7

Description

I recently went through a router failure and an upgrade. And with that local and external IP addresses changes. After an IP address update both relay and producer nodes are fetching updates from the blockchain (as in passive node) but not processing transactions. Or at least this is what Guild Live View is showing. Interestingly the producer node just minted a block that I take that will be lost on network write sync up.

Setup

  • Relay node: 192.168.0.12
  • Producer node: 192.168.0.10
  • IP: 82.44.254.62
  • Adapools: see here
  • Pooltool: see here
  • Pool.vet: see here

Topology

a) Relay

  1 {
  2   "Producers": [
  3     {
  4       "addr": "192.168.0.10",
  5       "port": 6000,
  6       "valency": 1
  7     },
  8     {
  9       "addr": "relays-new.cardano-mainnet.iohk.io",
 10       "port": 3001,
 11       "valency": 50 
 12     }
 13   ]
 14 }

b) Producer

  1 {
  2   "Producers": [
  3     {
  4       "addr": "192.168.0.12",
  5       "port": 6000,
  6       "valency": 1
  7     }
  8   ]
  9 }

                                                                                                                                                                                                          

Exceptions (might be related as well as might not):

a) Relay

20844 Apr 27 18:52:38 jolanta bash[677]: [jolanta:cardano.node.ChainDB:Notice:108] [2021-04-27 17:52:38.20 UTC] Chain extended, new tip: 1e774042b9adf5e4d4677dac54069ecce09926af382f5979e3db3200bfae31f2 at slot 2      7979667
20845 Apr 27 18:52:51 jolanta bash[677]: [jolanta:cardano.node.DnsSubscription:Error:2319] [2021-04-27 17:52:51.67 UTC] Domain: "relays-new.cardano-mainnet.iohk.io" Application Exception: 204.236.187.238:3001 Mu      xError MuxBearerClosed "<socket: 45> closed when reading data, waiting on next header True"
20846 Apr 27 18:52:51 jolanta bash[677]: [jolanta:cardano.node.ErrorPolicy:Warning:127] [2021-04-27 17:52:51.67 UTC] IP 204.236.187.238:3001 ErrorPolicySuspendPeer (Just (ApplicationExceptionTrace (MuxError MuxB      earerClosed "<socket: 45> closed when reading data, waiting on next header True"))) 20s 20s
20847 Apr 27 18:52:52 jolanta bash[677]: [jolanta:cardano.node.DnsSubscription:Notice:4453] [2021-04-27 17:52:52.72 UTC] Domain: "relays-new.cardano-mainnet.iohk.io" Connection Attempt Start, destination 3.9.116      .228:3001

Netstat

a) Relay

maciej@jolanta:~/cardano-node$ netstat -pant | grep 6000
(Not all processes could be identified, non-owned process info
 will not be shown, you would have to be root to see it all.)
tcp        0      0 192.168.0.12:6000       0.0.0.0:*               LISTEN      677/cardano-node    
tcp        0      0 192.168.0.12:6000       192.168.0.10:45311      ESTABLISHED 677/cardano-node    
tcp        0      0 192.168.0.12:37851      192.168.0.10:6000       ESTABLISHED 677/cardano-node 

b) Producer

maciej@wladyslawa:~/cardano-node$ netstat -pant | grep 6000
(Not all processes could be identified, non-owned process info
 will not be shown, you would have to be root to see it all.)
tcp        0      0 192.168.0.10:6000       0.0.0.0:*               LISTEN      718/cardano-node    
tcp        0      0 192.168.0.10:6000       192.168.0.12:37851      ESTABLISHED 718/cardano-node    
tcp        0      0 192.168.0.10:45311      192.168.0.12:6000       ESTABLISHED 718/cardano-node   

enter image description here

UPDATE:

Logs

Relay: Download

Producer: Download

7
  • 1
  • 2
    Thanks. Did the whole IP address change according to this very guide. Paragraph 18.4 coincashew.com/coins/overview-ada/…
    – matcheek
    Apr 27 at 20:59
  • 1
    Sorry not sure :) Apr 27 at 21:02
  • 1
    Have you verified that both machines are running with the correct IP and has firewalls allowing for the new IPs? The relay might be pushing blocks to the BP, but the BP may not be able to push its blocks back to the relay.
    – Kc Gibson
    Apr 27 at 22:57
  • 1
    did you disable "TraceMempool" in main config file ? can you send screen for your relay ?
    – Mahdi Gh
    May 2 at 10:12
4

"Failed to start all required subscriptions"

As recommended by @KC Gibson, reduce the valency for the IOHK node to two, and restart the node. The remote host is closing your connections, probably due to some connection limiting policy on their end.

I recently went through a router failure and an upgrade. And with that local and external IP addresses changes. After an IP address update both relay and producer nodes are fetching updates from the blockchain (as in passive node) but not processing transactions. Or at least this is what Guild Live View is showing.

If your issue is that gLiveView is not 'processing transactions', then you must first set TraceMempool to true in your mainnet-config.json to enable this feature.

Interestingly the producer node just minted a block that I take that will be lost on network write sync up.

cardano-node uses a 'pull' system for sharing blockchain data, so your ability to broadcast blocks depends on other nodes connecting to yours to pull data.

pool.vet suggests that you may not be well connected because you don't use topologyUpdater. Nevertheless, it reports that you should have one inbound connection from an IOHK node, and this is enough for you to publish blocks. Your netstat example doesn't show this, but if you run netstat -pant | grep cardano-node on your relay, you will see an incoming connection from IOHK.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.