SSH won't connect via IP or umbrel.local (rbpi4 full node)

Support and Troubleshooting
1

Using a Raspberry Pi 4B (2TB SSD + Ethernet).

I have run the following commands in my CMD (Win10):

ssh umbrel@192.168.1.112

Connecting via IP results in the command hanging (no response).

ssh umbrel@umbrel.local

Using umbrel.local results in “ssh: Could not resolve hostname umbrel.local: No such host is known.”
I have also tried ssh on another device (macOS) and both commands just hang.

Ping statistics for 192.168.1.112:
    Packets: Sent = 4, Received = 4, Lost = 0 (0% loss)

I can ping the IP and I receive a response meaning the node is connected and reachable.

I had issues with none of my apps starting, updating, or opening. I powered off the node multiple times, tested the connection, took the SD card and SSD out, and tried them again. Nothing worked, and then I seemed to have created another problem as the local web server is no longer reachable.

Please help. Thanks :slight_smile:

2

I have the same exact problem any help would be much appreciated.

3

same issue :frowning:

4

I have the exact same issue, help would be really appreciated…

5

I am experiencing the same issue. For those that posted earlier, any solution? Or just waiting for someone to reply to this thread :sweat_smile: ?

6

does a reboot and persistent attemp to reconnect via ssh allow you in?

7

Could you try generating a verbose output to see where the connection hangs (ssh -vvv umbrel@192.168.1.112)? You can also connect your Raspberry Pi to a video output to show a terminal which should be more helpful for debugging.

8

I actually used the Umbrel UI to shut down (in hindsight I should have used CL). I cannot SSH anymore, as it’s no longer connected to the network (I used CL and no responses. i also used a network scanning app and it’s definitely not connected to the network). I then connected a monitor to the RPi and I only get a blank screen. But the lights on the RPI and the SSD are still turned on.

So Umbrel and the RPi did not shut down properly and my only option is to unplug power supply and hope nothing gets corrupted? Any other options?

9

I unplugged the power supply and plugged it back in. I was able to access Umbrel UI and my node synced. No issues so far (hope that continues).

10

i have a similar post open where I can’t get in as well. my newly arrived server was sent ( i believe ) with a bad ssd. Keep an eye on your memory and disk space. You might be up against a wall. " top " is a linux command to see mem. “df -h” will help you see disk space.

1 Like
12

and if your system is sick, maybe these will help:

1. dmesg --follow

  • Rationale: Monitors kernel messages in real-time, useful for catching hardware or driver issues affecting Umbrel’s Docker containers or storage.
  • Example: sudo dmesg --follow
  • Output & Action: If you see Out of memory: Killed process (docker)—Umbrel’s containers are crashing due to RAM exhaustion. Increase RAM or reduce app load (e.g., stop unused apps with docker stop).

2. iotop

  • Rationale: Tracks disk I/O by process, critical for spotting if Umbrel’s Bitcoin sync or file apps are overloading the disk.
  • Example: sudo iotop -o (shows only active I/O processes)
  • Output & Action: High DISK READ/WRITE for bitcoind (e.g., 100 MB/s)—normal during sync, but if persistent post-sync, check disk health with smartctl.

3. nmon

  • Rationale: Provides an interactive dashboard for CPU, memory, disk, and network, helping assess Umbrel’s resource usage holistically.
  • Example: nmon (press c for CPU, m for memory, d for disk)
  • Output & Action: CPU at 90%+ with dockerd—Umbrel apps may be overtaxed; consider restarting Docker (sudo systemctl restart docker).

4. ss -tuln

  • Rationale: Shows listening sockets, ensuring Umbrel’s app ports (e.g., Bitcoin’s 8333) are open and not conflicting.
  • Example: ss -tuln | grep 8333
  • Output & Action: No output—Bitcoin port isn’t listening; restart the container (docker restart umbrel_bitcoin_1) or check firewall (ufw status).

5. pidstat -u 1

  • Rationale: Monitors CPU usage per process in real-time, pinpointing if Umbrel’s Docker containers are CPU hogs.
  • Example: pidstat -u 1
  • Output & Action: dockerd at 50%+ CPU continuously—investigate with docker logs for the offending app; consider limiting resources in docker-compose.yml.

6. vmstat -SM 1

  • Rationale: Reports memory, CPU, and I/O stats every second, useful for spotting bottlenecks during Umbrel app updates or syncs.
  • Example: vmstat -SM 1
  • Output & Action: High si/so (swap in/out)—system is swapping; add more RAM or reduce Umbrel app load.

