How to upgrade ES from 0.90.2 to 1.1.1


(nie_kl) #1

Hi everyone:
I used ES-0.90.2 as my index server.I have three pcs and each pc
has a ES instance.They run as a cluster.And they worked well,but now I want
upgrade my ES from 0.90.2 to 1.1.1.
At first,I read the document on the elasticsearch.org.I shutdown
my ES0.90.2,and unzip the ES1.1.1 in each pc. Then I copy the 'data' folder
in the ES0.90.2 to the same dictionary in ES1.1.1.At last,i run the three
ES1.1.1.But the shards became unassigned.Why,and how could i upgrade
rightly.
In my ES0.90.2,there is one index and more more than 30 types.I
creat 20 shards and 2 replica in my index.So there are 60 shards in total
and there are 20 shards in each pc.
When I start ES1.1.1,it seemed can find the meta infomation of the
old index data.Because,ES1.1.1 could know the name of old index data and
know how many shards the old index data has.
the status of my ES1.1.1 is:
{
"cluster_name" : "test",
"status" : "red",
"timed_out" : false,
"number_of_nodes" : 1,
"number_of_data_nodes" : 1,
"active_primary_shards" : 0,
"active_shards" : 0,
"relocating_shards" : 0,
"initializing_shards" : 0,
"unassigned_shards" : 60
}

--
You received this message because you are subscribed to the Google Groups "elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/d0e14772-3076-4130-8d49-66ba02a1aea0%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.


(Ivan Brusic) #2

How many nodes do you have? You mentioned that you unzip the distribution
on each PC, but you status only shows 1 node. Can you output the status for
each node? Did you copy the configs as well?

First of all, there is no need to copy the data unless you want a backup.
You can maintain the same configs and data by pointing elasticsearch to the
some directory that lives outside of the installation directory:
http://www.elasticsearch.org/guide/en/elasticsearch/reference/current/setup-dir-layout.html

Cheers,

Ivan

On Fri, May 16, 2014 at 4:01 AM, nie_kl@126.com wrote:

Hi everyone:
I used ES-0.90.2 as my index server.I have three pcs and each pc
has a ES instance.They run as a cluster.And they worked well,but now I want
upgrade my ES from 0.90.2 to 1.1.1.
At first,I read the document on the elasticsearch.org.I shutdown
my ES0.90.2,and unzip the ES1.1.1 in each pc. Then I copy the 'data' folder
in the ES0.90.2 to the same dictionary in ES1.1.1.At last,i run the three
ES1.1.1.But the shards became unassigned.Why,and how could i upgrade
rightly.
In my ES0.90.2,there is one index and more more than 30 types.I
creat 20 shards and 2 replica in my index.So there are 60 shards in total
and there are 20 shards in each pc.
When I start ES1.1.1,it seemed can find the meta infomation of the
old index data.Because,ES1.1.1 could know the name of old index data and
know how many shards the old index data has.
the status of my ES1.1.1 is:
{
"cluster_name" : "test",
"status" : "red",
"timed_out" : false,
"number_of_nodes" : 1,
"number_of_data_nodes" : 1,
"active_primary_shards" : 0,
"active_shards" : 0,
"relocating_shards" : 0,
"initializing_shards" : 0,
"unassigned_shards" : 60
}

--
You received this message because you are subscribed to the Google Groups
"elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an
email to elasticsearch+unsubscribe@googlegroups.com.
To view this discussion on the web visit
https://groups.google.com/d/msgid/elasticsearch/d0e14772-3076-4130-8d49-66ba02a1aea0%40googlegroups.comhttps://groups.google.com/d/msgid/elasticsearch/d0e14772-3076-4130-8d49-66ba02a1aea0%40googlegroups.com?utm_medium=email&utm_source=footer
.
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups "elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an email to elasticsearch+unsubscribe@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/CALY%3DcQB1HPpx2Om%2BHvW-%2B65ytZJ1jdvoDzyZsPogXpE4v2jm1A%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.


(Ayush Sangani) #3

Hi Ivan,

I could use your expertise in elasticsearch here:

I am upgrading from elasticsearch 0.90 to 1.1.1.

The process I am following for upgradation in staging is:

  1. I have new system in which I have installed elasticsearch 1.1.1 with same config as 0.90 just the cluster-name is different.

  2. I have copied the nodes folder inside data folder to same location at elasticsearch 1.1.1.

  3. Restarted the 1.1.1 elasticsearch service.

Problem: I am not able to see the indices present in 1.1.1 but I see indices present in the data folder.

Am I missing any step?

Please let me know if you can help me with this issue.

-Thanks


(system) #4