Snapshot repository

[jhahn]

I am running 2024R1.3.5 on our nagioslog server cluster with 3 nodes. each has 8vCPU and 32 GB ram. I am using an NFS Share as our repository, its mapped and accessable to the nagios server. Everytime I try to get a list of the snapshots, it comes back with no data. I see the snapshots in the repo, I even adjust the dates in the web interface to find the data but it just seems to keep spinning and nothing ever comes up. what should I do here?

Thanks,
Joe H

[helen9c]

Could it be a caching or indexing issueTunnel Rush Game? Try running a repository sync or refresh from the CLI to see if the snapshots appear.

[jhahn]

So I looked on line and I can not find anywhere where it allows me to run a repository Sync from the CLI. I can not even get to Admin>Snapshots & Maintenance from the web interface. when I run from the command line curl -XPOST 'http://localhost:9200/_export/state' I get {"count":0,"states":[]} basically telling me there are no back up jobs running. If you know how to run a Sync manually at the CLI or a doc you can point me to that I will dig deeper.


thanks,
Joe h

[jhahn]

As a side not, I ran curator with the following results

/usr/local/nagioslogserver/scripts# ./curator.sh show snapshots --repository repository
2025-11-19 16:45:57,495 INFO Job starting: show snapshots
2025-11-19 16:46:27,529 ERROR Unable to find all snapshots in repository: repository
2025-11-19 16:46:27,529 ERROR No snapshots found in Elasticsearch.

[helen9c]

Hey Joe, it sounds frustrating that nothing shows up even from CLI or Curator. A few things I’d check next:

• Double-check the NFS repo permissions - ES needs write access, not just read. Sometimes it silently refuses otherwise.

• Make sure the repo is actually registered in Elasticsearch. The folder can exist but ES won’t use it until it’s defined.

• Take a look at Elasticsearch logs on all nodes; sometimes one node failing can make snapshots invisible cluster-wide.

If nothing else works, maybe try creating a small test repo and see if snapshots appear there. That can help figure out if it’s the repo itself or something in the cluster.

Hope that helps a bit. These situations can be tricky.

[vedicsystems]

You need to check the permissions on your NFS mount point on the Nagios server. If the user running the Nagios service doesn't have read access to the snapshot files, the web interface won't be able to list them, even if the share is mounted. Verify the file ownership and permissions for the repository path.

[jmichaelson]

So Elasticsearch is creating snapshots (presumably correctly as they show up on the shared mount) but can't find them afterward? This is going to take some time to look into as the version of Elasticsearch is out of date and documentation is practically non-existent anymore. But I'll try to see what i can come up with for you.