Grant, Thanks for the info, that makes total sense.
On Tue, May 3, 2011 at 10:52 AM, Grant Schofield <[email protected]> wrote: > On May 3, 2011, at 12:35 PM, Jeff Pollard wrote: > > Grant, > > Thanks for the reply. I think I understand what you're saying, but it's > not 100% clear to me how exactly it applies to my question. I'll try > explaining my plan another way if that helps. > > Imagine we have 4 nodes running in a ring on EC2: [email protected], > [email protected], [email protected], [email protected]. Then one day node > [email protected] goes down and is unrecoverable. Our plan had been: > > # On a existing node in the cluster > riak-admin remove [email protected] > riak-admin ringready > # .. wait for TRUE > > # Boot a new EC2 machine configured as [email protected] (brand new IP) > # Then on that new machine... > riak start > riak-admin join [email protected] > riak-admin ringready > # .. wait for TRUE > > In that scenario, I imagine the issue is that removing and adding a new > node under a new node name is putting extra strain on the cluster to shift > data around? Simply replacing a new node under the same hostname (even > though it has a different IP) would mean less work for the cluster? And how > much extra work are > > > You're correct in assuming that there is extra strain on the cluster moving > that data around, but the amount of that churn will depend on how much data > you have in your Riak cluster and your use case. Removing and adding a node > is a great way to replace a broken node, but if you have a large amount of > data it might be quicker and load the cluster less to replace the broken > node with a node that has the same hostname or IP and the old nodes data > directory. > > If [email protected] dies due to a disk problem and your backups won't contain > the data you would need to replace the node as you described. The best > solution for replacing a dead node will vary based on circumstances, your > use case, and your availability needs. > > Grant > > > > Thanks! > > On Tue, May 3, 2011 at 6:24 AM, Grant Schofield <[email protected]> wrote: > >> >> On May 2, 2011, at 7:58 PM, Jeff Pollard wrote: >> >> I was reviewing the Riak Operations >> webinar<http://blog.basho.com/2011/04/15/follow-up-to-riak-operations-webinar/>, >> and it was mentioned that the preferred vm.args -name for EC2 environments >> should be "riak@hostname" because you don't have to "rename data or do >> anything weird" like you would if your nodes were named "[email protected]" >> (approximately 40:05 in the video). >> >> I was looking for some elaboration on this tip, namely: >> >> 1. What is meant by "rename data or do anything weird" >> >> When you bring a cluster together using data copied from a different set >> of nodes you need to re-ip the first node you plan to start but you also >> have to change the ring manually on that node so when the subsequent nodes >> join everything works properly. >> >> >> 1. Is "hostname" in riak@hostname a public DNS host that you configure >> in your DNS to map to the EC2 public hostname (ec2-50-18-...)? >> >> You can use DNS (public or private) or host file entries that reference >> the private IP of the node if you choose to use hostname. >> >> >> 1. Does anyone have any best practices around vm.args -name in EC2 >> environments? >> >> We haven't outlined any best practices ourselves, but I tend to believe >> using a hostname that you can change the IP for via DNS or a hosts file is a >> more flexible way of approaching the problem. >> >> Grant >> >> >> Thanks! >> _______________________________________________ >> riak-users mailing list >> [email protected] >> http://lists.basho.com/mailman/listinfo/riak-users_lists.basho.com >> >> >> > >
_______________________________________________ riak-users mailing list [email protected] http://lists.basho.com/mailman/listinfo/riak-users_lists.basho.com
