A tool to download and update clamav databases and database patch files for the purposes of hosting your own database mirror.
Copyright (C) 2021-2022 Cisco Systems, Inc. and/or its affiliates. All rights reserved.
This tool downloads the latest ClamAV databases along with the latest database patch files.
This project replaces the clamdownloader.pl
Perl script by Frederic Vanden Poel, formerly provided here: https://www.clamav.net/documents/private-local-mirrors
Run this tool as often as you like, but it will only download new content if there is new content to download. If you somehow manage to download too frequently (eg: by using cvd clean all
and cvd update
repeatedly), then the official database server may refuse your download request, and one or more databases may go on cool-down until it's safe to try again.
- Python 3.6 or newer.
- An internet connection with DNS enabled.
- The following Python packages. These will be installed automatically if you use
pip
, but may need to be installed manually otherwise:click
v7.0 or newercoloredlogs
v10.0 or newercolorama
requests
dnspython
v2.1.0 or newerrangehttpserver
You may install cvdupdate
from PyPI using pip
, or you may clone the project Git repository and use pip
to install it locally.
Install cvdupdate
from PyPI:
python3 -m pip install --user cvdupdate
When running cvd update
to update the databases, it will also check if there is a new version of the cvdupdate
package on Python's PyPI package repository. If there is a newer version of cvdupdate
, you will see a message prompting you to upgrade. It will look someething like this:
WARNING You are running cvdupdate version: 1.1.0.
WARNING There is a newer version on PyPI: 1.1.1. Please update!
To upgrade the cvdupdate
package through PyPI, run:
python3 -m pip install --user --upgrade cvdupdate
Use the --help
option with any cvd
command to get help.
cvd --help
Tip: You may not be able to run the
cvd
orcvdupdate
shortcut directly if your Python Scripts directory is not in yourPATH
environment variable. If you run into this issue, and do not wish to add the Python Scripts directory to your path, you can run CVD-Update like this:python -m cvdupdate --help
(optional) You may wish to customize where the databases are stored:
cvd config set --dbdir <your www path>
Run this to download the latest database and associated CDIFF patch files:
cvd update
Downloaded databases will be placed in ~/.cvdupdate/database
unless you customized it to use a different directory.
Newly downloaded databases will replace the previous database version, but the CDIFF patch files will accumulate up to a configured maximum before it starts deleting old CDIFFs (default: 30 CDIFFs). You can configure it to keep more CDIFFs by manually editing the config (default: ~/.cvdupdate/config.json
). The same behavior applies for CVD-Update log rotation.
Run this to serve up the database directory on http://localhost:8000
so you can test it with FreshClam.
cvd serve
Disclaimer: The
cvd serve
feature is not intended for production use, just for testing. You probably want to use a more robust HTTP server for production work.
Install ClamAV if you don't already have it and, in another terminal window, modify your freshclam.conf
file. Replace:
DatabaseMirror database.clamav.net
... with:
DatabaseMirror http://localhost:8000
Tip: A default install on Linux/Unix places
freshclam.conf
in/usr/local/etc/freshclam.conf
. If one does not exist, you may need to create it usingfreshclam.conf.sample
as a template.
Now, run freshclam -v
or freshclam.exe -v
to see what happens. You should see FreshClam successfully update it's own database directory from your private database server.
Run cvd update
as often as you need. Maybe put it in a cron
job.
Tip: Each command supports a
--verbose
(-V
) mode, which often provides more details about what's going on under the hood.
Cron is a popular choice to automate frequent tasks on Linux / Unix systems.
-
Open a terminal running as the user which you want CVD-Update to run under, do the following:
crontab -e
-
Press
i
to insert new text, and add this line:30 */4 * * * /bin/sh -c "~/.local/bin/cvd update &> /dev/null"
Or instead of
~/
, you can do this, replacingusername
with your user name:30 */4 * * * /bin/sh -c "/home/username/.local/bin/cvd update &> /dev/null"
-
Press , then type
:wq
and press to write the file to disk and quit.
About these settings:
I selected 30 */4 * * *
to run at minute 30 past every 4th hour. CVD-Update uses a DNS check to do version checks before it attempts to download any files, just like FreshClam. Running CVD-Update more than once a day should not be an issue.
CVD-Update will write logs to the ~/.cvdupdate/logs
directory, which is why I directed stdout
and stderr
to /dev/null
instead of a log file. You can use the cvd config set
command to customize the log directory if you like, or redirect stdout
and stderr
to a log file if you prefer everything in one log instead of separate daily logs.
DNS is required for CVD-Update to function properly (to gather the TXT record containing the current definition database version). You can select a specific nameserver to ensure said nameserver is used when querying the TXT record containing the current database definition version available
-
Set the nameserver in the config. Eg:
cvd config set --nameserver 208.67.222.222
-
Set the environment variable
CVDUPDATE_NAMESERVER
. Eg:CVDUPDATE_NAMESERVER="208.67.222.222" cvd update
The environment variable will take precedence over the nameserver config setting.
Note: Both options can be used to provide a comma-delimited list of nameservers to utilize for resolution.
Depending on your type of proxy, you may be able to use CVD-Update with your proxy by running CVD-Update like this:
First, set a custom domain name server to use the proxy:
cvd config set --nameserver <proxy_ip>
Then run CVD-Update like this:
http_proxy=http://<proxy_ip>:<proxy_port> https_proxy=http://<proxy_ip>:<proxy_port> cvd update -V
Or create a script to wrap the CVD-Update call. Something like:
#!/bin/bash
http_proxy=http://<proxy_ip>:<proxy_port>
export http_proxy
https_proxy=http://<proxy_ip>:<proxy_port>
export https_proxy
cvd update -V
Disclaimer: CVD-Update doesn't support proxies that require authentication at this time. If your network admin allows it, you may be able to work around it by updating your proxy to allow HTTP requests through unauthenticated if the User-Agent matches your specific CVD-Update user agent. The CVD-Update User-Agent follows the form
CVDUPDATE/<version> (<uuid>)
where theuuid
is unique to your installation and can be found in the~/.cvdupdate/state.json
file (or~/.cvdupdate/config.json
for cvdupdate <=1.0.2). See #9 for more details.Adding support for proxy authentication is a ripe opportunity for a community contribution to the project.
