Thread by @EthereumOnARM

Today it’s been 1 year since we started staking on a @Raspberry_Pi 4 running a local #Ethereum 1 node & 1 #eth2 node (Beacon+Validator) in the same Rpi

28 Block Proposals & 81,032 Attestations running a 15w powered device. Almost 0 downtime & some minor issues

+info & tips👇

Validator info

- Validator clients: @ethnimbus, @prylabs Prysm & @sigp_io Lighthouse
- Current client: Nimbus
- Proposed blocks: 28
- Attestations: 81,032
- Missed block proposals: 0
- Missed attestations: 1,061
- Balance: 34.4 ETH

Hardware setup

- @Raspberry_Pi 4 8 GB RAM
- Western Digital Blue 1TB 3D NAND
- Ugreen SATA-USB Case with ASMedia Chipset
- Heatsink & fan
- Official power supply

Software setup

- OS: @ubuntu Server 20.04.3
- Eth1 client: @go_ethereum 1.10.13
- Eth2 client: @ethnimbus 1.5.4

Although the @raspberry_pi may look like a fragile computer it is a very reliable and solid device. We’ve been running several models 24/7 since 2016 & never got a hardware issue

The overall experience running the node was great & pretty much we didn’t notice it was around.

Sure we did run into some minor issues and controlled shutdowns:

- A couple blackouts (no data corruption).
- A MicroSD fault (it wasn’t a new one). We attached the USB disk into a new Ethonarm flashed Rpi & kept going.

- 2 clients migrations. Prysm to Lighthouse & Lighthouse to Nimbus. Clients can export/import the slashing database to prevent a slashing event so it is quite safe.

- OS Security updates
- Some @ethnimbus missed attestations caused by a client param modification (our fault). The team looked into it in no time & helped us to fix it very quickly.

Finally, some tips:

- It is OK to be offline, actually, the protocol is designed this way so you don’t get strongly penalized even if you are offline for days. Do not get obsessed with this.
- So, don’t try complex setups or automation tools in order to avoid going offline.

- As complex setups can get you slashed. For instance, do not enable the validator systemd service or any other kind of automatic start.
- Run an UPS to avoid data corruption (more than going offline).

- Run a backup device with an Eth1 node and the Beacon Chain synced. If something goes wrong you can easily switch to it.
- Remember that you can use 3rd party providers such as @infura_io if you are not able to sync an Eth1 or Eth2 nodes on time

- Try to run a minority client if possible. It is good for the network health and it helps developers to make the clients more reliable.
- Wait 2-3 days to update the Eth clients unless the new version fixes some important bug or a security issue.

- Jump into Eth teams discords or open a bug on Github if something goes wrong (missed attestations, missed blocks, low peers, etc) and don’t hesitate to ask for info. Even if your question seems dumb to you it may be a bug or other important issue.

That 's it!

Exciting times ahead and lots of work to do... can’t wait to celebrate another year validating on the #Ethereum network (on a 100% Proof of Stake chain by then).

Happy Birthday Ethereum PoS