the use of cookies. For more information on our use of cookies please see our Privacy Policy.