Broken cluster


#1

Hi,

I had a 2.0.1 enterprise edition cluster configuration on AWS Ec2.
Secondary node of my cluster was scheduled for retirement and I had to reboot it to retrieve the instance.
Now my IP of the instance is changed after a stop and start action on AWS.
When I am trying to start the couchbase server it’s giving me error
[ns_server:error,2013-07-22T15:36:21.539,nonode@nohost:dist_manager<0.6430.0>:dist_manager:init:148]Configured address 10.151.11.204 seems to be invalid. Will refuse to to start for safety reasons.

Please help to retrieve the data and cluster.

Thanks


#2

Sounds like the ipaddress of the machine has changed from the one you installed it with.
So if you did not assign it a Elastic Ip when you first installed it probably still think its the old one.


#3

Yes, I didn’t assign the Elastic IP when I installed it. After a stop and start action the private IP of the instance changed as AWS provides DHCP IP to it’s instances. Now I am not able to start the cluster and I have to backup the data with the cluster. How to do this? Please help.


#4

you MIGHT be able to take a copy of you data files.
/opt/couchbase/var/lib/couchbase/data/* and move them to another machine.

CB docs sorta talks about it. but they pretty much say “Delete Everything” and start over.
http://www.couchbase.com/docs/couchbase-manual-2.0/couchbase-bestpractice-cloud-ip.html