> For the complete documentation index, see [llms.txt](https://documentation.tjhsst.edu/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://documentation.tjhsst.edu/machines/ceph.md).

# Ceph

The current set of G10 Ceph servers (karel, wumpus, and stobar) was acquired in 2018. Each server has 64GB of RAM, an 8-core AMD EPYC processor, 12 4TB hard drives, and two 400GB SSDs. See [here](/technologies/storage/ceph.md) for more info on how they are configured.

[Waitaha](/machines/ceph/waitaha.md), [Barrel](/machines/ceph/barrel.md), and [Valdes](/machines/ceph/valdes.md) make up the failover Ceph cluster.  The OSD disks themselves are on the [Apocalypse](/obsolete/apocalypse.md) array.

## History

The current Ceph servers were purchased in the summer of 2018 along with [Overlord](/machines/vm-servers/overlord.md), [Torch](/machines/vm-servers/torch.md), and [Waverider](/machines/vm-servers/waverider.md) (see [2018 Server Purchases](/machines/history/2018-purchases.md)).

After the events of the [Cephpocalypse](/machines/history/2018-cephpocalypse.md), karel, wumpus, and stobar became the new Ceph cluster. For a period of time, wumpus was running Ubuntu 16.04 LTS, leading to high latencies on the entire cluster. The ability to take it down (in the middle of a school day) to upgrade it (to 18.04 LTS with a newer kernel) is a testament to the fault-tolerancy and redundancy of Ceph.