This tool is to creates the following directories:
~/.cvdupdate
~/.cvdupdate/logs
~/.cvdupdate/databases
This tool creates the following files:
~/.cvdupdate/config.json
~/.cvdupdate/state.json
~/.cvdupdate/databases/<database>.cvd
~/.cvdupdate/databases/<database>-<version>.cdiff
~/.cvdupdate/logs/<date>.log
Tip: You can set custom
database
andlogs
directories with thecvd config set
command. It is likely you will want to customize thedatabase
directory to point to your HTTP server'swww
directory (or equivalent). Bare in mind that if you already downloaded the databases to the old directory, you may want to move them to the new directory.
Important: If you want to use a custom config path, you'll have to use it in every command. If you're fine with having it go in
~/.cvdupdate/config.json
, don't worry about it.
Familiarize yourself with the various commands using the --help
option.
cvd --help
cvd config --help
cvd update --help
cvd add --help
cvd clean --help
Print out the current list of databases.
cvd list -V
Print out the config to see what it looks like.
cvd config show
Do an update, use "verbose mode" to so you can get a feel for how it works.
cvd update -V
List out the databases again:
cvd list -V
The print out the config again so you can see what's changed.
cvd config show
And maybe take a peek in the database directory as well to see it for yourself.
ls ~/.cvdupdate/database
Have a look at the logs if you wish.
ls ~/.cvdupdate/logs
cat ~/.cvdupdate/logs/*
Maybe add an additional database that is not part of the default set of databases.
cvd add linux.cvd https://database.clamav.net/linux.cvd
List out the databases again:
cvd list -V
Test out your mirror with FreshClam on the same computer.
This tool includes a --serve
feature that will host the current database directory on http://localhost (default port: 8000).
You can test it by running freshclam
or freshclam.exe
locally, where you've configured freshclam.conf
with:
DatabaseMirror http://localhost:8000
Build docker image
docker build . --tag cvdupdate:latest
Run image, that will automaticly update databases in folder /srv/cvdupdate
and write logs to /var/log/cvdupdate
docker run -d \
-v /srv/cvdupdate:/cvdupdate/database \
-v /var/log/cvdupdate:/cvdupdate/logs \
cvdupdate:latest
Run image, that will automaticly update databases in folder /srv/cvdupdate
, write logs to /var/log/cvdupdate
and set owner of files to user with ID 1000
docker run -d \
-v /srv/cvdupdate:/cvdupdate/database \
-v /var/log/cvdupdate:/cvdupdate/logs \
-e USER_ID=1000 \
cvdupdate:latest
Default update interval is 30 */4 * * *
(see Cron Example)
You may pass custom update interval in environment variable CRON
For example - update every day in 00:00
docker run -d \
-v /srv/cvdupdate:/cvdupdate/database \
-v /var/log/cvdupdate:/cvdupdate/logs \
-e CRON='0 0 * * *' \
cvdupdate:latest
A Docker compose.yaml
is provided to:
- Regularly update a Docker volume with the latest ClamAV databases.
- Serve a database mirror on port 8000 using the Apache webserver.
Edit the compose.yaml
file if you need to change the default values:
- Port 8000
- USER_ID=0
- CRON=30 */4 * * *
docker compose build
docker compose up -d
docker compose down
Volumes are defined in compose.yaml
and will be auto-created when you run docker compose up
DRIVER VOLUME NAME
local cvdupdate_database
local cvdupdate_log
We'd love your help. There are many ways to contribute!
Join the ClamAV community on the ClamAV Discord chat server.
If you find an issue with CVD-Update or the CVD-Update documentation, please submit an issue to our GitHub issue tracker. Before you submit, please check to if someone else has already reported the issue.
If you find a bug and you're able to craft a fix yourself, consider submitting the fix in a pull request. Your help will be greatly appreciated.
If you want to contribute to the project and don't have anything specific in mind, please check out our issue tracker. Perhaps you'll be able to fix a bug or add a cool new feature.
By submitting a contribution to the project, you acknowledge and agree to assign Cisco Systems, Inc the copyright for the contribution. If you submit a significant contribution such as a new feature or capability or a large amount of code, you may be asked to sign a contributors license agreement comfirming that Cisco will have copyright license and patent license and that you are authorized to contribute the code.
The following steps are intended to help users that wish to contribute to development of the CVD-Update project get started.
-
Create a fork of the CVD-Update git repository, and then clone your fork to a local directory.
For example:
git clone https://github.com/<your username>/cvdupdate.git
-
Make sure CVD-Update is not already installed. If it is, remove it.
python3 -m pip uninstall cvdupdate
-
Use pip to install CVD-Update in "edit" mode.
python3 -m pip install -e --user ./cvdupdate
Once installed in "edit" mode, any changes you make to your clone of the CVD-Update code will be immediately usable simply by running the cvdupdate
/ cvd
commands.
This project has not selected a specific Code-of-Conduct document at this time. However, contributors are expected to behave in professional and respectful manner. Disrespectful or inappropriate behavior will not be tolerated.
CVD-Update is licensed under the Apache License, Version 2.0 (the "License"). You may not use the CVD-Update project except in compliance with the License.
A copy of the license is located here, and is also available online at apache.org.
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.