I will troubleshoot, tune or deploy your ceph cluster with cephadm
About this Gig
Ceph in HEALTH_WARN, slow, or rebuilding forever? I run production Ceph integrated with OpenStack for a Tier-1 telecom operator and deploy clusters with cephadm.
What I can do for you:
- Diagnose HEALTH_WARN/ERR: stuck PGs, down OSDs, near-full pools
- Speed up slow clusters: BlueStore, recovery/backfill and network tuning
- Design CRUSH rules and pools, replication vs erasure coding
- Deploy new clusters with cephadm (RBD, CephFS, RGW/S3)
- Integrate Ceph with OpenStack Glance, Cinder and Nova
- Set up Prometheus + Grafana monitoring and alerts
How it works:
1. Send me your "ceph -s" and "ceph osd tree" output before ordering
2. I confirm the scope and the right package
3. I work over SSH and explain every change
4. You get a written report or runbook at the end
Supported: Ceph Reef, Squid and Tentacle on Ubuntu, Rocky Linux and RHEL.
Customers live on the cluster? I plan every change to keep it online and safe.
Cloud provider:
Other
Expertise:
Installation
•
Backup
•
Migration
•
Debugging
•
Configuration
Cloud computing resource:
Other
FAQ
Q: What Ceph versions do you support?
Reef, Squid and Tentacle, deployed with cephadm. I can also help you upgrade older Pacific or Quincy clusters.
Q: Can you integrate Ceph with OpenStack?
Yes. Glance, Cinder, and Nova ephemeral. This is one of my core specialties.
Q: Can you fix my existing broken cluster?
Absolutely. Message me with ‘ceph status’ output and I’ll diagnose the issue.
Q: Do you set up monitoring?
Yes, Prometheus + Grafana dashboards are included in Premium and available as an add-on for other packages.
Can you fix a cluster that is still serving customers?
Yes. I work in small, reversible steps with throttled recovery so your clients stay online.

