Shulou(Shulou.com)05/31 Report--

This article will explain in detail the example analysis of upgrading and restarting the full cluster version of the Elasticsearch search server. The content of the article is of high quality, so the editor will share it with you for reference. I hope you will have a certain understanding of the relevant knowledge after reading this article.

The Elasticsearch search server requires a full cluster restart, and the main version upgrade is from 0.x to 1.x or from 2.x version of 1.x to rolling version support.

This process requires a complete cluster restart to upgrade, as shown below:

Step 1: turn off automatic allocation of fragments

When you shut down a node, the configuration process immediately attempts to replicate shards on other nodes in the original node cluster, resulting in a lot of I / O waste. You can avoid this by disabling allocation before closing a node:

123456PUT / _ cluster/settings {"persistent": {"cluster.routing.allocation.enable": "none"}}

If you upgrade from 0.90.x to 1.x, use this setting: no, but as follows:

1234567PUT / _ cluster/settings {"persistent": {"cluster.routing.allocation.disable_allocation": true, "cluster.routing.allocation.enable": "none"}}

Step 2: perform a synchronous refresh to brush the data in memory to the hard disk

1POST / _ flush/synced

Step 3: close the old cluster and configure the new cluster

Stop all Elasticsearch services on all nodes in the cluster, and each node can be upgraded according to the same procedure described.

Step 4: start the cluster and use API to check that all nodes have successfully joined the cluster

12GET _ cat / health GET _ cat / nodes

Step 5: wait for the cluster status to turn yellow

Once each node has restored the local, once each node has restored the local shard, the status will turn yellow, which means that all major shards have been recovered, but not all replica shards have been allocated. This can be expected because allocation is still disabled. The status turns yellow, which means that all major shards have been recovered, but not all copies have been allocated. This is to be expected because allocation is still disabled.

Step 6: re-enable allocation

Delay the allocation of replicas until all nodes join the cluster and allow master replicas to be assigned to nodes that already have local partial replicas, and re-enable shard allocation:

123456PUT / _ cluster/settings {"persistent": {"cluster.routing.allocation.enable": "all"}}

If you upgrade from 0.90.x to 1.x, use this setting: no, but as follows:

1234567PUT / _ cluster/settings {"persistent": {"cluster.routing.allocation.disable_allocation": false, "cluster.routing.allocation.enable": "all"}}

From now on, all data nodes on which the cluster allocates replica shards are securely indexed and searched.

So much for the example analysis of the full cluster upgrade and restart of the Elasticsearch search server. I hope the above content can be of some help and learn more. If you think the article is good, you can share it for more people to see.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.