HA Cluster User Guide - Marketo

HA Cluster User Guide - Marketo
High Availability Cluster
User Guide
v3.1.x
3000-hac-v3.1.x-000031-A
Copyright В© 2013 NexentaВ® Systems, ALL RIGHTS RESERVED
Notice: No part of this publication may be reproduced or transmitted in any form or by any means, electronic
or mechanical, including photocopying and recording, or stored in a database or retrieval system for any
purpose, without the express written permission of Nexenta Systems (hereinafter referred to as “Nexenta”).
Nexenta reserves the right to make changes to this documentation at any time without notice and assumes
no responsibility for its use. Nexenta products and services only can be ordered under the terms and
conditions of Nexenta Systems’ applicable agreements. All of the features described in this documentation
may not be available currently. Refer to the latest product announcement or contact your local Nexenta
Systems sales office for information on feature and product availability. This documentation includes the
latest information available at the time of publication.
Nexenta is a registered trademark of Nexenta Systems in the United States and other countries.
All other trademarks, service marks, and company names in this documentation are properties of their
respective owners.
Product Versions Applicable to this Documentation:
Product
Versions supported
NexentaStor
v3.1.x
HA Cluster
v3.x
ii
High Availability Cluster User Guide
Contents
Preface
1
................................................v
Introduction to HA Cluster
.................................1
About the Nexenta HA Cluster
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1
Storage Failover
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2
Exclusive Access to Storage
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2
SCSI-3 PGR for Additional Protection
Service Failover
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3
Sample Network Architecture
Additional Resources
2
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4
Configuring the HA Cluster
.................................5
About Configuring the HA Cluster
Prerequisites
. . . . . . . . . . . . . . . . . . . . . . . . .3
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .5
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .5
Resolving Hostname Conflicts
Defining Netmasks
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6
Binding the Nodes using SSH
Configuring the HA Cluster
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .7
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .8
Adding a Shared Volume to the HA Cluster
Importing a Volume to the HA Cluster
Removing a Shared Volume
3
. . . . . . . . . . . . . . . . . . . . . . . . . . .9
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .9
Heartbeat and Network Interfaces
About Heartbeat and Network Interfaces
Heartbeat Mechanism
. . . . . . . . . . . . . . . . . . . . . . . . . . 11
. . . . . . . . . . . . . . . . . . . . . . . . 11
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Modifying Heartbeat Interfaces
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Adding volume heartbeats
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Adding network heartbeats:
4
. . . . . . . . . . . . . . . . . . . . . . .8
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Adding serial heartbeats:
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Configuring Storage Failover
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
About Configuring Storage Failover
. . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
High Availability Cluster User Guide
iii
Contents
Cluster Configuration Data
Mapping Information
NFS/CIFS Failover
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Configuring iSCSI Targets for Failover
. . . . . . . . . . . . . . . . . . . . . . . . . . 17
Configuring Fibre Channel Targets for Failover
Miscellaneous Options for Fibre Channel
5
Advanced Setup
About Advanced Setup
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Setting Failover Mode
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Setting Automatic Failover Mode
. . . . . . . . . . . . . . . . . . . . . . . . . . . 22
Adding a Virtual IP Address and Hostname
Implementing Miscellaneous Options
System Operations
Viewing the HA Cluster Status
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Manually Triggering a Failover
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
Viewing Support Logs
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
Testing and Troubleshooting
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
Repairing a Broken Cluster Service
Replacing a Faulted Node
iv
. . . . . . . . . . . . . . . . . . . . . . . . . . . 24
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Verifying Shared Volume Status
8
. . . . . . . . . . . . . . . . . . . . . . . 22
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
About System Operations
7
. . . . . . . . . . . . . . . . . . . . . . 18
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Setting Manual Failover Mode
6
. . . . . . . . . . . . . . . . . . . . 18
Index
. . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
High Availability Cluster User Guide
Preface
This documentation presents information specific to Nexenta Systems, Inc.
products. The information is for reference purposes and is subject to
change.
Intended Audience
This documentation is intended for Network Storage Administrators. It
assumes that you have experience NexentaStor and with data storage
concepts, such as NAS, SAN, NFS, and ZFS.
Documentation History
The following table lists the released revisions of this documentation.
Table 1: Documentation Revision History
Revision
Date
Description
3000-hac-v3.1.x-000031-A
September, 2013
GA
Contacting Support
•
Visit the Nexenta Systems, Inc. customer portal http://
www.nexenta.com/corp/support/support-overview/account-management.
Login and browse the customers knowledge base.
Choose a method for contacting support:
•
Using the NexentaStor user interface, NMV (Nexenta Management
View):
a. Click Support.
b. Complete the request form.
c. Click Send Request.
•
Using the NexentaStor command line, NMC (Nexenta Management
Console):
a. At the command line, type support.
b. Complete the support wizard.
