Upgrade introducing new nodes

Hi,

Could you please check if my upgrade scenario given below will work
properly? I have three nodes in cluster (ES 1.3, no replication) and I want
to upgrade it to ES 1.4.

I wonder can I do the following:

  1. Create three new nodes running ES 1.4
  2. Add them to cluster
  3. Configure replication "number_of_replicas": 1
    4)Make sure replicas goes to new nodes by invoking curl -XPOST -d '{
    "commands" : [ { "allocate" ... etc.
  4. When all repicas are moved, enable one of new servers as master
  5. Disable the old master
  6. Stop old servers

Will this scenario work? Would it be the same to create snapshot and
restore it on new servers as new cluster?

The main point of why I'm not following official guide is: I want to be
able to come back quickly to previous configuration.

Chris

--
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/ee26c9da-187f-402c-b5b6-b2e8bdbc2699%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Yes but once you move a shard to a newer node then (usually) you can't
shift it back. This is due to changes in the underlying lucene segment.

You'll also need to make sure you have the same Java versions on all nodes
it you'll see problems.

But yes, snapshot and restore would be a better way to do this as you
maintain your previous state completely segregated.
On 12/01/2015 2:58 pm, "Krzysztof Madejski" krzysztof.madejski@epf.org.pl
wrote:

Hi,

Could you please check if my upgrade scenario given below will work
properly? I have three nodes in cluster (ES 1.3, no replication) and I want
to upgrade it to ES 1.4.

I wonder can I do the following:

  1. Create three new nodes running ES 1.4
  2. Add them to cluster
  3. Configure replication "number_of_replicas": 1
    4)Make sure replicas goes to new nodes by invoking curl -XPOST -d '{
    "commands" : [ { "allocate" ... etc.
  4. When all repicas are moved, enable one of new servers as master
  5. Disable the old master
  6. Stop old servers

Will this scenario work? Would it be the same to create snapshot and
restore it on new servers as new cluster?

The main point of why I'm not following official guide is: I want to be
able to come back quickly to previous configuration.

Chris

--
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/ee26c9da-187f-402c-b5b6-b2e8bdbc2699%40googlegroups.com
https://groups.google.com/d/msgid/elasticsearch/ee26c9da-187f-402c-b5b6-b2e8bdbc2699%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/CAEYi1X9yqCgJueGOTi3KfZT5U00PuLMJHn-c_7DNuGAfNtfSFg%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.

Thank you Mark for help!

I guess I'll go with replication as it doesn't require to stop indexing new
documents (and while my index is quite big it can take some time).

Chris

On Mon, Jan 12, 2015 at 3:51 AM, Mark Walkom markwalkom@gmail.com wrote:

Yes but once you move a shard to a newer node then (usually) you can't
shift it back. This is due to changes in the underlying lucene segment.

You'll also need to make sure you have the same Java versions on all nodes
it you'll see problems.

But yes, snapshot and restore would be a better way to do this as you
maintain your previous state completely segregated.
On 12/01/2015 2:58 pm, "Krzysztof Madejski" <
krzysztof.madejski@epf.org.pl> wrote:

Hi,

Could you please check if my upgrade scenario given below will work
properly? I have three nodes in cluster (ES 1.3, no replication) and I want
to upgrade it to ES 1.4.

I wonder can I do the following:

  1. Create three new nodes running ES 1.4
  2. Add them to cluster
  3. Configure replication "number_of_replicas": 1
    4)Make sure replicas goes to new nodes by invoking curl -XPOST -d '{
    "commands" : [ { "allocate" ... etc.
  4. When all repicas are moved, enable one of new servers as master
  5. Disable the old master
  6. Stop old servers

Will this scenario work? Would it be the same to create snapshot and
restore it on new servers as new cluster?

The main point of why I'm not following official guide is: I want to be
able to come back quickly to previous configuration.

Chris

--
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/ee26c9da-187f-402c-b5b6-b2e8bdbc2699%40googlegroups.com
https://groups.google.com/d/msgid/elasticsearch/ee26c9da-187f-402c-b5b6-b2e8bdbc2699%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 a topic in the
Google Groups "elasticsearch" group.
To unsubscribe from this topic, visit
https://groups.google.com/d/topic/elasticsearch/_hHPmrqNXEo/unsubscribe.
To unsubscribe from this group and all its topics, send an email to
elasticsearch+unsubscribe@googlegroups.com.
To view this discussion on the web visit
https://groups.google.com/d/msgid/elasticsearch/CAEYi1X9yqCgJueGOTi3KfZT5U00PuLMJHn-c_7DNuGAfNtfSFg%40mail.gmail.com
https://groups.google.com/d/msgid/elasticsearch/CAEYi1X9yqCgJueGOTi3KfZT5U00PuLMJHn-c_7DNuGAfNtfSFg%40mail.gmail.com?utm_medium=email&utm_source=footer
.

For more options, visit https://groups.google.com/d/optout.

--
Krzysztof Madejski
Koordynator Koduj Dla Polski, Fundacja ePaństwo
Code for Poland program manager, ePF Foundation

krzysztof.madejski@epf.org.pl
+48 664 082 823

--
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/CAHd0-5eDvL%3DRoEfhoiuxMq0ovc4Z-eNVq9dLmAtdnJF%3DBYp_nw%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.