You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@cassandra.apache.org by Philippe <wa...@gmail.com> on 2011/11/10 00:07:19 UTC
Moving experiences ?
Hello,
I am going to need to move some nodes to rebalance my cluster. How safe is
this to do on a cluster with writes & reads ?
Thanks
Re: Moving experiences ?
Posted by Radim Kolar <hs...@sendmail.cz>.
Dne 10.11.2011 8:16, Maki Watanabe napsal(a):
> I missed the news.
> How the "nodetool move" work in recent version (0.8.x or later?)
> Just stream appropriate range of data between nodes?
yes
Re: Moving experiences ?
Posted by Maki Watanabe <wa...@gmail.com>.
I missed the news.
How the "nodetool move" work in recent version (0.8.x or later?)
Just stream appropriate range of data between nodes?
2011/11/10 Peter Schuller <pe...@infidyne.com>:
> Keep in mind that if you're using an older version of Cassandra a move
> is actually a decommission followed by bootstrap - so neighboring
> nodes will temporarily own a larger part of the ring while a node is
> being moved.
--
w3m
Re: Moving experiences ?
Posted by Peter Schuller <pe...@infidyne.com>.
> I am going to need to move some nodes to rebalance my cluster. How safe is
> this to do on a cluster with writes & reads ?
Moving nodes while serving live traffic is supported. Of course, there
is always some inherent risk in making ring topology changes (make
sure you know the process), but it is supported.
Keep in mind that if you're using an older version of Cassandra a move
is actually a decommission followed by bootstrap - so neighboring
nodes will temporarily own a larger part of the ring while a node is
being moved.
--
/ Peter Schuller (@scode, http://worldmodscode.wordpress.com)
Re: changing ownership when nodes have same token
Posted by hani elabed <ha...@gmail.com>.
I have had this happen to me when I copied the installation from one node
to the other( 2 Virtual machines with same configuration). I had already
started the first node with the default setup, i.e a single node without
providing a seed IP address. See step 3 in here
http://wiki.apache.org/cassandra/GettingStarted
what I did is to delete everything cassandra from both nodes, and start
from scratch and provide a seed IP, look at config/cassandra.yaml and
follow Step 3 above, to the letter, i.e provide ListenAddress and
ThriftAddress...
To delete the old cassandra install do something like this (ruby code
below, change to 1.0.2 or your version);
if File.exists? "/home/vagrant/apache-cassandra-1.0.0"
execute_command "rm -rf /home/vagrant/apache-cassandra-1.0.0"
end
if File.exists? "/var/log/cassandra"
execute_command "rm -rf /var/log/cassandra"
end
if File.exists? "/var/lib/cassandra"
execute_command "rm -rf /var/lib/cassandra"
end
Good luck Feng
Hani
On Thu, Nov 10, 2011 at 4:31 PM, Feng Qu <ma...@gmail.com> wrote:
> Hello,
>
> I notice that when starting a new node with same configuration(cluster
> name, seeds, token etc) as an existing ring member, the new node will take
> over the ownership from existing ring member. Is this expected behavior? I
> would like to see Cassandra prevents new node from joining the ring as in
> this case new node is mis-configured.
>
> Feng
>
changing ownership when nodes have same token
Posted by Feng Qu <ma...@gmail.com>.
Hello,
I notice that when starting a new node with same configuration(cluster name, seeds, token etc) as an existing ring member, the new node will take over the ownership from existing ring member. Is this expected behavior? I would like to see Cassandra prevents new node from joining the ring as in this case new node is mis-configured.
Feng
Re: Moving experiences ?
Posted by "Rodrigo K. Ferreira" <rk...@gmail.com>.
I did some test about that in the past months, and its safe if you have a
high replication factor on the cluster and high read consistency on the
clients.
But, if you have a big amount of data, it will get much time to rebalance
the nodes.
On Wed, Nov 9, 2011 at 9:07 PM, Philippe <wa...@gmail.com> wrote:
> Hello,
> I am going to need to move some nodes to rebalance my cluster. How safe is
> this to do on a cluster with writes & reads ?
>
> Thanks
>