Photo by Zoltan Tukacs on Unsplash
So you've become a Chia farmer and want to maximize the probability of getting a reward? Chiadog helps with automated monitoring and sends you a mobile notification in case something appears to malfunction.
| Subsystem | Notification (example values) | Priority |
|---|---|---|
| Harvester | Your harvester appears to be offline! No events for the past 400 seconds. | HIGH |
| Harvester | Disconnected HDD? The total plot count decreased from 100 to 40. | HIGH |
| Harvester | Connected HDD? The total plot count increased from 0 to 42. | LOW |
| Harvester | Experiencing networking issues? Harvester did not participate in any challenge for 120 seconds. It's now working again. | NORMAL |
| Harvester | Seeking plots took too long: 21.42 seconds! | NORMAL |
| Full Node | Experiencing networking issues? Skipped 42 signage points! | NORMAL |
| Full Node | Block found!! | LOW |
| Wallet | Cha-ching! Just received 2.0 XCH ☘️ | LOW |
| Daily Stats | Hello farmer! 👋 Here's what happened in the last 24 hours: Received ☘️: 2.00 XCH️ Proofs 🧾: 176 found! - 176 partials submitted 📑 - 0 blocks found 📦 Search 🔍: - average: 0.46s - over 5s: 2 occasions - over 15s: 1 occasions Plots 🌱: 42, new: 2 Eligible plots 🥇: 0.08 average Skipped SPs |
LOW |
Please refer to Status Reference page for detailed explanations of the notifications.
It parses the debug.log generated by the chia process and runs various checks to determine the health of your farmer.
Among others, it can detect if your node has lost sync and the farmer is no longer participating in challenges, or if
one of your external HDDs disconnected and your harvester doesn't have access to the full amount of plots you have.
It only requires read-access to your debug.log file and internet connection to send out notifications. It's highly
recommended that you run chiadog in a sandboxed environment. Please use the official docker image.
Furthermore, following best security practices, you should keep your wallet separate.
You may use one (or more) of the following integrations to receive notifications from chiadog.
| Integration | Advantages | Cost |
|---|---|---|
| Pushover | High priority notifications that can override your phone's silent mode. | $5 one time purchase after 30 day trial. |
| Pushcut | Alternative to Pushover | |
| You probably already have an email. No additional apps. | Free | |
| Slack | Quick & easy setup. | Free |
| Discord | Quick & easy setup. | Free |
| Telegram | Quick & easy setup. | Free |
| Shell script (beta) | Execute anything in your own script. | Free |
| MQTT | Well-suited for Home Automation. | Free |
| Grafana | For hardware monitoring. | Free |
| Ifttt | Can be used to send push notifications or to do other API integrations depending on incoming data. | Free |
For detailed guide on how to test and configure, please refer to INTEGRATIONS.md.
- Linux, MacOS & Windows
- Python 3.7+
- Git
- Enabled
INFOlogs on your chia farmer
The instructions below are specific to Linux and MacOS, for installing chiadog on Windows, please refer to this
separate README section.
First configure the log level to INFO. This ensures that all logs necessary for chiadog to operate are available
under ~/.chia/mainnet/log/debug.log.
chia configure -log-level=INFO
Then restart your farmer to apply the changes:
chia start --restart farmer
Check that logs are coming in:
cat ~/.chia/mainnet/log/debug.log
The new recommended way of using chiadog is via the official docker image.
For updating from previous version, see section below.
- Clone the repository
git clone https://github.com/martomi/chiadog.git
cd chiadog
- Run the install script.
./install.sh
- Copy the example config file
cp config-example.yaml config.yaml
- Open up
config.yamlin your editor and configure it to your preferences. The example is large, feel free to omit any portions where you're fine with the defaults!
Skip this if you followed the above section.
cd chiadog
git fetch
git checkout main
git pull
./install.sh
-
Open
config.yamland configurefile_log_consumer:- You need to enable the file log consumer to read local chia log files
- Double-check that the path to your chia logs is correct
-
Start the watchdog
./start.sh
- Verify that your plots are detected. Within a few seconds you should see INFO log:
Detected new plots. Farming with 42 plots.
If you see the above log message, you can be certain that chiadog is running correctly. It'll remain silent until the
next scheduled daily notification or until any issues are detected.
You can repeat the above process for every machine where you are running a harvester. Use
the notification_title_prefix
in config.yaml to give every machine a unique notification prefix so that you can easily distinguish them.
If you don't want to setup chiadog on each machine separately, you can also monitor multiple remote harvesters and run
chiadog on a single machine. Please refer to the wiki page
on Remote Monitoring Multiple Harvesters.
You can enable more verbose logging from config.yaml by changing INFO to DEBUG. You should see logs for every
keep-alive event from the harvester.
Chia has not yet introduced timezone aware log timestamps. Until they do make sure the TZ environment variable matches
between the machine producing the logs and running chiadog. If running chiadog on Docker, pass in -e TZ=UTC, substituting
UTC with your timezone, for example Europe/London.
There are failure-cases in which chiadog is helpless. For example, your computer completely freezes or shuts down.
Perhaps your entire home network goes down. chiadog won't be able to send you a notification.
There's a way however: in the config under the section of keep_alive_monitor, you can enable
pinging to a remote service that will act as a watchdog of chiadog. A second level of redundancy, if you wish!
You may chose your favourite service for that, I've tested it with HealthChecks.io. It's free to signup and create an endpoint that expects to receive pings every 10 minutes. If it does not, it will notify you. It has integrations with Pushover, Email, Slack, Discord and more.
. ./venv/bin/activate
nohup python3 -u main.py --config config.yaml > output.log 2>&1 &
To stop chiadog, you can find the Process ID (PID) via ps aux | grep main.py and then softly interrupt the process
with kill -SIGINT <pid_here>.
Alternatively to the original chiadog docker image, you can setup a systemd service which runs chiadog as a limited user and blocks access to key chia locations.
Contributions are always welcome! Please refer to CONTRIBUTING documentation.