Full Synchronisation with Geth Repeated Tree Node Error

Question

I am doing a university project on the functionality of the EVM. I need to use debug.traceTransaction() function which requires a fully synchronized node. I have tried to sync Geth on a university VM with the following:

geth console --cache=2048

The node is quick to synch (good internet speed, 8GB of RAM and an SSD). But each time I test it by running debug.traceTransaction() from the console, it gives me the "missing trie node" error.

I have resynched fully 4 times. There is nothing else running on the machine at the same time, so resouces should not be a problem. I am perhaps getting bad info from a nearby peer? There are a few VM's running ethereum nodes within the campus network.. could this be the problem? Is there a way to specify only nodes of a particular IP address range should be synched with?

Answer 1

After resynchronising many times, I finally got a complete state history by setting cache=1024, despite the fact the VM had 8GB of RAM, it was crashing if set at 2GB and nothing else running on machine!

Answer 2

It does not look like you are running Geth with the appropriate APIs enabled. This is outlined here.

geth --cache=2048 --rpc --rpcapi "eth,net,web3,debug" console