> For the complete documentation index, see [llms.txt](https://docs.alephzero.org/aleph-zero/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.alephzero.org/aleph-zero/use/validate/running-an-aleph-node-on-testnet/building-and-running-from-source-advanced/running-the-binary.md).

# Running the binary

#### File descriptor limit

Since you are not running the node in a docker, there is one additional OS level configuration you have to take care of. The node uses a lot of file descriptors, both for network connections as well as database handling. To check the limit run:

```bash
ulimit -Hn
```

If the returned value is at least 10000 you are set. Otherwise increase the limit to the recommended value by running:

```bash
DESCRIPTOR_CONFIG="$USER hard nofile 10000"
sudo echo $DESCRIPTOR_CONFIG >> /etc/security/limits.conf
```

Relogin and check the limit again to ensure the change took effect.

#### Run!

First, source the script you created two chapters back and set the config variables:

```bash
source set_env.sh
```

Once all of the variables mentioned in the previous chapters are set, you can start the node with the following command:

```bash
aleph-node \
  --validator \
  --enable-pruning \
  --database rocksdb
  --execution Native \
  --database paritydb \
  --name "${NAME}" \
  --base-path "${BASE_PATH}" \
  --pool-limit "${POOL_LIMIT}" \
  --chain "${CHAIN}" \
  --node-key-file "${NODE_KEY_PATH}" \
  --backup-path "${BACKUP_PATH}" \
  --rpc-port "${RPC_PORT}" \
  --ws-port "${WS_PORT}" \
  --port "${PORT}" \
  --validator-port "${VALIDATOR_PORT}" \
  --public-addr "${PUBLIC_ADDR}" \
  --public-validator-addresses "${PUBLIC_VALIDATOR_ADDRESS}" \
  --rpc-cors all \
  --no-mdns \
  --ws-max-connections "${WS_MAX_CONNECTIONS}" \
  --enable-log-reloading \
  --rpc-methods Unsafe \
  --bootnodes "${BOOT_NODES}" \
  --telemetry-url "${TELEMETRY_URL} ${TELEMETRY_VERBOSITY_LEVEL}" > ${LOG_FILE} 2>&1
```

#### Further improvements

We recommend restarting the node automatically whenever it crashes. This is a rare occurrence, but were the node to crash you might miss the performance targets for a session and receive lower rewards. You should also set up some monitoring to check whether the node operates correctly – the prometheus endpoint is likely to be useful – including checking for crashloops. The details of such setups are beyond the scope of this guide.