[clug] JOB: Ceph admins for short or long term contract

[Chris Smart]

Hi all,

We're in need of Linux admins with solid experience running Ceph in production, for short or long term (2 yrs+) contracts (whatever suits). Our two main clusters have about 1500 OSDs each and are running Red Hat Ceph 3 (Luminous) with Bluestore (SAS with NVMe for DB/WAL), deployed with Ceph Ansible and underpin OpenStack clusters.

We are looking for admins to help our team with some or all of the following:
- Tuning of the clusters, including pgnum calculations, primaries and even pg distribution, upmap, hotspots, etc
- Tuning of Ceph storage nodes (768GB RAM, 72 threads, LACP bonds and multiple networks)
- Crush maps for additional resiliency and separating data across old and new nodes
- Monitoring and alerts (Prometheus and Alert Manager)
- Day 2 automation with Ansible
- Adding more Ceph mon nodes (expand from 3 to 5)
- Upgrading existing Ceph 3 to containerised Ceph 3
- Upgrading Ceph 3 to Ceph 4, then probably 5
- Teaching and training existing staff (as required)
- Whatever else you think needs to be done

If you or someone you know is interested (and an Australian citizen), then please contact me offline and name your price.

Thanks,
-c