1

I runned a node on testnet successfully, then I wanted to do it again but in the mainnet. I installed by following "Coincashew" guide and the developers cardano guide. I am encountering the following error and problems. First, node makes a very slow chunk validation, by my experience on testnet this is really weird, but, it might be something from mainnet?, I don't really know, this is a sample of the process:

[vmi92943:cardano.node.ChainDB:Info:7] [2022-07-10 16:07:57.49 UTC] Validated chunk no. 996 out of 2322. Progress: 42.89%
[vmi92943:cardano.node.ChainDB:Info:7] [2022-07-10 16:07:57.49 UTC] Validating chunk no. 997 out of 2322. Progress: 42.89%
[vmi92943:cardano.node.ChainDB:Info:7] [2022-07-10 16:07:58.09 UTC] Validated chunk no. 997 out of 2322. Progress: 42.94%
[vmi92943:cardano.node.ChainDB:Info:7] [2022-07-10 16:07:58.09 UTC] Validating chunk no. 998 out of 2322. Progress: 42.94%
[vmi92943:cardano.node.ChainDB:Info:7] [2022-07-10 16:07:58.75 UTC] Validated chunk no. 998 out of 2322. Progress: 42.98%
[vmi92943:cardano.node.ChainDB:Info:7] [2022-07-10 16:07:58.75 UTC] Validating chunk no. 999 out of 2322. Progress: 42.98%
[vmi92943:cardano.node.ChainDB:Info:7] [2022-07-10 16:07:59.39 UTC] Validated chunk no. 999 out of 2322. Progress: 43.02%
[vmi92943:cardano.node.ChainDB:Info:7] [2022-07-10 16:07:59.39 UTC] Validating chunk no. 1000 out of 2322. Progress: 43.02%
[vmi92943:cardano.node.ChainDB:Info:7] [2022-07-10 16:07:59.88 UTC] Validated chunk no. 1000 out of 2322. Progress: 43.07%
[vmi92943:cardano.node.ChainDB:Info:7] [2022-07-10 16:07:59.88 UTC] Validating chunk no. 1001 out of 2322. Progress: 43.07%
[vmi92943:cardano.node.ChainDB:Info:7] [2022-07-10 16:08:00.52 UTC] Validated chunk no. 1001 out of 2322. Progress: 43.11%

Right after that chunk validation the following is shown:

[vmi92943:cardano.node.ChainDB:Info:7] [2022-07-10 16:28:05.79 UTC] Found a valid last location at chunk 2322 with tip d9138d5dec83c9b1283a8122d3fa0602e81d78d9c652ca646dbe05dd6aaaf4c6@50157911.
[vmi92943:cardano.node.ChainDB:Info:7] [2022-07-10 16:28:05.79 UTC] Opened imm db with immutable tip at d9138d5dec83c9b1283a8122d3fa0602e81d78d9c652ca646dbe05dd6aaaf4c6 at slot 50157911 and chunk 2322
[vmi92943:cardano.node.ChainDB:Info:7] [2022-07-10 16:28:05.79 UTC] Started opening Volatile DB
[vmi92943:cardano.node.ChainDB:Info:7] [2022-07-10 16:28:13.88 UTC] Opened vol db
[vmi92943:cardano.node.ChainDB:Info:7] [2022-07-10 16:28:13.88 UTC] Started opening Ledger DB
[vmi92943:cardano.node.ChainDB:Error:7] [2022-07-10 16:29:23.72 UTC] Invalid Snapshot DiskSnapshot {dsNumber = 49806209, dsSuffix = Nothing}InitFailureRead (ReadFailed (DeserialiseFailure 866812015 "end of input"))

Then, the node starts replaying ledger from genesis:

[vmi92943:cardano.node.ChainDB:Info:7] [2022-07-10 16:28:13.88 UTC] Replaying ledger from genesis
[vmi92943:cardano.node.ChainDB:Info:7] [2022-07-10 16:28:14.29 UTC] Replayed block: slot 0 out of 50157911. Progress: 0.00%
[vmi92943:cardano.node.ChainDB:Info:7] [2022-07-10 16:28:17.04 UTC] Replayed block: slot 21599 out of 50157911. Progress: 0.04%
[vmi92943:cardano.node.ChainDB:Info:7] [2022-07-10 16:28:21.05 UTC] Replayed block: slot 43199 out of 50157911. Progress: 0.09%

After this the node starts syncing for a while, but all of the sudden it stops, showing the following:

The problem first showed up with cardano-node version 1.34.1 I thought that maybe updating to 1.35.1 would solve the problem, but it didn't. This output are from 1.35.1 I don't know if the first two issues posted happened in 1.34.1 because I didn't reload the node

[vmi92943:cardano.node.ChainDB:Notice:15464] [2022-07-10 20:45:26.25 UTC] Chain extended, new tip: ffc34e48133acb25d215f4e5bbbbed795439a5b6465e2f16296df0f78bc0d04b at slot 50211725
[vmi92943:cardano.node.ChainDB:Notice:15464] [2022-07-10 20:45:27.86 UTC] Chain extended, new tip: 52be37abf928860cf634c67d234403a0eec60b138fa4cbc6306dad04b12f8588 at slot 50211749
./startRelayNode1.sh: line 9: 429709 Killed                  /home/cardano/.local/bin/cardano-node run +RTS -N -A16m -qg -qb -RTS --topology ${TOPOLOGY} --database-path ${DB_PATH} --socket-path ${SOCKET_PATH} --host-addr ${HOSTADDR} --port ${PORT} --config ${CONFIG}

I runned everything again and I noticed this warning

[vmi92943:cardano.node.shutdown:Warning:7] [2022-07-10 20:44:12.01 UTC] Will terminate upon reaching NoShutdown
9
  • How much free diskspace do you have on the chain DB volume (try df -h), and how much RAM and swap space (try free -h)? The mainnet chain DB is about 70GB at the moment and 16GB of RAM is needed. Jul 10, 2022 at 21:21
  • I have enough disk space currently using arround 40 gb. I have an 150gb of storage. I have 8gb of RAM. I decided to use 8gb instead of 16gb because I read about many SPOs that are doing well with 8gb of RAM Jul 10, 2022 at 21:40
  • How much swapspace do you have? If sticking with 8GB of RAM make sure you have a good amount of swapspace allocated eg 10GB. Cardano-node at times will use more than 8GB of RAM and then use the swapspace. While your chain is syncing run top or free -h to monitor RAM and swap usage. Jul 10, 2022 at 22:08
  • I have been watching RAM usage and it never reaches te point where it need to make use of swap. I am planing on reinstalling the entire node, can you confirm to me the ideal version of cardano-node for mainnet? Jul 10, 2022 at 22:53
  • Yeah that's probably a good idea. In terms of versions I would go straight to 1.35.0. Version 1.34.1 is stable but will require a few hours of chain work when moving to 1.35.0, so you can avoid that down time. I updated my pool to 1.35.0 recently and things look stable. There will likely be a 1.35.1 but that should just be a code update, no chain/DB changes. Jul 10, 2022 at 23:30

1 Answer 1

1

Summary of comments:

System with 8GB of RAM and no swap space defined, so the process would run out of memory and terminate.

Swap space will be set up (10GB) and things should be ok.

1
  • Be aware that while swap space may allow the node to run, if its hitting swap space because there is not enough RAM, it will be slower than it if was just in RAM. For a regular node this may not be a problem, but for a block producer, the poorer performance may result in lost slot battles. Jul 11, 2022 at 8:24

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.