Repository verification exception when registering new ES snapshot

Hi,

I am trying to snapshot an elasticsearch repository.

I have three different nodes across three different centos servers, e.g. 10.212.145.10, 10.212.145.20, 10.212.145.30.

I have added the path.repo field in my elasticsearch.yml file (etc/elasticsearch/elasticsearch.yml) on all my nodes:

path.repo: ["/mnt/elasticsearch_repository/backup1"]

I have created the above directory on all my nodes and run a:

chown -R elasticsearch:elasticsearch /mnt/elasticsearch_repository/backup1

to ensure no access issues.

I have run the following to register the repository:

curl -XPUT -H "Content-Type: application/json" 'http://10.212.145.10:9200/_snapshot/backup' -d '{
    "type": "fs",
    "settings": {
        "location": "/mnt/elasticsearch_repository/backup1",
        "compress": true
    }
}'

However, when I do this I get the following error:

{"error":{"root_cause":[{"type":"repository_verification_exception","reason":"[backup] [[0dWhaMt8TYK-e6ck1tMuQQ, 'RemoteTransportException[[ES02][10.212.145.10:9300][internal:admin/repository/verify]]; nested: RepositoryVerificationException[[backup] a file written by master to the store [/mnt/elasticsearch_repository/backup1] cannot be accessed on the node [{ES02}{0dWhaMt8TYK-e6ck1tMuQQ}{FHIuvrRvSymkpW2h8pzrqQ}{10.212.145.20}{10.212.145.20:9300}{dilm}{ml.machine_memory=12429500416, xpack.installed=true, ml.max_open_jobs=20}]. This might indicate that the store [/mnt/elasticsearch_repository/backup1] is not shared between this node and the master node or that permissions on the store don't allow reading files written by the master node]; nested: NoSuchFileException[/mnt/elasticsearch_repository/backup1/tests-i039sRXORcSfMOvtR6LHtg/master.dat];'], [JSrN7i0yRgiPZYo5JGkRUA, 'RemoteTransportException[[ES03][10.212.145.30:9300][internal:admin/repository/verify]]; nested: RepositoryVerificationException[[backup] a file written by master to the store [/mnt/elasticsearch_repository/fitbbackup1]....... 

Any help would be greatly appreciated.

Thanks

It needs to be a shared volume, e.g. NFS, available to all nodes. Local directories do not work.

I have used a mounted shared drive - does this not work? How do I change to NFS?

I have also used the above method before and it worked fine.

The error indicates it is not working now though. Did you set it up correctly?

I only did the steps outline above - is there something more I should have done ?

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.