Constantly parsing mappings on cluster restart?

Elastic Stack Elasticsearch
1

Hey all,

I have a single node cluster with a few hundred indices, with each index
broken up into a few (between 2-8) types. I attempted a restart of the
cluster, and now it's pounding on the CPU and not really indexing anything.
ES is also pouring into the logs messages of the following flavor:

parsed mapping [default], and got different sources
original:
{"default":{"_source":{"compress":true},"_size":{"enabled":true,"store":"yes"},"properties":{}}}
parsed:
{"default":{"_size":{"enabled":true,"store":"yes"},"_source":{"compress":true},"properties":{}}}

It should be the same document, except for the order? Is there something I
did wrong?

Cheers for any help you can provide!
Matt

2

Which version are you using?

On Sun, Nov 13, 2011 at 7:33 PM, Matt matt.chu@gmail.com wrote:

Hey all,

I have a single node cluster with a few hundred indices, with each index
broken up into a few (between 2-8) types. I attempted a restart of the
cluster, and now it's pounding on the CPU and not really indexing anything.
ES is also pouring into the logs messages of the following flavor:

parsed mapping [default], and got different sources
original:

{"default":{"_source":{"compress":true},"_size":{"enabled":true,"store":"yes"},"properties":{}}}
parsed:

{"default":{"_size":{"enabled":true,"store":"yes"},"_source":{"compress":true},"properties":{}}}

It should be the same document, except for the order? Is there something I
did wrong?

Cheers for any help you can provide!
Matt

3

Found the problem..., opened an issue:
Using root object level mapping (_size, _source) can cause reparsing of the mapping on startup · Issue #1458 · elastic/elasticsearch · GitHub, will be fixed
in 0.18 and master.

On Sun, Nov 13, 2011 at 7:53 PM, Shay Banon kimchy@gmail.com wrote:

Which version are you using?

On Sun, Nov 13, 2011 at 7:33 PM, Matt matt.chu@gmail.com wrote:

Hey all,

I have a single node cluster with a few hundred indices, with each index
broken up into a few (between 2-8) types. I attempted a restart of the
cluster, and now it's pounding on the CPU and not really indexing anything.
ES is also pouring into the logs messages of the following flavor:

parsed mapping [default], and got different sources
original:

{"default":{"_source":{"compress":true},"_size":{"enabled":true,"store":"yes"},"properties":{}}}
parsed:

{"default":{"_size":{"enabled":true,"store":"yes"},"_source":{"compress":true},"properties":{}}}

It should be the same document, except for the order? Is there something
I did wrong?

Cheers for any help you can provide!
Matt

4

Awesome, cheers!