惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

J
Java Code Geeks
Engineering at Meta
Engineering at Meta
GbyAI
GbyAI
MongoDB | Blog
MongoDB | Blog
Blog — PlanetScale
Blog — PlanetScale
腾讯CDC
U
Unit 42
freeCodeCamp Programming Tutorials: Python, JavaScript, Git & More
Apple Machine Learning Research
Apple Machine Learning Research
M
MIT News - Artificial intelligence
人人都是产品经理
人人都是产品经理
Hugging Face - Blog
Hugging Face - Blog
MyScale Blog
MyScale Blog
小众软件
小众软件
博客园 - 三生石上(FineUI控件)
N
Netflix TechBlog - Medium
阮一峰的网络日志
阮一峰的网络日志
博客园 - Franky
Recent Announcements
Recent Announcements
A
About on SuperTechFans
Stack Overflow Blog
Stack Overflow Blog
The GitHub Blog
The GitHub Blog
D
Docker
H
Hackread – Cybersecurity News, Data Breaches, AI and More

OneUptime Blog

How to Monitor Azure App Services (PaaS) with OpenTelemetry Grafana Stack vs OneUptime: DIY Observability or Unified Platform? Your AI Workloads Are About to Blow Up Your Observability Bill The Great Observability Consolidation Is Here How to Write Custom Object Classes for Ceph How to Write Custom Ceph Manager Modules How to Write a ceph.conf Configuration File How to Use Rook-Ceph with OpenShift How to Use Rook-Ceph with Longhorn for Comparison How to Configure Volume Snapshot Class for RBD in Rook How to Configure VolumeReplicationClass Scheduling Intervals in Rook How to Set Up Volume Replication with Rook-Ceph How to Create Volume Group Snapshots with Rook CSI How to Visualize Ceph Network Performance in Grafana How to Enable Virtual Host-Style Bucket Access in Rook How to View Runtime Configuration via Admin Socket How to View Quota Settings and Update Stats in Ceph RGW How to View PG Scaling Recommendations with autoscale-status How to View PG Distribution via Admin Socket How to View Performance Metrics in the Ceph Dashboard How to View Connection Status via Admin Socket How to View Ceph Cluster Summary Dashboard via CLI How to Version Control Rook-Ceph Configuration How to Version Control Ceph Infrastructure with Terraform How to Verify Kubernetes Node Requirements for Rook-Ceph Deployment How to Verify Health Before and After Rook Upgrades How to Verify Data Integrity with Deep Scrubbing How to Verify Complete Rook-Ceph Cleanup How to Verify Backup Integrity from Ceph Snapshots How to Use Rook-Ceph with Velero for Kubernetes Backup
How to View OSD Performance Counters in Ceph
Nawaz Dhandala · 2026-03-31 · via OneUptime Blog

What Are Performance Counters?

Ceph daemons expose internal performance counters via the admin socket. These counters track IOPS, byte throughput, operation latency histograms, journal performance, and queue depths. They are the primary source of low-level I/O diagnostics.

Accessing via ceph tell

The simplest way to get performance counters without SSH:

ceph tell osd.0 perf dump

This returns a large JSON blob. Filter for key metrics:

ceph tell osd.0 perf dump | python3 -c "
import json, sys
data = json.load(sys.stdin)
op = data.get('osd', {})
print('op_r:', op.get('op_r', 0))
print('op_w:', op.get('op_w', 0))
lat = op.get('op_r_latency', {})
print('op_r_latency (avg s):', lat.get('avgtime', 0))
"

Accessing via the Admin Socket

On the OSD host directly:

ceph daemon osd.0 perf dump

Key Performance Counters

CounterDescription
osd.op_rTotal read operations
osd.op_wTotal write operations
osd.op_r_latencyRead latency histogram
osd.op_w_latencyWrite latency histogram
osd.op_r_process_latencyTime processing reads
filestore.journal_latencyJournal write latency (FileStore)
bluestore.submit_latBlueStore submit latency
osd.op_in_bytesBytes received
osd.op_out_bytesBytes sent

Real-Time Monitoring with ceph daemon

Poll counters continuously:

watch -n 2 "ceph daemon osd.0 perf dump | python3 -c \"
import json,sys
d=json.load(sys.stdin)
osd=d.get('osd',{})
print('reads:', osd.get('op_r',0))
print('writes:', osd.get('op_w',0))
print('read_latency_avg:', osd.get('op_r_latency',{}).get('avgtime',0))
\""

Cluster-Wide Counter Collection

To collect performance counters from all OSDs at once:

ceph tell osd.* perf dump

This queries counters from every OSD daemon. To see all available counter definitions with types and descriptions for a specific OSD:

ceph tell osd.0 perf schema

BlueStore-Specific Counters

For BlueStore OSDs, check storage layer metrics:

ceph daemon osd.0 perf dump | python3 -c "
import json,sys
d=json.load(sys.stdin)
bs=d.get('bluestore',{})
print('kv_sync_lat:', bs.get('kv_sync_lat',{}).get('avgtime',0))
print('kv_final_lat:', bs.get('kv_final_lat',{}).get('avgtime',0))
"

Summary

Ceph OSD performance counters are accessed via ceph tell osd.<id> perf dump for remote access or ceph daemon osd.<id> perf dump locally. Focus on op_r_latency, op_w_latency, and BlueStore kv_sync_lat to diagnose I/O performance problems. Use ceph tell osd.* perf dump for a cluster-wide view.