Hello everyone. I have an elastic cluster with version 7.16.2. A crontab script is run on one node, which makes a snapshot. Now, the situation is like this, it was decided to add a repository snapshot registration and deletion to the script. And is this normal behavior for the script to constantly register a snapshot repository, taking a snapshot and store on NFS share , and delete? The fact is that when you register a snapshot repository, elastic makes some changes to the "location". But, I could be wrong. Also, we all know that if an elastic file gets damaged, then everything is lost. Please tell me, does the constant registration and deletion of the repository snapshot by a script run by crontab affect data integrity? The essence of this method is that I do not have authorization in the cluster, therefore, I considered this method of registering and deleting the repository snapshot to be one of the ways to secure elastic.
I don't understand why you would want to do that.
No, it is definitely not normal to register and delete repositories frequently. Creating multiple snapshots in a single repository is going to use a lot less storage space than having lots of repositories with a single snapshot in them.
Hello, thank you for your reply. But, I'm not going to create multiple repositories with a single snapshot. Every day, the script will create and register a snapshot repository with a permanent "location". This means that there will be one repository where there will be several snapshots. And the snapshot repository will be registered and deleted daily. And the repository itself, where the snapshots are stored, will be permanent for snapshots. I just don't want my snapshot repository to remain available, so I'm deleting it. And when performing a backup, the script runs daily registration and deletion of the repository snapshot.