Installation details
#ScyllaDB version: 6.1
#Cluster size: 3 x 2
os (RHEL/CentOS/Ubuntu/AWS AMI): Debian
Me again. After finally found a solution to massive insertion (close to the theoritical 10k rps per core) I found something else.
Boostraping a new cluster : 3 nodes on 2 two dc (6 in total) at once I hit this on random node :
Startup failed: std::runtime_error (A node with address X.X.X.X already exists, cancelling join. Use replace_address if you want to replace this node.)
It was a previous attempt of startup failling with
shard 0:main] init - Startup failed: seastar::rpc::remote_verb_error (called make_nonvoters before Raft upgrade finished
I tried disabling raft (but how?). And btw once a node is in this situation I’m stuck (all I can do is changing IP). The node is nowhere except in gossip. I tried to follow the procedure to delete node or fix raft without success …
Annoying