Red Cluster!

摘自:http://blog.kiyanpro.com/2016/03/06/elasticsearch/reroute-unassigned-shards/

There are 3 cluster states:

  1. green: All primary and replica shards are active
  2. yellow: All primary shards are active, but not all replica shards are active
  3. red: Not all primary shards are active

When cluster health is red, it means cluster is dead. And that means you can do nothing until it’s recovered, which is very bad indeed. I will share with you how to deal with one common situation: when cluster is red due to unassigned shards.

Steps

The general idea is pretty simple: find those shards which are unassigned, manually assign them to a node with reroute API. Let’s see how we can do that step by step. Then we can combine them into a configurable simple script.

Step 1: Check Unassigned Shards

To get cluster information, we usually use cat APIs. There is a GET /_cat/shards endpoint to show a detailed view of what nodes contain which shards[1].

Cat shards

1
2
3
4
5
6
7
8
9
# cat shards verbose
curl "http://your.elasticsearch.host.com:9200/_cat/shards?v"
 
# cat shards index
curl "http://your.elasticsearch.host.com:9200/_cat/shards/wiki2"
# example return
# wiki2 0 p STARTED 197 3.2mb 192.168.56.10 Stiletto
# wiki2 1 p STARTED 205 5.9mb 192.168.56.30 Frankie Raye
# wiki2 2 p STARTED 275 7.8mb 192.168.56.20 Commander Kraken

By piping cat shards to fgrep, we can get all unassigned shards.

Get unassigned shards

1
2
3
4
5
6
# cat shards with fgrep
curl "http://your.elasticsearch.host.com:9200/_cat/shards" | fgrep UNASSIGNED
# example return
# wiki1 0 r UNASSIGNED ALLOCATION_FAILED
# wiki1 1 r UNASSIGNED ALLOCATION_FAILED
# wiki1 2 r UNASSIGNED ALLOCATION_FAILED

If you don’t want to deal with shell script, you can also find these unassigned shards using another endpoint POST /_flush/synced[2]. This endpoint is actually not just some information. It allows an administrator to initiate a synced flush manually. This can be particularly useful for a planned (rolling) cluster restart where you can stop indexing and don’t want to wait the default 5 minutes for idle indices to be sync-flushed automatically. It returns with a json response.

_flush/synced

1
curl -XPOST "http://your.elasticsearch.host.com:9200/twitter/_flush/synced"

If there are failed shards in the response, we can iterate through a failures array to get all unassigned ones.

Example response with failed shards

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
{
"_shards": {
"total": 4,
"successful": 1,
"failed": 1
},
"twitter": {
"total": 4,
"successful": 3,
"failed": 1,
"failures": [
{
"shard": 1,
"reason": "unexpected error",
"routing": {
"state": "STARTED",
"primary": false,
"node": "SZNr2J_ORxKTLUCydGX4zA",
"relocating_node": null,
"shard": 1,
"index": "twitter"
}
}
]
}
}

Step 2: Reroute

The reroute command allows to explicitly execute a cluster reroute allocation command including specific commands[3] . An unassigned shard can be explicitly allocated on a specific node.

Reroute example

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
curl -XPOST 'localhost:9200/_cluster/reroute' -d '{
"commands" : [ {
"move" :
{
"index" : "test", "shard" : 0,
"from_node" : "node1", "to_node" : "node2"
}
},
{
"allocate" : {
"index" : "test", "shard" : 1, "node" : "node3"
}
}
]
}'

There are 3 kinds of commands you can use:

move: Move a started shard from one node to another node. Accepts index and shard for index name and shard number, from_node for the node to move the shard from, and to_node for the node to move the shard to.

cancel: Cancel allocation of a shard (or recovery). Accepts index and shard for index name and shard number, and node for the node to cancel the shard allocation on. It also accepts allow_primary flag to explicitly specify that it is allowed to cancel allocation for a primary shard. This can be used to force resynchronization of existing replicas from the primary shard by cancelling them and allowing them to be reinitialized through the standard reallocation process.

allocate: Allocate an unassigned shard to a node. Accepts the index and shard for index name and shard number, and node to allocate the shard to. It also accepts allow_primary flag to explicitly specify that it is allowed to explicitly allocate a primary shard (might result in data loss).

Combining step 2 with the unassigned shards from Step 1, we can reroute all unassigned shards 1 by 1, thus getting faster cluster recovery from red state.

Example Solutions

Python

Below is a python script I wrote using POST /_flush/synced and POST /reroute

Shell Script

Below is a shell script I found elsewhere in a blog post[4]

1
2
3
4
5
6
7
8
9
10
11
12
13
14
for shard in $(curl -XGET http://localhost:9200/_cat/shards | grep UNASSIGNED | awk '{print $2}'); do
curl -XPOST 'localhost:9200/_cluster/reroute' -d '{
"commands" : [ {
"allocate" : {
"index" : "t37", # index name
"shard" : $shard,
"node" : "datanode15", # node name
"allow_primary" : true
}
}
]
}'
sleep 5
done

EDIT: Based on Vincent’s comment I updated the shell script:

Possible Unassigned Shard Reasons

FYI, these are the possible reasons for a shard be in a unassigned state[1]:

Name Comment
INDEX_CREATED Unassigned as a result of an API creation of an index
CLUSTER_RECOVERED Unassigned as a result of a full cluster recovery
INDEX_REOPENED Unassigned as a result of opening a closed index
DANGLING_INDEX_IMPORTED Unassigned as a result of importing a dangling index
NEW_INDEX_RESTORED Unassigned as a result of restoring into a new index
EXISTING_INDEX_RESTORED Unassigned as a result of restoring into a closed index
REPLICA_ADDED Unassigned as a result of explicit addition of a replica
ALLOCATION_FAILED Unassigned as a result of a failed allocation of the shard
NODE_LEFT Unassigned as a result of the node hosting it leaving the cluster
REROUTE_CANCELLED Unassigned as a result of explicit cancel reroute command
REINITIALIZED When a shard moves from started back to initializing, for example, with shadow replicas
REALLOCATED_REPLICA A better replica location is identified and causes the existing replica allocation to be cancelled

References

  1. ElasticSearch Document Cat Shards
  2. ElasticSearch Document Synced Flush
  3. ElasticSearch Document Cluster Reroute
  4. How to fix your elasticsearch cluster stuck in initializing shards mode?

