Details
-
Improvement
-
Status: Resolved
-
Low
-
Resolution: Not A Problem
-
None
-
None
Description
According to http://stackoverflow.com/questions/6832285/how-do-you-use-the-cassandra-tool-sstableloader and other references, it looks like running sstableloader in many cases necessitates duplicating (a subset of) the destination node's Cassandra install folder. This is primarily to have the right cassandra.yaml in the right place, as far as I can see.
It seems like it would be much less of a pain to use the bulk loader if it took a parameter for where to look for cassandra.yaml (or supported the individual settings it needed to gossip correctly directly on command line)
The whole business about making a duplicate loopback interface also seems a bit Byzantine.
Since the command-line tool appears just to be a fairly thin wrapper around code that is used elsewhere in Cassandra, disentangling it from the standard Cassandra configuration might be fairly difficult. But then again I'm pretty clueless so I'd be happy to be corrected.