Message not fully read (response) for / failed to send shard started to

Elastic Stack Elasticsearch
1

Hi,

Today I saw these errors in the logs:

Message not fully read (response) for (master node)
failed to send shard started to (data)

I've checked that all nodes are on the same version (0.19.12) and all using
java 1.6.0_32. I'm using unicast discovery.

Any clues/ideas? We are running a different version of ES (0.18 something)
on the same servers but they have a different cluster-name, so they are not
part of our 0.19.12 cluster...could the cluster names get mixed up somehow?

-T

--
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.
For more options, visit https://groups.google.com/groups/opt_out.

2

forgot to say...restarting the data node fixed the issue...but I'd like to
get to the bottom of this to make sure we're stable in production.
thanks
-T

On Tuesday, March 12, 2013 7:42:18 PM UTC-7, Tinou Bao wrote:

Hi,

Today I saw these errors in the logs:

Message not fully read (response) for (master node)
failed to send shard started to (data)

I've checked that all nodes are on the same version (0.19.12) and all
using java 1.6.0_32. I'm using unicast discovery.

Any clues/ideas? We are running a different version of ES (0.18 something)
on the same servers but they have a different cluster-name, so they are not
part of our 0.19.12 cluster...could the cluster names get mixed up somehow?

-T

--
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.
For more options, visit https://groups.google.com/groups/opt_out.

3

If you are mixing versions on the same network with multicast, each node has to communicate to each other to see if they belong to the same cluster.
But, they can not talk to each other if they don't use the same major version.

That's the reason of this message.

HTH

--
David :wink:
Twitter : @dadoonet / @elasticsearchfr / @scrutmydocs

Le 13 mars 2013 à 03:42, Tinou Bao tinou.bao@gmail.com a écrit :

Hi,

Today I saw these errors in the logs:

Message not fully read (response) for (master node)
failed to send shard started to (data)

I've checked that all nodes are on the same version (0.19.12) and all using java 1.6.0_32. I'm using unicast discovery.

Any clues/ideas? We are running a different version of ES (0.18 something) on the same servers but they have a different cluster-name, so they are not part of our 0.19.12 cluster...could the cluster names get mixed up somehow?

-T

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

--
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.
For more options, visit https://groups.google.com/groups/opt_out.