Technical Report NetApp Element Software Remote Replication Feature Description and Deployment Guide Pavani Krishna Goutham Baru, NetApp January 2019 | TR-4741 Abstract This document describes different types of remote replication supported by NetApp ® storage clusters. It also provides a general description of system features, the setup process, and configurations that you must implement in various networking scenarios.
27
Embed
NetApp Element Software Remote Replication · Technical Report NetApp Element Software Remote Replication Feature Description and Deployment Guide Pavani Krishna Goutham Baru, NetApp
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Technical Report
NetApp Element Software Remote Replication Feature Description and Deployment Guide Pavani Krishna Goutham Baru, NetApp
January 2019 | TR-4741
Abstract
This document describes different types of remote replication supported by NetApp® storage
clusters. It also provides a general description of system features, the setup process, and
configurations that you must implement in various networking scenarios.
4.3 Data Consistency ............................................................................................................................................ 9
4.4 Data Efficiency ................................................................................................................................................ 9
5 Choosing the Appropriate Replication Method ................................................................................. 9
5.1 Considerations for Setting Up Synchronous Replication................................................................................. 9
5.2 Consideration for Setting Up Asynchronous Replication............................................................................... 10
5.3 Consideration for Setting Up Snapshot Replication ...................................................................................... 10
6 Remote Replication Topologies and Considerations...................................................................... 10
6.1 Scenario 1: End-to-End Maximum Transmission Unit of 9000 and Same Native VLAN End to End ............ 10
6.2 Scenario 2: End-to-End MTU of 9000 Is Supported; Site A and Site B Are on Different Native VLANs ....... 11
6.3 Scenario 3: An MTU of 9000 Is Supported Locally at Each Site; WAN Connection Between Sites ............. 12
7 Cluster Pairing Element Walk-Through ............................................................................................ 12
7.1 Pairing Clusters Using MVIP ......................................................................................................................... 12
7.2 Pairing Clusters with a Pairing Key ............................................................................................................... 14
Message: Resuming Local Sync ............................................................................................................................ 25
Message: Resuming Data Transfer ....................................................................................................................... 25
Message: Active ..................................................................................................................................................... 25
Where to Find Additional Information ..................................................................................................... 25
Version History .......................................................................................................................................... 25
LIST OF TABLES
Table 1) Synchronous replication latency and packet loss. ............................................................................................ 9
Table 2) Comparison of replication modes. .................................................................................................................. 10
Table 3) Time period (in hours) required for performing an initial or full data synchronization. .................................... 23
Table 4) TCP port requirements for replication. ............................................................................................................ 24
LIST OF FIGURES
Figure 1) Bond management and bond storage. ............................................................................................................ 6
Figure 3) Management and storage network connectivity. ............................................................................................. 7
Figure 4) Remote replication states. ............................................................................................................................... 8
Figure 5) Remote replication on the same layer 2 network. ......................................................................................... 11
Figure 6) Remote replication over different VLANs. ..................................................................................................... 11
Figure 7) Remote replication over different MTU networks. ......................................................................................... 12
Asynchronous replication (async) continuously replicates data from a source cluster to a target cluster
without waiting for the acknowledgments from the target cluster. During asynchronous replication, writes
are acknowledged to the client (application) after they are committed on the source cluster.
Asynchronous replication is appropriate for the following use cases:
• The disaster recovery site is far from the source and the application does not tolerate latencies induced by the network.
• There are bandwidth limitations on the network connecting the source and target clusters.
Snapshot-Only Replication
This feature replicates changed data at discrete points of time to the remote cluster. Only snapshots
created on the source cluster are replicated. Active writes from the source volume are not. Snapshot
replication does not affect asynchronous or synchronous replication. The snapshots are replicated
periodically as configured by the user.
2 Prerequisites for Remote Replication
The following are the characteristics and requirements of replicated volumes:
The ports required for remote replication should not be blocked by a
firewall. Refer to Appendix A: TCP Port Requirements
• There should be full end-to-end connectivity between the source cluster and the target cluster.
• Cluster pairing and volume pairing should be performed before remote replication.
• The administrator must know the password for at least one cluster, but preferably both clusters.
• There should be sufficient space on the remote cluster to create a volume as large as the primary site.
• A volume can be paired with only one volume at a time. Cascading replication is not supported.
• A cluster can pair with a maximum of four other clusters.
• Any number of volumes can replicate writes in the Active state. However, only 10 volumes at a time per node can be incorporated into a replication startup sequence. Therefore, if more than 10 volumes begin a replication startup sequence at the same time, only 10 of them move through the process at a time. The remaining volumes wait for a spot in the queue to open.
3 Cluster Pairing and Volume Pairing
This section introduces cluster pairing and volume pairing as a prelude to replication.
3.1 Prerequisites
Cluster Pairing
The following prerequisites are required to establish cluster pairing:
• Establish full network connectivity between the source and target clusters, including the management virtual IP addresses (MVIPs).
• Every node in the source cluster must be able to ping every other node in the target cluster.
• You must have cluster administrator privileges for one or both clusters that are being paired.
• The versions of the NetApp Element software present on a cluster must be compatible with the software versions of the other cluster. See NetApp Element Software User Guide.
