Second Node setup, Status OFFLINE but uptime 1 hour+?

I bet this is a simple fix. But, I decided to get a second Raspberry Pi 4 device and another 4TB external drive to my setup to run another Storj node behind my home router.

I finished getting everything setup including making additional port forwards to the second device, with the port number increasing by 1 (23 for SSH, 14003, 28968, etc), but after I got the docker, watch tower and everything setup, checking up on the CLI dashboard.sh screen I notice that the node status says OFFLINE (but uptime was over an hour):

Status OFFLINE
Uptime 20m28s

                   Available     Used     Egress     Ingress
     Bandwidth           N/A      0 B        0 B         0 B (since Aug 1)
          Disk       3.32 TB      0 B
Internal 127.0.0.1:7778

I ALSO noticed that the “ID” field at the top appears to be the same as my first node setup. So, should that be the same or should it be different somehow, and something is trying to reference my first node setup behind the same router? Is my suspicion true that the ID needs to be different, or that I need to do something different with the certificate that I made earlier on my M1 Mac mini (before I transferred it over to the default storage identity directories) on the Raspberry Pi?

And do I need to get you guys logs? I can get you logs of my first and second nodes. Second dashboard is working fine. I feel like I’m real close, right? Thanks in advance.

My docker launch command:

docker run -d --restart unless-stopped --stop-timeout 300 \
    -p 28968:28968/tcp \
    -p 28968:28968/udp \
    -p 127.0.0.1:14003:14003 \
    -e WALLET="REDACTED" \
    -e EMAIL="REDACTED" \
    -e ADDRESS="REDACTED.ddns.net:28968" \
    -e STORAGE="3.32TB" \
    --mount type=bind,source="/home/pi/.local/share/storj/identity/storagenode",destination=/app/identity \
    --mount type=bind,source="/mnt/storj",destination=/app/config \
    --name storagenode storjlabs/storagenode:latest

The Uptime on the dashboard mean how long your node running, not how long it was online.
So, you need to fix the error in the port mapping. The right part of the port mapping must not be changed unless you modified a listening port in the config/additional options in the docker run command.
So, the correct port mapping should looks like

That change allowed it to work! Thank you. However, web dashboard doesn’t appear to work; CLI dashboard appears shows in terminal.

  1. What could prevent the CLI showing online and working, but no browser letting me access the web GUI using the external no-ip domain and port assigned by my no-ip settings that the CLI dashboard is giving me as a URL to try and view?

  2. And as I have two nodes that both appear online for me, how could I more accurately monitor my potential earnings and withheld amounts since I have two nodes pointing to the same Storj wallet address? Will they combine no matter which web GUI I access? I appreciate the assistance.

You need to use article How to remote access the web dashboard - Node Operator to have an access to your dashboard on your phone or PC/Mac.

The other way is to remove 127.0.0.1: from port mapping for the dashboard port. This will allow you to connect to you node via LAN.
However, I would ask you to do not forward port for dashboard on your router, because the dashboard is not protected and you will expose your personal information to the whole internet. If you want to have a remote access outside your network - use the above guide.
The alternative is to use [Tech Preview] Multinode Dashboard Binaries for all your nodes, it doesn’t requires to forward any additional port except standard node port which will be forwarded to make the node work.

I understand. One of the storage nodes had the remote thing but I undid it to your suggestion. So now it’s all completely secure again.

Last question. Is there a way to look at payout information for all nodes I would set up? Because I can see the panel information for each individual node. So, is there an integrated way to see the collective earnings to get a more accurate reading of one estimated payouts would be?

Thanks. I will probably try that later. But I think I’ve done enough commandline stuff for a week; this is all very new to me. Thank you very much for the help. They are both online now.

1 Like