Rocket Pool Support

Troubleshooting and self-help resources.

Is the transaction pending, or failed?

Go to etherscan.io and search for your node wallet address.\n\nIs the transaction pending, or did it fail?

It’s pending

It failed

Stuck Transaction

Transactions get stuck in a pending state when the gas settings specified were too low.

You can either wait for gas prices to lower enough and see if the transaction goes through on its own, or overwrite it.

If you decide to overwrite it, you can either speed it up or cancel and resubmit it.

Failed Transaction

If a transaction takes too long, it may be dropped.

You will have to re-submit it.

Full Disk

Is your node using a second drive to store the chain data?

I am using a second drive.

I am storing the chain data on the same drive as the operating system.

Second Drive Remaining Space

Replace /path/to/chaindata/mount in the following command and run it:

df -h /path/to/chaindata/mount

Do you have at least 50G free?

I do

I do not

Prune Geth

You should have enough free space to prune Geth.

Find the directions to do that here

Resync Geth

Since you do not have enough free space to prune, you will have to resync geth’s data.

The process is similar to pruning, but will take longer.

To begin, run:

rocketpool service resync-eth1

and confirm when prompted.

If you’re using an infura fallback, you should also run:

docker stop rocketpool_node

to avoid hitting your free tier limits. This container uses the API but isn’t required unless you’re launching new minipools or need to auto-claim RPL rewards.

When rocketpool node sync shows that eth1 has fully synced (this may take a few days), run docker start rocketpool_node.

Root Partition Size

Run df -h /

Is the “Size” column representative of the full capacity of your SSD?

Yes, the partition is using my whole SSD.

No, the partition is smaller than my SSD capacity.

Root Partition Remaining Space

Run df -h /. Do you have at least 50G free?

Yes, I have at least 50G free.

No, I have less than 50G free.

LVM

When you installed Ubuntu, it may not have partitioned the full disk.

Run lsblk, and look for a section similar to:

└─nvme0n1p3               259:3    0  1.8T  0 part
  └─ubuntu--vg-ubuntu--lv 253:0    0  100G  0 lvm  /

Note that the disk capacity is 1.8T, but the volume is only 100G.

Did this happen to you?

Yes, this happened to me.

No, something else is wrong.

Grow LVM Volume

Run:

sudo lvextend -An -l +100%FREE -r /dev/ubuntu-vg/ubuntu-lv

to resize the root volume. If this command doesn’t print any errors, run sudo vgcfgbackup to backup the new configuration.

Does df -h / show free space now?

I got an error

No, the disk is still full.

Yes, it does.

Find us on Discord

Unfortunately, we recommend that you seek help in the #support channel on Rocket Pool’s Discord.

Please use the link below to join.

Done!

In that case, you should be all done!

How many attestations are you missing?

Are you missing all your attestations, or just some?

All

Some

Is it due to Doppelgänger Protection?

Did you recently restart your validator?

Yes

No

Doppelgänger Detection

If you recently restarted your validator and are missing attestations, it is likely because you have Doppelgänger Detection enabled.

With Doppelgänger Detection, your validator will skip attestations for 1-3 epochs while monitoring the beacon chain to ensure no other validators are currently attesting with your validator keys. This can help prevent you from accidentally getting slashed, either by running multiple validators with the same keys or by switching clients too quickly.

Please wait and see if your attestations resume after a few epochs.

Is your drive full?

Run df -h /. Is the volume where you store your chain data (/ by default) completely full?

Yes

No

Peer Count

Does your Grafana dashboard show a low or erratic peer count?

Yes

No

Port Forwarding

The main reason for a low peer count is misconfigured port forwarding settings.

Head to YouGetSignal and test your P2P ports (9001 and 30303 by default) against your Node’s public IP address.

If you’re staking from home, the pre-populated IP address is probably correct. If you’re staking from a VPS, enter your node’s IP address by hand.

If the ports show as “Closed”, you will need to configure port fowarding in your router (or with your VPS provider, if you’re not staking from home).

Unfortunately this process is different for each router, and you will have to find instructions specific to your router model.

If your ports show as “Open” and you still have erratic or low peers, the other culprits could be performance issues or bandwidth limitations. Please seek further help on the Rocket Pool Discord.

Is NTP working?

Run timedatectl status. Does it show NTP service: active?

Yes

No

Install Chrony

If NTP is inactive, your system clock isn’t being kept in sync with the rest of the network.

The easiest way to fix this is to install chrony by running:

sudo apt update
sudo apt install chrony

Did snap install docker?

Run snap list. Does it print a line containing "Docker"?

Yes

No

Snapd Remediation

Occasionally, we have observed Nodes with conflicting versions of docker installed.

This appears to be the case with you. To resolve this issue, run the following commands:

sudo systemctl stop snap.docker.dockerd.service
sudo snap remove --purge docker
sudo systemctl restart docker.socket docker.service

How much data is missing?

Make sure the smartnode is running with rocketpool service status. Is all your data missing, or only some?

All

Some

Did snap install docker?

Run snap list. Does it print a line containing "Docker"?

Yes

No

Is the Prometheus Data Source correctly set?

In Grafana, navigate to Configuration->Data Sources. Is the Prometheus Data Source set to "default"?

Yes

No

Reset Prometheus Data Source

Sometimes, Grafana seems to un-link from Prometheus.

Simply click on the Prometheus data source and set as the default, using the button at the top of the settings.

Is UFW allowing Prometheus access to Node-Exporter?

If you run:

sudo ufw status

do you see a line like the following one?

9103                       ALLOW       172.23.0.0/16              # Allow prometheus access to node-exporter

Yes

No

Follow the guides to update your UFW settings.

Your UFW settings are preventing Prometheus from accessing the Node-Exporter container, which it needs to do in order to query some metrics.

Please follow this section of the guides to allow it access: specifically, the yellow “NOTE” box.

Is Node-Exporter running?

Run rocketpool service status and look for a line that looks like this:

rocketpool_exporter        /bin/node_exporter --path. ...   Up

Does it exist, and say “Up”?

Yes

No

Is rocketpool_node running?

Run rocketpool service status and look for a line that looks like this:

rocketpool_node            /go/bin/rocketpool -m 0.0. ...   Up                                                                                                                                          

Does it exist, and say “Up”?

Yes

No

Start the service.

It seems like the whole smart node service isn’t running.

Some containers are automatically stopped in certain situations (i.e. pruning).

Please try to restart them (assuming you are not longer pruning): rocketpool service start

Submit Feedback
Thanks for your feedback!