docker-library
/
docs
mirror of https://github.com/docker-library/docs.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Merge pull request #798 from joelwurtz/patch-2 

Explain cluster not supported for elasticsearch
This commit is contained in:
Tianon Gravi 2017-01-13 12:29:24 -08:00 committed by GitHub
parent 14d8d20482 b8960fd741
commit 1127b378f5
1 changed files with 12 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
elasticsearch/content.md
View File

@ -10,18 +10,26 @@ Elasticsearch is a registered trademark of Elasticsearch BV.
# How to use this image
## Host Setup
## Cluster
**Note:** since 5.0, Elasticsearch only listens on `localhost` by default, so this image sets `network.host` to `0.0.0.0` (given that `localhost` is not terribly useful in the Docker context).
**Note:** since 5.0, Elasticsearch only listens on `localhost` by default on both http and transport, so this image sets `http.host` to `0.0.0.0` (given that `localhost` is not terribly useful in the Docker context).
As a result, Elasticsearch is more strict about the bootstrap checks that it performs, especially when checking the value of `vm.max_map_count` which is not namespaced and thus must be set to an acceptable value on the host (as opposed to simply using `--sysctl` on `docker run`).
As a result, this image does not support clustering out of the box and extra configuration must be set in order to support it.
Supporting clustering implies having Elasticsearch in a production mode which is more strict about the bootstrap checks that it performs, especially when checking the value of `vm.max_map_count` which is not namespaced and thus must be set to an acceptable value on the host (as opposed to simply using `--sysctl` on `docker run`).
One example of adding clustering support is to pass the configuration on the `docker run`:
```console
$ docker run -d --name elas elasticsearch -Etransport.host=0.0.0.0 -Ediscovery.zen.minimum_master_nodes=1
```
See the following sections of the upstream documentation for more information:
- [Setup Elasticsearch » Important System Configuration » Virtual memory](https://www.elastic.co/guide/en/elasticsearch/reference/5.0/vm-max-map-count.html)
- [Setup Elasticsearch » Bootstrap Checks » Maximum map count check](https://www.elastic.co/guide/en/elasticsearch/reference/5.0/_maximum_map_count_check.html)
This [comment in elastic/elasticsearch#4978](https://github.com/elastic/elasticsearch/issues/4978#issuecomment-258676104) shows why this change was added in upstream. Production mode is when listening on non-localhost.
This [comment in elastic/elasticsearch#4978](https://github.com/elastic/elasticsearch/issues/4978#issuecomment-258676104) shows why this change was added in upstream.
> Elasticsearch will not start in production mode if `vm.max_map_count` is not high enough. [...] If the value on your system is NOT high enough, then your cluster is going to crash and burn at some stage and you will lose data.