Green Sky
a3126d581b
82460b212 feat: add ngc events 24b54722a fix: Ensure we have allocators available for the error paths. 48dbcfebc cleanup: Remove redundant `-DSODIUM_EXPORT` from definitions. 0cef46ee9 cleanup: Fix a few more clang-tidy warnings. 0c5b918e9 cleanup: Fix a few more clang-tidy warnings. 4d3c97f49 cleanup: Enforce stricter identifier naming using clang-tidy. a549807df refactor: Add `mem` module to allow tests to override allocators. 6133fb153 chore: Add devcontainer setup for codespaces. 620e07ecd chore: Set a timeout for tests started using Conan c0ec33b16 chore: Migrate Windows CI from Appveyor to Azure DevOps 8ed47f3ef fix incorrect documentation a1e245841 docs: Fix doxygen config and remove some redundant comments. b0f633185 chore: Fix the Android CI job 7469a529b fix: Add missing `#include <array>`. 2b1a6b0d2 add missing ngc constants getter declarations and definitions 2e02d5637 chore: Add missing module dependencies. REVERT: 67badf694 feat: add ngc events git-subtree-dir: external/toxcore/c-toxcore git-subtree-split: 82460b2124216af1ac9d63060de310a682a2fd15 |
||
---|---|---|
.. | ||
bash-completion/completions | ||
docker | ||
src | ||
websocket | ||
BUILD.bazel | ||
CMakeLists.txt | ||
README.md | ||
tox-bootstrapd.conf | ||
tox-bootstrapd.service | ||
tox-bootstrapd.sh |
Instructions
These instructions are primarily tested on Debian Linux, Wheezy for SysVinit and Jessie for systemd, but they should work on other POSIX-compliant systems too.
For systemd
users
Setting up
For security reasons we run the daemon under its own user.
Create a new user by executing the following:
sudo useradd --home-dir /var/lib/tox-bootstrapd --create-home --system --shell /sbin/nologin --comment "Account to run Tox's DHT bootstrap daemon" --user-group tox-bootstrapd
Restrict access to home directory:
sudo chmod 700 /var/lib/tox-bootstrapd
Copy tox-bootstrapd.conf
file to where ExecStart=
from tox-bootstrapd.service
points to. By default it's /etc/tox-bootstrapd.conf
.
sudo cp tox-bootstrapd.conf /etc/tox-bootstrapd.conf
Go over everything in the copied tox-bootstrapd.conf
file. Set options you want and add actual working nodes to the bootstrap_nodes
list, instead of the example ones, if you want your node to connect to the Tox network. Make sure pid_file_path
matches PIDFile=
from tox-bootstrapd.service
.
Copy tox-bootstrapd.service
to /etc/systemd/system/
:
sudo cp tox-bootstrapd.service /etc/systemd/system/
You must uncomment the next line in tox-bootstrapd.service, if you want to use port number < 1024:
#CapabilityBoundingSet=CAP_NET_BIND_SERVICE
and, possibly, install libcap2-bin
or libcap2
package, depending of your distribution.
Reload systemd units definitions, enable service for automatic start (if needed), start it and verify it's running:
sudo systemctl daemon-reload
sudo systemctl enable tox-bootstrapd.service
sudo systemctl start tox-bootstrapd.service
sudo systemctl status tox-bootstrapd.service
Get your public key and check that the daemon initialized correctly:
sudo grep "tox-bootstrapd" /var/log/syslog
Updating
You want to make sure that the daemon uses the newest toxcore, as there might have been some changes done to the DHT, so it's advised to update the daemon at least once every month.
To update the daemon first stop it:
sudo systemctl stop tox-bootstrapd.service
Then update your toxcore git repository, rebuild the toxcore and the daemon and make sure to install them.
Check if tox-bootstrapd.service
in toxcore git repository was modified since the last time you copied it, as you might need to update it too.
Reload tox-bootstrapd.service
if you have updated modified it:
sudo systemctl daemon-reload
After all of this is done, simply start the daemon back again:
sudo systemctl start tox-bootstrapd.service
Troubleshooting
- Check daemon's status:
sudo systemctl status tox-bootstrapd.service
- Check the log for errors:
sudo grep "tox-bootstrapd" /var/log/syslog
# or
sudo journalctl --pager-end
# or
sudo journalctl -f _SYSTEMD_UNIT=tox-bootstrapd.service
-
Make sure tox-bootstrapd user has write permission for keys and pid files.
-
Make sure tox-bootstrapd has read permission for the config file.
-
Make sure tox-bootstrapd location matches its path in tox-bootstrapd.service file.
For SysVinit
users
Setting up
For security reasons we run the daemon under its own user.
Create a new user by executing the following:
sudo useradd --home-dir /var/lib/tox-bootstrapd --create-home --system --shell /sbin/nologin --comment "Account to run Tox's DHT bootstrap daemon" --user-group tox-bootstrapd
Restrict access to home directory:
sudo chmod 700 /var/lib/tox-bootstrapd
Copy tox-bootstrapd.conf
file to where CFGFILE
variable from tox-bootstrapd.sh
points to. By default it's /etc/tox-bootstrapd.conf
.
sudo cp tox-bootstrapd.conf /etc/tox-bootstrapd.conf
Go over everything in the copied tox-bootstrapd.conf
file. Set options you want and add actual working nodes to the bootstrap_nodes
list, instead of the example ones, if you want your node to connect to the Tox network. Make sure pid_file_path
matches PIDFILE
from tox-bootstrapd.sh
.