Note: Cluster pairings are bidirectional.
• Firewall settings must permit communication between the source cluster and the target cluster.
Volume Pairing
The following prerequisites are required to establish volume pairing:
• Cluster pairing is a prerequisite for volume pairing between the clusters.
• The access mode on the destination volume should be the replication target, and the source volume should be in read/write mode.
• The source volume should be smaller than or of equal size to the target volume. If the volume size of the source is larger than the volume size of the target, replication transitions to an error state.
• The quality of service (QoS) settings for the target volume should be the same as for the source. This configuration allows the same volume performance for the target as for the source if a failover occurs. During synchronous replication, the QoS settings should be identical on both the source and the target to avoid volume throttling.
3.2 Management Virtual IP Address and Storage Virtual IP Address
Bonded networks aggregate two physical interfaces into one logical interface. This bonding of interfaces
is essential for high availability. The interfaces can be configured into two bonds on a storage node called
Bond1G and Bond10G. Bond10G is used for the storage network, and Bond 1G is used for the
management network. It is a best practice to assign a high-bandwidth network to storage traffic.
Management Virtual IP Address
A management virtual IP (MVIP) address is assigned to the logical interface designated for management
traffic of the cluster. Separating the interfaces of management traffic and storage traffic provides the
following advantages:
• Management traffic does not affect the bandwidth on the storage network.
• Traffic is isolated across different switches.
• Traffic is isolated across different virtual LANs (VLANs).
6.3 Scenario 3: An MTU of 9000 Is Supported Locally at Each Site; WAN Connection Between Sites
An MTU of 9000 is supported locally at each site, and a WAN connection between the sites supports an
MTU of approximately only 1500.
• The storage-side gateway on the cluster is configured on the Bond10G interface. The gateway router supports an MTU of 9000 on the interface facing the storage cluster.
• The gateway router’s WAN interface supports the WAN connection’s MTU (1500 in this case). Congestion can be caused by the difference in the supported MTU.
• The routers at each location must support Path MTU discovery, and the ICMP replies with ICMP type 3 messages.
Figure 7) Remote replication over different MTU networks.
Path MTU Discovery Overview
Path MTU discovery is an algorithm described and implemented in TCP stacks. This algorithm attempts
to discover the largest IP datagram that can be sent without fragmentation on an IP path and maximizes
data transfer throughput.
Path MTU discovery is implemented when the IP sender has set the Don’t Fragment (DF) flag in the IP
header. If an IP packet with this flag set reaches the router with a next-hop MTU that is too small, the
packet cannot be sent without fragmentation. In this case, the router discards the packet and sends an
ICMP Fragmentation Needed but DF Set error to the sender of the packet. When the sender receives this
error, the sender can send smaller MTU packets to reach the destination.
7 Cluster Pairing Element Walk-Through
7.1 Pairing Clusters Using MVIP
You can pair two clusters by using MVIP. Cluster administrator access is mandatory for the pairing, and
you must perform authentication before starting the cluster pairing. To use a pairing key, follow these
6. Choose the volume ID option and select the desired replication model. Select the name of the target cluster. Enter the administrator and password for credentials.
7. Select Home and check the volume pairs for pairing completion.
11 Recovery Point Objective and Recovery Time Objective
A replication architecture depends on your business requirements. The parameters that define the
business requirements are maximum permissible data loss, speed of recovery, and application timeout.
The speed of recovery depends on the network architecture between the source and replicated nodes.
Refer to the Interoperability Matrix Tool (IMT) on the NetApp Support site to validate that the exact product and feature versions described in this document are supported for your specific environment. The NetApp IMT defines the product components and versions that can be used to construct configurations that are supported by NetApp. Specific results depend on each customer’s installation in accordance with published specifications.
Software derived from copyrighted NetApp material is subject to the following license and disclaimer:
THIS SOFTWARE IS PROVIDED BY NETAPP “AS IS” AND WITHOUT ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE, WHICH ARE HEREBY DISCLAIMED. IN NO EVENT SHALL NETAPP BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
NetApp reserves the right to change any products described herein at any time, and without notice. NetApp assumes no responsibility or liability arising from the use of products described herein, except as expressly agreed to in writing by NetApp. The use or purchase of this product does not convey a license under any patent rights, trademark rights, or any other intellectual property rights of NetApp.
The product described in this manual may be protected by one or more U.S. patents, foreign patents, or pending applications.
Data contained herein pertains to a commercial item (as defined in FAR 2.101) and is proprietary to NetApp, Inc. The U.S. Government has a non-exclusive, non-transferrable, non-sublicensable, worldwide, limited irrevocable license to use the Data only in connection with and in support of the U.S. Government contract under which the Data was delivered. Except as provided herein, the Data may not be used, disclosed, reproduced, modified, performed, or displayed without the prior written approval of NetApp, Inc. United States Government license rights for the Department of Defense are limited to those rights identified in DFARS clause 252.227-7015(b).
Trademark Information
NETAPP, the NETAPP logo, and the marks listed at http://www.netapp.com/TM are trademarks of NetApp, Inc. Other company and product names may be trademarks of their respective owners.