Reroute Unassigned Shards——遇到主shard 出现的解决方法就是重新路由的更多相关文章

  1. 运行jar提示“没有主清单属性”的解决方法

    以下记录的是我export jar包后运行遇到问题的解决方法,如有错误,欢迎批评指正. 1.运行导出jar包,提示"没有主清单属性" 2.回想自己导出jar的操作是否有误,重新ex ...

  2. mysql插入报主键冲突,解决方法主键索引重新排序

    1.备份表结构 create table table_bak like table_name; 2.备份表数据 insert into table_bak select * from table_na ...

  3. JAVA之中出现无法加载主类的情况解决方法

    j今天打代码的时候出现了无法加载主类的情况,我就收集了一些,java无法加载主类的方法 ava无法加载主类解决办法 今天启动项目,又遇到找不到或无法加载主类的情况,清除项目后无法编译,class文件下 ...

  4. eclipse 导出jar 没有主清单属性的解决方法

    eclipse编写导出的jar文件,运行出现了没有主清单属性,问题在哪里呢?有下面几种方法: 1. 导出jar文件的时候选择[可运行的jar文件]而不是[Jar文件]即可,如下图: 2. 在jar文件 ...

  5. KETTLE 主键不唯一解决方法

    SELECT 某一列, COUNT( 某一列 ) FROM 表 GROUP BY 某一列 HAVING

  6. How to resolve unassigned shards in Elasticsearch——写得非常好

    How to resolve unassigned shards in Elasticsearch 转自:https://www.datadoghq.com/blog/elasticsearch-un ...

  7. springboot打包成jar包后找不到xml,找不到主类的解决方法

    springboot打包成jar包后找不到xml,找不到主类的解决方法 请首先保证你的项目能正常运行(即不打包的时候运行无误),我们在打包时经常遇到如下问题: springboot打包成jar包后找不 ...

  8. Recovering unassigned shards on elasticsearch 2.x——副本shard可以设置replica为0在设置回来

    Recovering unassigned shards on elasticsearch 2.x 摘自:https://z0z0.me/recovering-unassigned-shards-on ...

  9. 如何在Elasticsearch中解析未分配的分片(unassigned shards)

    一.精确定位到有问题的shards 1.查看哪些分片未被分配 curl -XGET localhost:9200/_cat/shards?h=index,shard,prirep,state,unas ...

随机推荐

  1. 数组进行多少次OP操作,才干有序

    1 题目描写叙述: 有一个数组:2,1,4,3.对于数组,有一种操作op(idx):将该index相应的数字移到首位.比方: op(3):  2 1 43 -> 3 2 1 4 op(1):   ...

  2. ubuntu13.10升级到ubuntu14.04

    Upgrading from Ubuntu 13.10 To upgrade from Ubuntu 13.10 on a desktop system: Open Software Sources. ...

  3. caffe训练CIFAR数据库

    CIFAR-10是一个用于普适物体识别的数据集.Cifar-10由60000张32*32的RGB彩色图片构成,50000张训练图片,10000张测试图片,分为10类.cifar下载地址: http:/ ...

  4. 对于NAS,IP SAN以及iSCSCI SAN存储的一些认识和理解

    一直以来用户对于在选购存储产品上有许多不清楚,市场上有NAS, FC SAN,IP SAN和iSCSCI SAN产品,到底哪种类型的产品更适合支撑企业的应用系统呢? 我们经常可以听到用户讲: “NAS ...

  5. 3. Longest Substring Without Repeating Character[M] 最大不重复子串

    题目 Given a string, find the length of the longest substring without repeating characters. Example 1: ...

  6. PHP配置优化:php-fpm配置解读

    PHP-FPM是一个PHP FastCGI管理器,php-fpm.conf配置文件用于控制PHP-FPM管理进程的相关参数,比如工作子进程的数量.运行权限.监听端口.慢请求等等. 我们在编译安装PHP ...

  7. 杂文 | 金沙江创投朱啸虎谈微信小程序

    我一直在想怎么去描述和朱啸虎的聊天. 最简单的开头一定是说“独角兽捕手朱啸虎”看好小程序了,这个赛道蕴含了许多无比巨大.充沛的机会,所以约到朱啸虎同学深聊这个赛道的创业.但是这样的开头太平,不好玩.事 ...

  8. Activity全屏沉浸状态

    public class MainActivity extends AppCompatActivity { private static final String TAG = "MainAc ...

  9. uva 11624 Fire! 【 BFS 】

    按白书上说的,先用一次bfs,求出每个点起火的时间 再bfs一次求出是否能够走出迷宫 #include<cstdio> #include<cstring> #include&l ...

  10. Kattis - CD

    CD Jack and Jill have decided to sell some of their Compact Discs, while they still have some value. ...