BitShares Forum

Main => Technical Support => Topic started by: runestone on April 05, 2018, 10:33:40 pm

Title: Running a cluster of full nodes
Post by: runestone on April 05, 2018, 10:33:40 pm

Assume you have multiple full nodes / witness nodes (e.g. 10 seperate servers, spread across the world). Can those 10 witness_node processes share the same --data-dir ("witness_node_data_dir") or will it cause some sort of file lock / race condition issue?

The reason for asking is that every time you setup a new witness node is takes many hours before the blockchain have been retrived and stored in the witness_node_data_dir.

If they different processes cannot work on the data data-dir is it then possible to copy/paste a fully loaded data-dir to another "empty" witness node?

Title: Re: Running a cluster of full nodes
Post by: Fox on April 06, 2018, 03:18:36 am

You may copy the /blockchain directory to a new host and start the witness_node to save time.