Look at the variable declarations in the beginning of tox-bootstrapd.sh
init script to see if you need to change anything for it to work on your system. The default values must be fine for most users and we assume that you use those next.
If you have configured the daemon to use any port numbers that are lower than 1024, you need to execute the command below, as by default non-privileged users cannot open ports <1024. The change persists through reboot:
sudo setcap CAP_NET_BIND_SERVICE=+eip /usr/local/bin/tox-bootstrapd
Copy tox-bootstrapd.sh
init script to /etc/init.d/tox-bootstrapd
(note the disappearance of ".sh" ending):
sudo cp tox-bootstrapd.sh /etc/init.d/tox-bootstrapd
Set permissions for the init system to run the script:
sudo chmod 755 /etc/init.d/tox-bootstrapd
Make the init system aware of the script, start the daemon and verify it's running:
sudo update-rc.d tox-bootstrapd defaults
sudo service tox-bootstrapd start
sudo service tox-bootstrapd status
Get your public key and check that the daemon initialized correctly:
sudo grep "tox-bootstrapd" /var/log/syslog
Updating
You want to make sure that the daemon uses the newest toxcore, as there might have been some changes done to the DHT, so it's advised to update the daemon at least once every month.
To update the daemon first stop it:
sudo service tox-bootstrapd stop
Then update your toxcore git repository, rebuild the toxcore and the daemon and make sure to install them.
Check if tox-bootstrapd.sh
in toxcore git repository was modified since the last time you copied it, as you might need to update it too.
After all of this is done, simply start the daemon back again:
sudo service tox-bootstrapd start
Troubleshooting
- Check daemon's status:
sudo service tox-bootstrapd status
- Check the log for errors:
sudo grep "tox-bootstrapd" /var/log/syslog
-
Check that variables in the beginning of
/etc/init.d/tox-bootstrapd
are valid. -
Make sure tox-bootstrapd user has write permission for keys and pid files.
-
Make sure tox-bootstrapd has read permission for the config file.
-
Make sure tox-bootstrapd location matches its path in the
/etc/init.d/tox-bootstrapd
init script.
For Docker
users:
Setting up
If you are familiar with Docker and would rather run the daemon in a Docker container, you may download the latest official docker image. To download the latest image run:
docker pull toxchat/bootstrap-node:latest
docker run --rm -it --entrypoint=sha256sum toxchat/bootstrap-node:latest /usr/local/bin/tox-bootstrapd
This will print the SHA256 checksum of the latest binary, which should agree with the SHA256 checksum in the Dockerfile.
If you want to build the bootstrap node from source, check out the latest release:
git checkout $(git tag --list | grep -P '^v(\d+).(\d+).(\d+)$' | \
sed 's/v/v /g' | sed 's/\./ /g' | \
sort -snk4,4 | sort -snk3,3 | sort -snk2,2 | tail -n 1 | \
sed 's/v /v/g' | sed 's/ /\./g')
and run the following from the top level c-toxcore directory:
tar c $(git ls-files) | docker build -f other/bootstrap_daemon/docker/Dockerfile -t toxchat/bootstrap-node -
sudo useradd \
--home-dir /var/lib/tox-bootstrapd \
--create-home \
--system \
--shell /sbin/nologin \
--comment "Account to run Tox's DHT bootstrap daemon" \
--user-group tox-bootstrapd
sudo chmod 700 /var/lib/tox-bootstrapd
docker run -d --name tox-bootstrapd --restart always \
--user "$(id -u tox-bootstrapd):$(id -g tox-bootstrapd)" \
-v /var/lib/tox-bootstrapd/:/var/lib/tox-bootstrapd/ \
--ulimit nofile=32768:32768 \
-p 443:443 \
-p 3389:3389 \
-p 33445:33445 \
-p 33445:33445/udp \
toxchat/bootstrap-node
We create a new user and protect its home directory in order to mount it in the Docker image, so that the kyepair the daemon uses would be stored on the host system, which makes it less likely that you would loose the keypair while playing with or updating the Docker container.
You can check logs for your public key or any errors:
docker logs tox-bootstrapd
Note that the Docker container runs a script which pulls a list of bootstrap nodes off https://nodes.tox.chat/ and adds them in the config file.
Updating
You want to make sure that the daemon uses the newest toxcore, as there might have been some changes done to the DHT, so it's advised to update the daemon at least once every month.
To update the daemon, all you need is to erase current container with its image:
docker stop tox-bootstrapd
docker rm tox-bootstrapd
docker rmi toxchat/bootstrap-node
Then rebuild and run the image again:
tar c $(git ls-files) | docker build -f other/bootstrap_daemon/docker/Dockerfile -t toxchat/bootstrap-node -
docker run -d --name tox-bootstrapd --restart always \
--user "$(id -u tox-bootstrapd):$(id -g tox-bootstrapd)" \
-v /var/lib/tox-bootstrapd/:/var/lib/tox-bootstrapd/ \
--ulimit nofile=32768:32768 \
-p 443:443 \
-p 3389:3389 \
-p 33445:33445 \
-p 33445:33445/udp \
toxchat/bootstrap-node
Troubleshooting
- Check if the container is running:
docker ps -a
- Check the log for errors:
docker logs tox-bootstrapd