7. lsof -i -P -n

  • Rationale: Lists open files and network connections, helping identify what’s using Umbrel’s ports or files.
  • Example: lsof -i -P -n | grep 8333
  • Output & Action: No process on 8333—Bitcoin container may have crashed; restart it (docker restart umbrel_bitcoin_1).

8. watch -n 2 docker stats

  • Rationale: Continuously monitors Docker container resource usage, ideal for daily Umbrel wellness checks.
  • Example: watch -n 2 docker stats
  • Output & Action: umbrel_bitcoin_1 at 90% MEM—RAM is maxed; stop unused apps or upgrade hardware.

9. smartctl -a /dev/sda

  • Rationale: Checks SMART data for disk health, vital since Umbrel’s blockchain data can stress drives.
  • Example: sudo smartctl -a /dev/sda (replace /dev/sda with your drive)
  • Output & Action: Reallocated_Sector_Ct > 0—disk is failing; back up Umbrel data (docker volume ls to find volumes) and replace the drive.

10. htop --filter docker

  • Rationale: Filters processes by name in an interactive view, making it easy to spot Umbrel’s Docker processes.
  • Example: htop --filter docker
  • Output & Action: Multiple dockerd threads at high CPU—check logs (docker logs) for misbehaving apps.

11. lsblk -f

  • Rationale: Shows block devices and filesystems, ensuring Umbrel’s storage mounts are intact.
  • Example: lsblk -f
  • Output & Action: /umbrel not listed—mount point is gone; remount with mount /dev/sdb1 /umbrel (adjust device).

12. fuser /umbrel

  • Rationale: Identifies processes using a directory, useful if Umbrel’s data dir is locked or inaccessible.
  • Example: sudo fuser /umbrel
  • Output & Action: Lists PIDs (e.g., 1234)—kill them (kill -9 1234) if they’re blocking Umbrel.

13. strace -p

  • Rationale: Traces system calls of a process, great for debugging hung Umbrel containers.
  • Example: sudo strace -p $(pidof bitcoind)
  • Output & Action: Repeated read() timeouts—network issue; check connectivity (ping 8.8.8.8).

14. tcpdump -i eth0

  • Rationale: Captures network packets, helping diagnose Umbrel’s app connectivity (e.g., Lightning Network).
  • Example: sudo tcpdump -i eth0 port 9735
  • Output & Action: No packets on 9735—Lightning port blocked; open it (sudo ufw allow 9735).

15. iostat -x 1

  • Rationale: Provides detailed disk I/O stats, revealing if Umbrel’s storage is a bottleneck.
  • Example: iostat -x 1
  • Output & Action: %util near 100%—disk is saturated; move data to a faster drive (e.g., SSD).

16. watch -d free -h

  • Rationale: Highlights memory changes every 2 seconds, catching sudden Umbrel memory spikes.
  • Example: watch -d free -h
  • Output & Action: free drops to <1GB—memory leak; restart Docker (sudo systemctl restart docker).

17. lsmod | grep overlay

  • Rationale: Checks if the overlay kernel module (used by Docker) is loaded, ensuring Umbrel’s containers can run.
  • Example: lsmod | grep overlay
  • Output & Action: No output—module missing; load it (sudo modprobe overlay) or reboot.

18. shuf /umbrel/statuses/*

  • Rationale: Randomly samples Umbrel status files, giving a quick health snapshot.
  • Example: shuf -n 5 /umbrel/statuses/*
  • Output & Action: Errors like sync failed—check specific app logs (docker logs umbrel_bitcoin_1).

19. timeout 10m tail -f /var/log/syslog

  • Rationale: Limits log tailing to 10 minutes, preventing terminal hangups while monitoring Umbrel-related system events.
  • Example: timeout 10m tail -f /var/log/syslog
  • Output & Action: docker: timeout—Docker crashed; restart it (sudo systemctl restart docker).

20. atop -a 1

  • Rationale: Logs all system activity per process with 1-second intervals, perfect for post-mortem Umbrel analysis.
  • Example: sudo atop -a 1
  • Output & Action: DSK column high for bitcoind—disk I/O issue; optimize storage or check health.
2 Likes
13

So useful. Really appreciate you sharing this! The community really benefits from replies like this. At least for those willing to search instead of creating new, repetitive posts. It seems responses to posts have decreased with new users who want everything spoon fed… Anyway, thanks again.

1 Like
14

my pleasure.