Ceph：Designing and Implementing Scalable Storage Systems
Michael Hackett Vikhyat Umrao Karan Singh Nick Fisk Anthony D'Atri Vaibhav Bhembre更新时间：2021-07-02 13:15:58
最新章节：Leave a review - let other readers know what you thinkcoverpage
Title Page
Copyright
Ceph: Designing and Implementing Scalable Storage Systems
About Packt
Why Subscribe?
Packt.com
Contributors
About the Authors
Packt Is Searching for Authors Like You
Preface
Who This Book Is For
What This Book Covers
To Get the Most out of This Book
Download the Example Code Files
Conventions Used
Get in Touch
Reviews
Ceph - Introduction and Beyond
Introduction
Ceph – the beginning of a new era
Software-defined storage – SDS
Cloud storage
Unified next-generation storage architecture
RAID – the end of an era
RAID rebuilds are painful
RAID spare disks increases TCO
RAID can be expensive and hardware dependent
The growing RAID group is a challenge
The RAID reliability model is no longer promising
Ceph – the architectural overview
Planning a Ceph deployment
Setting up a virtual infrastructure
Getting ready
How to do it...
Installing and configuring Ceph
Creating the Ceph cluster on ceph-node1
How to do it...
Scaling up your Ceph cluster
How to do it…
Using the Ceph cluster with a hands-on approach
How to do it...
Working with Ceph Block Device
Introduction
Configuring Ceph client
How to do it...
Creating Ceph Block Device
How to do it...
Mapping Ceph Block Device
How to do it...
Resizing Ceph RBD
How to do it...
Working with RBD snapshots
How to do it...
Working with RBD clones
How to do it...
Disaster recovery replication using RBD mirroring
How to do it...
Configuring pools for RBD mirroring with one way replication
How to do it...
Configuring image mirroring
How to do it...
Configuring two-way mirroring
How to do it...
See also
Recovering from a disaster!
How to do it...
Working with Ceph and OpenStack
Introduction
Ceph – the best match for OpenStack
Setting up OpenStack
How to do it...
Configuring OpenStack as Ceph clients
How to do it...
Configuring Glance for Ceph backend
How to do it…
Configuring Cinder for Ceph backend
How to do it...
Configuring Nova to boot instances from Ceph RBD
How to do it…
Configuring Nova to attach Ceph RBD
How to do it...
Working with Ceph Object Storage
Introduction
Understanding Ceph object storage
RADOS Gateway standard setup installation and configuration
Setting up the RADOS Gateway node
How to do it…
Installing and configuring the RADOS Gateway
How to do it…
Creating the radosgw user
How to do it…
See also…
Accessing the Ceph object storage using S3 API
How to do it…
Configuring DNS
Configuring the s3cmd client
Configure the S3 client (s3cmd) on client-node1
Accessing the Ceph object storage using the Swift API
How to do it...
Integrating RADOS Gateway with OpenStack Keystone
How to do it...
Integrating RADOS Gateway with Hadoop S3A plugin
How to do it...
Working with Ceph Object Storage Multi-Site v2
Introduction
Functional changes from Hammer federated configuration
RGW multi-site v2 requirement
Installing the Ceph RGW multi-site v2 environment
How to do it...
Configuring Ceph RGW multi-site v2
How to do it...
Configuring a master zone
Configuring a secondary zone
Checking the synchronization status
Testing user bucket and object sync between master and secondary sites
How to do it...
Working with the Ceph Filesystem
Introduction
Understanding the Ceph Filesystem and MDS
Deploying Ceph MDS
How to do it...
Accessing Ceph FS through kernel driver
How to do it...
Accessing Ceph FS through FUSE client
How to do it...
Exporting the Ceph Filesystem as NFS
How to do it...
Ceph FS – a drop-in replacement for HDFS
Operating and Managing a Ceph Cluster
Introduction
Understanding Ceph service management
Managing the cluster configuration file
How to do it...
Adding monitor nodes to the Ceph configuration file
Adding an MDS node to the Ceph configuration file
Adding OSD nodes to the Ceph configuration file
Running Ceph with systemd
How to do it...
Starting and stopping all daemons
Querying systemd units on a node
Starting and stopping all daemons by type
Starting and stopping a specific daemon
Scale-up versus scale-out
Scaling out your Ceph cluster
How to do it...
Adding the Ceph OSD
Adding the Ceph MON
There's more...
Scaling down your Ceph cluster
How to do it...
Removing the Ceph OSD
Removing the Ceph MON
Replacing a failed disk in the Ceph cluster
How to do it...
Upgrading your Ceph cluster
How to do it...
Maintaining a Ceph cluster
How to do it...
How it works...
Throttle the backfill and recovery:
Ceph under the Hood
Introduction
Ceph scalability and high availability
Understanding the CRUSH mechanism
CRUSH map internals
How to do it...
How it works...
CRUSH tunables
The evolution of CRUSH tunables
Argonaut – legacy
Bobtail – CRUSH_TUNABLES2
Firefly – CRUSH_TUNABLES3
Hammer – CRUSH_V4
Jewel – CRUSH_TUNABLES5
Ceph and kernel versions that support given tunables
Warning when tunables are non-optimal
A few important points
Ceph cluster map
High availability monitors
Ceph authentication and authorization
Ceph authentication
Ceph authorization
How to do it…
I/O path from a Ceph client to a Ceph cluster
Ceph Placement Group
How to do it…
Placement Group states
Creating Ceph pools on specific OSDs
How to do it...
The Virtual Storage Manager for Ceph
Introductionc
Understanding the VSM architecture
The VSM controller
The VSM agent
Setting up the VSM environment
How to do it...
Getting ready for VSM
How to do it...
