Elasticsearch Create a snapshot repository error


(Canon88) #1

When I use
curl -XPUT 'http://192.168.184.33:9200/_snapshot/elkbak' -d '{"type": "fs","settings": {"location": "/data/elk_bak","compress": true}}'.
Shows the following error:
{"error":{"root_cause":[{"type":"repository_verification_exception","reason":"[elkbak] [AvxdLuwDTGCif9MujDW8mw, 'RemoteTransportException[[logserver-es-02][192.168.184.34:9300][internal:admin/repository/verify]]; nested: RepositoryVerificationException[[elkbak] store location [/data/elk_bak] is not accessible on the node [{logserver-es-02}{AvxdLuwDTGCif9MujDW8mw}{192.168.184.34}{192.168.184.34:9300}{master=true}]]; nested: NotSerializableExceptionWrapper[access_denied_exception: /data/elk_bak/tests-UI5DJFuzRLWs2XuQ2SIcgA/data-AvxdLuwDTGCif9MujDW8mw.dat];']]"}],"type":"repository_verification_exception","reason":"[elkbak] [AvxdLuwDTGCif9MujDW8mw, 'RemoteTransportException[[logserver-es-02][192.168.184.34:9300][internal:admin/repository/verify]]; nested: RepositoryVerificationException[[elkbak] store location [/data/elk_bak] is not accessible on the node [{logserver-es-02}{AvxdLuwDTGCif9MujDW8mw}{192.168.184.34}{192.168.184.34:9300}{master=true}]]; nested: NotSerializableExceptionWrapper[access_denied_exception: /data/elk_bak/tests-UI5DJFuzRLWs2XuQ2SIcgA/data-AvxdLuwDTGCif9MujDW8mw.dat];']]"},"status":500}

This is my configuration.
logserver-es-01

cluster.name: my_logsearch
node.name: logserver-es-01
node.master: true
node.data: true
index.number_of_shards: 5
index.number_of_replicas: 1
path.logs: "/data/logs/elasticsearch"
path.data: "/data/elasticsearch"
network.host: 192.168.184.33
http.port: 9200
bootstrap.mlockall: true
transport.tcp.compress: true

discovery.zen.fd.connect_on_network_disconnect : true
discovery.zen.initial_ping_timeout : 10s
discovery.zen.fd.ping_interval : 2s
discovery.zen.fd.ping_retries : 10

network.tcp.keep_alive : true
indices.recovery.max_size_per_sec: 0
indices.recovery.concurrent_streams: 5
discovery.zen.ping.multicast.enabled: false
discovery.zen.ping.timeout: 3s
discovery.zen.ping.unicast.hosts: ["logserver-es-01","logserver-es-02"]
path.repo: ["/data/elk_bak"]
logserver-es-02

node.name: logserver-es-02
....
path.repo: ["/data/elk_bak"]

What is the problem, anybody help me. Thank you!


(Joar Svensson) #2

Are you sure the directory "/data/elk_bak" is accessible and that the permissions meet the requirements?


(Canon88) #3

The elk_bak permissions:
drwxr-xr-x 3 nobody nobody 4096 Sep 19 14:50 elk_bak
The Elasticsearch process, is to use elasticsearch start. Is this problem?


(Christian Dahlqvist) #4

The repository path must be a shared file system accessible to all nodes in the cluster. is this the case in your setup?


(Canon88) #5

Yes. All the nodes in the cluster mount NFS_Server folder.
NFS_Server:
[root @ localhost conf] # more/etc/exports
/backup 192.168.184.33 (rw, sync, no_root_squash)
/backup 192.168.184.34 (rw, sync, no_root_squash)

Cluster nodes 1

mount -t NFS 192.168.184.19: /backup/data/elk_bak /

Cluster node 2
mount -t NFS 192.168.184.19: /backup/data/elk_bak /


(Christian Dahlqvist) #6

Is the path "/backup/data/elk_bak/" or "/data/elk_bak/"?


(Canon88) #7

Is the path /data/elk_bak/
Cluster nodes 1
mount -t nfs 192.168.184.19:/backup /data/elk_bak/
Cluster nodes 2
mount -t nfs 192.168.184.19:/backup /data/elk_bak/

Cluster nodes 1
elasticsearch.yml
...
path.repo: ["/data/elk_bak"]

Cluster nodes 2
elasticsearch.yml
...
path.repo: ["/data/elk_bak"]


(Kenneth Mroz) #8

Was there a solution to this? currently running into the same issue?


(Amruta) #9

Is any solution there for this issue, I am running into same issue.