3

so I synced towards Alonzo mainnet with cabal build and interrupted at 92% and somehow ~/cardano/db/node.socket is missing. And no matter how many times I restart the node with

cardano-node run \
--config /home/user/cardano/mainnet-config.json \
--database-path /home/user/cardano/db/ \
--socket-path /home/user/cardano/db/node.socket \
--host-addr 127.0.0.1 \
--port 3001 \
--topology /home/user/cardano/mainnet-topology.json

it gets stuck at

Listening on http://127.0.0.1:12798
[hud-PS63:cardano.node.networkMagic:Notice:5] [2021-09-16 02:17:41.27 UTC] NetworkMagic 764824073
[hud-PS63:cardano.node.basicInfo.protocol:Notice:5] [2021-09-16 02:17:41.27 UTC] Byron; Shelley
[hud-PS63:cardano.node.basicInfo.version:Notice:5] [2021-09-16 02:17:41.27 UTC] 1.29.0
[hud-PS63:cardano.node.basicInfo.commit:Notice:5] [2021-09-16 02:17:41.27 UTC] 48429531f0d3d71fadce9a5971bf56a6df396f2d
[hud-PS63:cardano.node.basicInfo.nodeStartTime:Notice:5] [2021-09-16 02:17:41.27 UTC] 2021-09-16 02:17:41.275947702 UTC
[hud-PS63:cardano.node.basicInfo.systemStartTime:Notice:5] [2021-09-16 02:17:41.27 UTC] 2017-09-23 21:44:51 UTC
[hud-PS63:cardano.node.basicInfo.slotLengthByron:Notice:5] [2021-09-16 02:17:41.27 UTC] 20s
[hud-PS63:cardano.node.basicInfo.epochLengthByron:Notice:5] [2021-09-16 02:17:41.27 UTC] 21600
[hud-PS63:cardano.node.basicInfo.slotLengthShelley:Notice:5] [2021-09-16 02:17:41.27 UTC] 1s
[hud-PS63:cardano.node.basicInfo.epochLengthShelley:Notice:5] [2021-09-16 02:17:41.27 UTC] 432000
[hud-PS63:cardano.node.basicInfo.slotsPerKESPeriodShelley:Notice:5] [2021-09-16 02:17:41.27 UTC] 129600
[hud-PS63:cardano.node.basicInfo.slotLengthAllegra:Notice:5] [2021-09-16 02:17:41.27 UTC] 1s
[hud-PS63:cardano.node.basicInfo.epochLengthAllegra:Notice:5] [2021-09-16 02:17:41.27 UTC] 432000
[hud-PS63:cardano.node.basicInfo.slotsPerKESPeriodAllegra:Notice:5] [2021-09-16 02:17:41.27 UTC] 129600
[hud-PS63:cardano.node.basicInfo.slotLengthMary:Notice:5] [2021-09-16 02:17:41.27 UTC] 1s
[hud-PS63:cardano.node.basicInfo.epochLengthMary:Notice:5] [2021-09-16 02:17:41.27 UTC] 432000
[hud-PS63:cardano.node.basicInfo.slotsPerKESPeriodMary:Notice:5] [2021-09-16 02:17:41.27 UTC] 129600
[hud-PS63:cardano.node.basicInfo.slotLengthAlonzo:Notice:5] [2021-09-16 02:17:41.27 UTC] 1s
[hud-PS63:cardano.node.basicInfo.epochLengthAlonzo:Notice:5] [2021-09-16 02:17:41.27 UTC] 432000
[hud-PS63:cardano.node.basicInfo.slotsPerKESPeriodAlonzo:Notice:5] [2021-09-16 02:17:41.27 UTC] 129600
[hud-PS63:cardano.node.addresses:Notice:5] [2021-09-16 02:17:41.27 UTC] [SocketInfo 0.0.0.0:3001]
[hud-PS63:cardano.node.diffusion-mode:Notice:5] [2021-09-16 02:17:41.27 UTC] InitiatorAndResponderDiffusionMode
[hud-PS63:cardano.node.dns-producers:Notice:5] [2021-09-16 02:17:41.27 UTC] [DnsSubscriptionTarget {dstDomain = "relays-new.cardano-mainnet.iohk.io", dstPort = 3001, dstValency = 2}]
[hud-PS63:cardano.node.ip-producers:Notice:5] [2021-09-16 02:17:41.27 UTC] IPSubscriptionTarget {ispIps = [], ispValency = 0}

Is there a log to check what happened during the interruption? where is this stored (following cabal build documentation)

2
  • Please check if there is enough disk space. Node.socket will be created only when the node is running
    – raghu
    Sep 16 at 19:23
  • used about 117gb and 17 gb left. I reran, synced to >91% (database at 14 Gb), stopped the sync - checked andnode.socket= still in db folder, once I rerun cardano-node run ... node.socket= is gone. is 17gb too little?
    – Hud
    Sep 21 at 19:39
2

At this stage, it is normal that your node.socket does not yet exist because your launch has not passed the step IPSubscriptionTarget {ispIps = [], ispValency = 0}. It usually happens because your cardano-node has not been gracefully stopped.

If it continues from there after 30 min or so (assuming that you have a decent CPU with enough memory) it would be a sign of a hard shutdown. We sometimes have the impression that the program has frozen but it is because your cardano-node needs to scan and verify all the data again so it can take a while.

When you see a line like

Opened vol db

your node.socket will be here again.

1
  • 1
    Yes there has been some issues where my PC shuts down unexpectedly. But waiting for the node.socket to reappear just needs time!
    – Hud
    Sep 26 at 14:37
0

Try exporting the CARDANO_NODE_SOCKET_PATH again. An exported variable will only remain alive while the user who issued the command is active in their session. You can add this also to a startup script in that user's space.

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.