Cnode Relay stuck at "starting"

Hope someone can help me out. I’ve been running relay nodes for almost a month now with no issues. However, today, I opened gliveview on one of my relay nodes and it was obvious something was wrong. The status initially ready “Synced 100%” but it wasn’t doing anything. I stopped the node and rebooted the server and now the node says active/running but in gliveview I’m just getting it stuck at Status “Starting” - it’s been over an hour and the same thing.

cnode.service - Cardano Node
Loaded: loaded (/etc/systemd/system/cnode.service; enabled; vendor preset: enabled)
Active: active (running) since Mon 2021-03-01 23:38:39 UTC; 48min ago
Main PID: 63464 (
Tasks: 14 (limit: 4683)
Memory: 3.2G
CGroup: /system.slice/cnode.service
├─63464 /bin/bash /opt/cardano/cnode/scripts/
└─63578 cardano-node run --topology /opt/cardano/cnode/files/topology.json --config /opt/cardano/cnode/files/config.json --database-path /o>

Mar 01 23:38:39 systemd[1]: Started Cardano Node.
Mar 01 23:38:40 cnode[63464]: WARN: A prior running Cardano node was not cleanly shutdown, socket file still exists. Cleaning up.
Mar 01 23:38:41 cnode[63578]: Listening on

Any ideas?

1 Like

Going on 5 hours now and still just stuck at “Starting” - anyone? Can I just delete the database and have it re-sync?

It is generally a good idea to make a backup of your block data. I only keep these files/dirs …

immutable  ledger  protocolMagicId

all else is volatile and should go.

If you tar czf ... the above you will get a tgz of about 3.5GB

This is useful to do every now and then and allows you to spin up new nodes real quick.

Yes, I could see the benefit in doing that, but any idea why the node is stuck like it is? Would prefer to sort that out in case it happens again…

I run Docker images. If something goes wrong, I can restart a new node in a matter of seconds and don’t need to worry about what funky state my box has gotten into.

There is CIP about this here.

Were you able to get the help you needed for your problem yet? I started experiencing this with one of my nodes recently.

I ended up basically re-installing cardano node on that server and it fixed the issue right away. Maybe give it a try?

I believe I saw something similar on the forum and someone used a killall to resolve the problem with the socket staying open.