I am planning on syncing my geth node with Ropsten using Ubuntu on AWS. From my pains I've experienced so far I'm uncovering that I'll need : 8 cores, 64GB RAM, 100Mbps Internet connection, 300GB SSD drive and that I should run a cache size --cache=8192

Let me know if that seems correct or please let me know what is minimum to get a full sync using fast mode.

It took me 3 days to get all the blocks minus the last 65 of them using Ubuntu 16, 1gb ram, 2 gb swap file, 32gb drive, Download: 302.30 Mbit/s, Upload: 232.94 Mbit/s

1 Answer 1


I am using standard $10 DigitalOcean machine 2 GB memory, 1 vCPU core, 50 GB SSD, 2 TB transfer. Only thing I fixed is adding at least 100 Gb storage. Ropsten network was synced in 20-24 hours if fast mode.

  • 1
    thats actually great news, mine is still not synced and is going on day 4. Would you mind sharing your launch commands? I'm using geth --testnet --syncmode "fast" --ipcpath /home/ubuntu/.ethereum/geth.ipc --rpc --rpcapi db,eth,net,web3,personal --cache=2048 --rpcport 8545 --rpcaddr --rpccorsdomain "*" --rpcapi "db,eth,net,web3,personal" --port "30303" --networkid 3
    – JJacquet
    Commented Jun 18, 2018 at 15:15
  • 1
    First of all I cleaned up database from past experiments and used this command in supervisord config: command=geth --testnet --networkid=3 --fast --port 30303 --datadir=/var/lib/ethereum/ropsten --rpc --rpcport=18545 --rpccorsdomain "*" --rpcaddr= --rpcapi="db,eth,net,web3,personal,miner"
    – Alex G.P.
    Commented Jun 18, 2018 at 15:19
  • 1
    I'm so nervous about using removeDB because it took me so many days to get here.
    – JJacquet
    Commented Jun 18, 2018 at 15:21
  • 2
    @JJacquet I suppose you can just move database to another folder.
    – Alex G.P.
    Commented Jun 18, 2018 at 16:23

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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