UPDATE CARDANO NODE to 1.27.0 - FOR CNTOOLS users

Out of curiosity, do people compile on one relay and copy the binary over to another to reduce downtime? If so, is there a proper way of doing that? If not, why is better to compile on each node? Should we also compile on the block producer?

U can download the binari files and upload to the new server

1 Like

And this can also be done with the block producer?

Yes, it’s the same node as a Relay

1 Like

Really Thanks you Alex!

This is full update for Full-Node.

Could you guide how to update for air-gap (ubuntu Virtual box). Air-gap is not for run full-node, and just for do some information-pool-changing when we need.

Many Thanks!

nano env and check if u have commented the check for updated line

1 Like

Hello Alex,
2 questions, please:

  • Removing a relay, could you remind me how to modify the topology of the BP ?
  • Before rotating KES with CNTOOLS, do the keys have to be decrypted (I think yes) ?
    Big thank you as always…
    Alain
  • Removing a relay, could you remind me how to modify the topology of the BP ?

go to cd $CNODE_HOME/files
nano topology.json and delete the relay

Save the file and restart the node

  • Before rotating KES with CNTOOLS, do the keys have to be decrypted (I think yes) ?

Yes, the pool files need to be decrypted

Thanks You so much, Alex! I’m really appreciate!

Hi Alex,
Thanks for answering. When I modified the BP’s topology.json file as you told me, I found a one-line text. So I replaced it with the declaration of the 2 relays as in your tuto. Then I ran the topologyUpdater script (did I do well ?). Now I have this content in the topology.json:

{ “resultcode”: “402”,
“datetime”:“2021-05-26 15:04:23”,
“clientIp”: “a.b.c.d”,
“iptype”: 4,
“msg”: “IP is not (yet) allowed to fetch this list”,
“Producers”: [ { “addr”: “relays-new.cardano-mainnet.iohk.io”, “port”: 3001, “valency”: 2, “debug”:“default fallback result” } ] }

On the gliveview, the 2 relays are well communicating with the BP (IN and OUT). Do you think my BP’s topology is ok ? Poolvet is still indicating an error (unreachable) with the removed relay…

try to check the port from relay… it is opened to accept connection from any?

you can test from producer typing

telnet Relay-IP Relay-port (ex telnet 192.168.1.1 6000)

you should see connected

On my BP, I restored the topology.json without executing the topologyUpdater script ! So the content of topology.json has not changed !
On the BP, I see my 2 relays IN and OUT
On the relays, my BP is visible IN and OUT
I think all is ok now thanks to you, one more a time !

I will soon withdraw this pool and delegate to Charity Pool …
Sincerely,
Alain

I cannot get 1.26 upgraded to 1.27.

Ive tried multiple times. Tried multiple guides, still stuck on 1.26. Do i need to remove any folders or configs before doing this update?

Also, ghcup can never be installed no matter what i try…

Screenshot 2021-06-07 150711

cardano-node - -version
cardano-cli - -version

Both are version 1.26.1

Ok, did u rebuilt the node?

I think so, I followed coincashews guide and your guide to upgrade. Do i need to remove any folders or paths before doing this?

All im really trying to do is retire my pool. But i need to upgrade to 1.27 before i can do that.

But… not ok, u need to upgrade the node using another guide

1 Like

Thanks, i got it working. After installing ghc the path wasnt being added to bashrc. so i added that manually:
export PATH="$HOME/.cabal/bin:$HOME/.ghcup/bin:$PATH"

Then everything else fell into place.

glade u fixed