Ceph Makefile

Makefile

# --- Configuration ---
MON_IP ?= 192.168.0.2  # Primary monitor node IP
POOL_NAME ?= $(shell sudo ceph osd pool ls | grep "cephfs.cephfs.data" || echo "default_pool")
POOL_SIZE ?= 3
OSD_DEVICES ?= /dev/sdb  # Modify as per your system
FS_NAME ?= cephfs
CEPH_IMAGE ?= quay.io/ceph/ceph:v19.2.1

# Cluster Nodes
MON_NODES = mon1 mon2 mon3
OSD_NODES = osd1 osd2 osd3
MGR_NODES = mgr1 mgr2
MDS_NODES = mds1 mds2
RGW_NODES = rgw1

# SSH Credentials
SSH_USER ?= ceph-admin

# --- High-Level Targets ---

.PHONY: setup-cluster expand-cluster recover-cluster monitor-cluster upgrade-cluster status clean list-pools mount-cephfs

## Full Cluster Setup (One Command)
setup-cluster: install bootstrap add-mon add-osd add-mgr add-mds add-rgw create-pool enable-dashboard status
	@echo "✅ Ceph Cluster Setup Completed!"

## Expand the Cluster with New Storage Nodes
expand-cluster: add-osd
	@echo "✅ Ceph Cluster Expanded!"

## Detect and Recover Failed OSDs
recover-cluster:
	@echo "🔍 Checking for failed OSDs..."
	@FAILED_OSDS=$$(sudo ceph health detail | grep -oP 'osd.\d+' | grep -oP '\d+'); \
	if [ -n "$$FAILED_OSDS" ]; then \
		for osd in $$FAILED_OSDS; do \
			echo "🔄 Replacing OSD $$osd"; \
			sudo ceph osd out $$osd; \
			sudo ceph osd purge $$osd --yes-i-really-mean-it; \
			sudo ceph orch apply osd --all-available-devices; \
		done; \
		echo "✅ OSD Recovery Completed!"; \
	else \
		echo "✅ No failed OSDs detected."; \
	fi

## Recover a failed Ceph daemon
recover-daemon:
	@echo "🔧 Checking for failed Ceph daemons..."
	@FAILED_DAEMONS=$$(ceph health detail | grep 'CEPHADM_FAILED_DAEMON' | grep -oP 'daemon \S+ on \S+' | awk '{print $$2}'); \
	if [ -n "$$FAILED_DAEMONS" ]; then \
		for daemon in $$FAILED_DAEMONS; do \
			echo "🔄 Attempting to recover $$daemon..."; \
			sudo ceph orch restart $$daemon || sudo ceph orch redeploy $$daemon; \
			echo "✅ Recovery attempted for $$daemon."; \
		done; \
	else \
		echo "✅ No failed daemons detected."; \
	fi

## Recover Ceph Dashboard (Restart MGR and Re-enable Dashboard)
recover-dashboard:
	@echo "🔍 Checking Ceph Manager and Dashboard..."
	@MGR_STATUS=$$(ceph mgr dump | grep "active_name"); \
	if [ -z "$$MGR_STATUS" ]; then \
		echo "⚠️ No active Ceph Manager detected! Restarting..."; \
		sudo ceph orch restart mgr; \
	fi
	@echo "🔄 Restarting Ceph Dashboard..."
	sudo ceph mgr module disable dashboard || true
	sudo ceph mgr module enable dashboard
	@echo "✅ Ceph Dashboard restarted successfully!"
	@echo "📌 You can access it at:"
	@ceph mgr services | grep dashboard

## Enable Monitoring (Prometheus & Grafana)
monitor-cluster:
	@echo "📊 Setting up monitoring..."
	sudo ceph mgr module enable prometheus
	sudo ceph mgr module enable dashboard
	sudo ceph dashboard create-self-signed-cert
	@echo "✅ Monitoring enabled! Access Grafana via the Ceph dashboard."

## Upgrade the Ceph Cluster
upgrade-cluster:
	@echo "⬆️ Upgrading Ceph to $(CEPH_IMAGE)..."
	sudo ceph orch upgrade start --image $(CEPH_IMAGE)
	@echo "✅ Ceph upgrade initiated!"

## Check Cluster Status
status:
	@echo "🔍 Checking Ceph Cluster Status..."
	sudo ceph status

## Clean up (Reset Cluster)
clean:
	@echo "🗑️ WARNING: This will remove all Ceph data!"
	sudo cephadm rm-cluster --force --zap-osds
	@echo "✅ Ceph Cluster Removed!"

# --- Core Tasks ---

.PHONY: install bootstrap add-mon add-osd add-mgr add-mds add-rgw create-pool enable-dashboard

install:
	@echo "📥 Installing Ceph dependencies..."
	sudo apt update && sudo apt install -y ceph cephadm ceph-common

bootstrap:
	@echo "🚀 Bootstrapping Ceph cluster..."
	sudo cephadm bootstrap --mon-ip $(MON_IP)

add-mon:
	@echo "🔗 Adding Monitor nodes..."
	for node in $(MON_NODES); do \
		ssh $(SSH_USER)@$$node "sudo cephadm add mon"; \
	done

add-osd:
	@echo "📦 Adding OSD nodes..."
	for node in $(OSD_NODES); do \
		ssh $(SSH_USER)@$$node "sudo ceph orch apply osd --all-available-devices"; \
	done

add-mgr:
	@echo "🛠️ Adding Manager nodes..."
	for node in $(MGR_NODES); do \
		ssh $(SSH_USER)@$$node "sudo cephadm add mgr"; \
	done

add-mds:
	@echo "📂 Adding Metadata Servers..."
	for node in $(MDS_NODES); do \
		ssh $(SSH_USER)@$$node "sudo ceph orch apply mds $(FS_NAME) --placement=1"; \
	done

add-rgw:
	@echo "🌍 Adding RADOS Gateway (S3-Compatible Object Storage)..."
	for node in $(RGW_NODES); do \
		ssh $(SSH_USER)@$$node "sudo ceph orch apply rgw rgw.$(shell hostname) --placement=1"; \
	done

create-pool:
	@echo "💾 Creating Ceph Pool: $(POOL_NAME)..."
	sudo ceph osd pool create $(POOL_NAME) 128 128 replicated
	sudo ceph osd pool set $(POOL_NAME) size $(POOL_SIZE)

enable-dashboard:
	@echo "🌍 Enabling Ceph Dashboard..."
	sudo ceph mgr module enable dashboard
	sudo ceph dashboard create-self-signed-cert

## List all Ceph Pools
list-pools:
	@echo "📌 Listing Ceph Pools..."
	@echo "-----------------------------------"
	@echo "Name                     | Data Protection | Application | PG Status"
	@echo "-------------------------|----------------|------------|-----------"
	@sudo ceph osd pool ls detail | awk '/pool/{printf "%-25s", $$2} /size/{printf "| Replication: x%s ", $$3} /application/{printf "| %s ", $$3} /pg_num/{printf "| %s\n", $$2}'
	@echo "-----------------------------------"

## Mount CephFS on a local directory
mount-cephfs:
	@echo "📂 Mounting CephFS..."
	sudo mkdir -p /mnt/cephfs
	sudo mount -t ceph :/ /mnt/cephfs -o name=admin,secret=$(shell sudo ceph auth get-key client.admin)
	@echo "✅ CephFS Mounted at /mnt/cephfs"