Tips & tricks for Ceph based on 3 cases from our recent experience

PLEASE NOTE: our blog has MOVED to https://blog.flant.com/! New articles from Flant’s engineers will be posted there only. Check it out and subscribe to keep updated!

When using Ceph as a network storage in projects with various loads, we may face different tasks that do not seem trivial. Here are a few examples:

• migrating data into the new Ceph instance with partial use of past servers in the new cluster;
• solving the problem of allocation of disk space in Ceph.

To deal with these tasks we need to correctly remove OSDs while keeping the data intact. It is especially important in the case of massive volumes of data. That is what we will cover in this article.

The methods described below are suitable for any version of Ceph (unless special notes are given). In addition, we are going to take into account the fact that huge amounts of data can be stored in Ceph, thus we will split some steps into smaller ones — to prevent data loss and other problems.

Please, read the rest here.