I’ve been trying to set up a cardano pool for days and haven’t had success because the socket on the relay/block-producer goes missing. gLive also shows the status as starting and it never changes. Anyone knows what is going on?
What method did you use to configure your node?
How did you determine that socket is missing - what is the error for that?
Patience is the key… u will need to wait… starting is ok, perhaps the node re-sync the db (mandatory with 1.26.1 version)
If u press top this time u should see a high cpu usage…
PS: the socket will be created after the node will start
Yes I have the exact same thing.
My Ubuntu server restarted, looked like everything was paused/stuck on the relay. Running
cardano-cli query tip --mainnet showed an error message of
Network.Socket.connect: <socket: 11>: does not exist.
I was following the Coin Cashew guide and the socket in their guide resides under the db folder. Checking in there my socket has gone.
Following @Alexd1985 recommendation of waiting. Using top I see CPU maxed - I assume retracing its downloaded block chain to see where to resume from.
yes, so when the node is in starting phase the socket is not available
it could take time to open all the databases which the node is using…
you can check the logs of the node and match it with a successfully started one:
I think my issue with the socket going missing was related to using AWS t2.medium instances slowing down due to low CPU credits. I upgraded to R5 instances and the issues went away. I’m paying a lot more now but at least the nodes work reliably.
I have this same issue
cardano-cli: Network.Socket.connect: <socket: 11>: does not exist (No such file or directory)[ec2-user@ip-10-0-1-xxx
I upgraded to m5a from t3a and the problem persists. I just can’t get cardano-cli query blah blah 0 to respond wihtout this message.
As far as I can tell, everything looks correct - but I would not be posting here if it were. I am missng something. gLiveView works on my Relays fine, but gLiveView only displays the tip on the BP - The ports on my BP are open only to the two relays on their ports and subnet and my personal SSH.
I think my NACLs are fine, but I must be wrong about something.
Does anyone have any suggestions to resolve this error on AWS?
sudo systemctl status cardano-node
yes, check if ^ is running, the next thing I would try is to copy the db from the relay to your BP
You can stop relay-1 and BP and then from your BP run a command like this, and then restart the BP
scp -r ubuntu@relay-ip:/opt/cardano/cnode/db /opt/cardano/cnode
I just fixed part of the issue with ‘Security Group’ changes. It looks like I fixed gLiveView on my BP …
Now that gLiveView is synching correct, I can see that I need to wait fo the db to synchronize for a few hours. I am guessing that gLiveView gets data from more than one place … Is ths correct?
I’m using m5a on aws and this seems really slow …
As a side thought, when ADA gets to $50, I’m building a TaberNACL and going to relax …
If your relay db is 100% synched copying it to your BP using scp will take like 10 minutes vs 12 hrs if decide wait for the db to download.
Will do …
Here is the solution to my/the/any socket 11 isuue in AWS. I run this in .bashrc to solve the issue.
sudo chown -R ec2-user cardano-node