Database Reference
In-Depth Information
Here we issue the
loadbalance
argument to Nodetool to start sending data from 1.5 to other
nodes:
$ bin/nodetool -host 192.168.1.5 loadbalance
This will begin the load-balancing process. While that's happening, we can check the status by
using the
streams
argument with the host we want to check out, and then make sure that our
other node is still OK:
eben@morpheus$ bin/nodetool streams -h 192.168.1.5
Mode: Leaving: streaming data to other nodes
Not sending any streams.
Not receiving any streams.
eben@morpheus$ bin/nodetool streams -h 192.168.1.7
Mode: Normal
Not sending any streams.
Not receiving any streams.
Once the load balancing is complete, we can check the logs to see what happened during the
balancing.
Here we can see that the node on 1.5 started leaving the cluster after we load balanced, sending
its data to the other node, then bootstrapped itself, and then started accepting data again. This
has the effect of load balancing.
DEBUG 10:46:58,727 Leaving: old token was 20846671262289044293293447172905883342
DEBUG 10:46:58,746 Pending ranges:
/192.168.1.7:
(41654880048427970483049687892424207188,
20846671262289044293293447172905883342]
INFO 10:46:58,746 Leaving: sleeping 30000 for pending range setup
DEBUG 10:46:59,323 Disseminating load info ...
DEBUG 10:47:28,748 Node /192.168.1.5 ranges
[(41654880048427970483049687892424207188,20846671262289044293293447172905883342]]
DEBUG 10:47:28,749 Range
(41654880048427970483049687892424207188,20846671262289044293293447172905883342]
will be responsibility of /192.168.1.7
DEBUG 10:47:28,750 Ranges needing transfer are
[(41654880048427970483049687892424207188,20846671262289044293293447172905883342]]
INFO 10:47:28,750 Leaving: streaming data to other nodes
DEBUG 10:47:28,753 Beginning transfer process to /192.168.1.7 for ranges
(41654880048427970483049687892424207188,20846671262289044293293447172905883342]
INFO 10:47:28,753 Flushing memtables for Keyspace1...
INFO 10:47:28,753 Performing anticompaction ...