Installing VSM
How to do it...
Creating a Ceph cluster using VSM
How to do it...
Exploring the VSM dashboard
Upgrading the Ceph cluster using VSM
VSM roadmap
VSM resources
More on Ceph
Introduction
Disk performance baseline
Single disk write performance
How to do it...
Multiple disk write performance
How to do it...
Single disk read performance
How to do it...
Multiple disk read performance
How to do it...
Results
Baseline network performance
How to do it...
Ceph rados bench
How to do it...
How it works...
RADOS load-gen
How to do it...
How it works...
There's more...
Benchmarking the Ceph Block Device
How to do it...
How it works...
Benchmarking Ceph RBD using FIO
How to do it...
Ceph admin socket
How to do it...
Using the ceph tell command
How to do it...
Ceph REST API
How to do it...
Profiling Ceph memory
How to do it...
The ceph-objectstore-tool
How to do it...
How it works...
Using ceph-medic
How to do it...
How it works...
See also
Deploying the experimental Ceph BlueStore
How to do it...
See Also
Deploying Ceph
Preparing your environment with Vagrant and VirtualBox
System requirements
Obtaining and installing VirtualBox
Setting up Vagrant
The ceph-deploy tool
Orchestration
Ansible
Installing Ansible
Creating your inventory file
Variables
Testing
A very simple playbook
Adding the Ceph Ansible modules
Deploying a test cluster with Ansible
Change and configuration management
Summary
BlueStore
What is BlueStore?
Why was it needed?
Ceph's requirements
Filestore limitations
Why is BlueStore the solution?
How BlueStore works
RocksDB
Deferred writes
BlueFS
How to use BlueStore
Upgrading an OSD in your test cluster
Summary
Erasure Coding for Better Storage Efficiency
What is erasure coding?
K+M
How does erasure coding work in Ceph?
Algorithms and profiles
Jerasure
ISA
LRC
SHEC
Where can I use erasure coding?
Creating an erasure-coded pool
Overwrites on erasure code pools with Kraken
Demonstration
Troubleshooting the 2147483647 error
Reproducing the problem
Summary
Developing with Librados
What is librados?
How to use librados?
Example librados application
Example of the librados application with atomic operations
Example of the librados application that uses watchers and notifiers
Summary
Distributed Computation with Ceph RADOS Classes
Example applications and the benefits of using RADOS classes
Writing a simple RADOS class in Lua
Writing a RADOS class that simulates distributed computing
Preparing the build environment
RADOS class
Client librados applications
Calculating MD5 on the client
Calculating MD5 on the OSD via RADOS class
Testing
RADOS class caveats
Summary
Tiering with Ceph
Tiering versus caching
How Cephs tiering functionality works
What is a bloom filter
Tiering modes
Writeback
Forward
Read-forward
Proxy
Read-proxy
Uses cases
Creating tiers in Ceph
Tuning tiering
Flushing and eviction
Promotions
Promotion throttling
Monitoring parameters
Tiering with erasure-coded pools
Alternative caching mechanisms
Summary
Troubleshooting
Repairing inconsistent objects
Full OSDs
Ceph logging
Slow performance
Causes
Increased client workload
Down OSDs
Recovery and backfilling
Scrubbing
Snaptrimming
Hardware or driver issues
Monitoring
iostat
htop
atop
Diagnostics
Extremely slow performance or no IO
Flapping OSDs
Jumbo frames
Failing disks
Slow OSDs
Investigating PGs in a down state
Large monitor databases
Summary
Disaster Recovery
What is a disaster?
Avoiding data loss
What can cause an outage or data loss?
RBD mirroring
The journal
The rbd-mirror daemon
Configuring RBD mirroring
Performing RBD failover
RBD recovery
Lost objects and inactive PGs
Recovering from a complete monitor failure
Using the Cephs object store tool
Investigating asserts
Example assert
Summary
Operations and Maintenance
Topology
The 40 000 foot view
Drilling down
OSD dump
OSD list
OSD find
CRUSH dump
Pools
Monitors
CephFS
Configuration
Cluster naming and configuration
The Ceph configuration file
Admin sockets
Injection
Configuration management
Scrubs
Logs
MON logs
OSD logs
Debug levels
Common tasks
Installation
Ceph-deploy
Flags
Service management
Systemd: the wave (tsunami?) of the future
Upstart
sysvinit
Component failures
Expansion
Balancing
Upgrades
Working with remote hands
Summary
Monitoring Ceph
Monitoring Ceph clusters
Ceph cluster health
Watching cluster events
Utilizing your cluster
OSD variance and fillage
Cluster status
Cluster authentication
Monitoring Ceph MONs
MON status
MON quorum status
Monitoring Ceph OSDs
OSD tree lookup
OSD statistics
OSD CRUSH map
Monitoring Ceph placement groups
PG states
Monitoring Ceph MDS
Open source dashboards and tools
Kraken
Ceph-dash
Decapod
Rook
Calamari
Ceph-mgr
Prometheus and Grafana
Summary
Performance and Stability Tuning
Ceph performance overview
Kernel settings
pid_max
kernel.threads-max vm.max_map_count
XFS filesystem settings
Virtual memory settings
Network settings
Jumbo frames
TCP and network core
iptables and nf_conntrack
Ceph settings
max_open_files
Recovery
OSD and FileStore settings
MON settings
Client settings
Benchmarking
RADOS bench
CBT
FIO
Fill volume then random 1M writes for 96 hours no read verification:
Fill volume then small block writes for 96 hours no read verification:
Fill volume then 4k random writes for 96 hours occasional read verification:
Summary
Other Books You May Enjoy
Leave a review - let other readers know what you think
更新时间：2021-07-02 13:15:58