High Availability Cluster User Guide
v
Preface
Comments
Your comments and suggestions to improve this documentation are greatly
appreciated. Send any feedback to [email protected] and
include the documentation title, number, and revision. Refer to specific
pages, sections, and paragraphs whenever possible.
vi
High Availability Cluster User Guide
1
Introduction to HA Cluster
This section includes the following topics:
•
About the Nexenta HA Cluster
•
Storage Failover
•
Exclusive Access to Storage
•
Service Failover
•
Sample Network Architecture
•
Additional Resources
About the Nexenta HA Cluster
The Nexenta HA (High-Availability) Cluster consists of two NexentaStor
appliances and provides a storage volume-sharing service. Neither system
is designated as the primary or secondary system. You manage both
systems actively for shared storage, although only one system provides the
access to a shared volume at a time. After you create a volume on one
server and share it with the other server, then, when HAC detects a system
failure, it transfers ownership of the shared volumes to the other server in
the Cluster pair.
HA Cluster provides server monitoring and failover. Protection of services,
such as iSCSI, involves cooperation with other modules such as the SCSI
Target plugin.
An HA Cluster includes:
•
NexentaStor Appliances
Runs a defined set of services and monitors each other for failures.
HAC connects these NexentaStor appliances through various
communication channels, through which they exchange heartbeats
that provide information about their states and the services that
reside on them.
•
Cluster Service
A transferable unit that consists of:
•
Application start-up and shutdown code
High Availability Cluster User Guide
1
Introduction to HA Cluster
•
Network identity and appliance data
You can migrate services between cluster appliances manually, or
automatically, if one appliance fails.
Storage Failover
The primary benefit of HA Cluster is to detect storage system failures and
transfer ownership of the shared volumes to the alternate NexentaStor
appliance. All configured volume services failover to the other server. HA
Cluster ensures service continuity during exceptional events, including
power outages, disk failures, appliances that run out of memory or crash,
and other failures.
Currently, the minimum time to detect that an appliance has failed is
approximately 10 seconds. The failover and recovery time is largely
dependent on the amount of time it takes to failover the data volume on the
alternate appliance. Best practices to reduce the failover time include using
fewer zvols and file systems for each data volume. When using fewer file
systems, you may want to use other properties, such as reservations and
quotas, to control resource contention between multiple applications.
In the default configuration, HA Cluster implements failover storage services
if network connectivity is lost. HA Cluster automatically determines which
network device to monitor based on the services that are bound to an
interface. It checks all nodes in the cluster, so even if a node is not running
any services, HA Cluster continues to monitor the unused interfaces. If the
state of one changes to offline, it prevents failover to this node for services
that are bound to that interface. When the interface recovers, HA Cluster
enables failover for that interface again.
Other types of failure protection include link aggregation for network
interfaces and MPxIO for protection against SAS link failures.
Exclusive Access to Storage
You access a shared volume exclusively through the appliance that currently
owns the corresponding volume-sharing service. To ensure this exclusivity,
HA Cluster provides reliable fencing through the utilization of multiple types
of heartbeats. Fencing is the process of isolating a node in an HA cluster,
and/or protecting shared resources when a node malfunctions. Heartbeats,
or pinging, allow for constant communication between the servers. The
most important of these is the disk heartbeat. Generally, additional
heartbeat mechanisms increase reliability of the cluster's fencing logic; the
disk heartbeats, however, are essential.
HA Cluster can reboot the failed appliance in certain cases:
•
2
Failure to export the shared volume from the active node to the
passive node. This functionality is analogous to Stonith, the
technique for fencing in computer clusters.
High Availability Cluster User Guide
Introduction to HA Cluster
In addition, NexentaStor HA Cluster provides a number of other fail safe
mechanisms:
•
When you start a volume sharing service, make sure that the IP
address associated with that service is NOT attached to any
interface. The cluster automatically detects and reports if an
interface is using the IP address. If it is, the local service does not
perform the start-up sequence.
On disc systems which support a SCSI reservation, you can place a disc
before accessing the file systems, and set the system to panic if it loses the
reservation. This feature also serves to protect the data on a disc system.
HA Cluster also supports SCSI-2 reservations.
!
Note:
SCSI-3 PGR for Additional Protection
HA Cluster employs SCSI-3 PGR for additional protection.
SCSI-3 reservation enables write access for multiple nodes to a device and
simultaneously blocks access for other nodes. It uses the concept of Write
Exclusive Registrants Only. Therefore, only registered systems can perform
write operations.
You can enable PGR by issuing SCSI reservations on the devices in a volume
before you import them. This feature enforces data integrity which prevents
the pool from importing into two nodes at any time.
Nexenta recommends that you always deploy the HA Cluster with at least
one or more heartbeat channels (ethernet or serial).This configuration
ensures that the cluster always has exclusive access that is independent of
the storage interconnects used in the Cluster.
Service Failover
As discussed previously, system failures result in the failover of ownership
of the shared volume to the alternate node. As part of the failover process,
HA Cluster migrates the storage services that are associated with the shared
volume and restarts the services on the alternate node.
Sample Network Architecture
A sample cluster hardware setup includes:
•
Two x86/64-bit systems with a shared storage (SAS, iSCSI, or FC
storage)
•
Two network interface cards (not mandatory, but good practice)
High Availability Cluster User Guide
3
Introduction to HA Cluster
The following illustration is an example of an HA Cluster deployment of a
Nexenta iSCSI environment. The host server attaches to iSCSI LUNs in the
JBOD, which are connected to the Nexenta appliances nodeA and nodeB.
The Nexenta appliances use the active/passive function of the HA cluster.
NodeA services one group of iSCSI luns while nodeB presents a NAS storage
LUN.
The following diagram show an example of HA Cluster configuration.
CLIENT SIDE NETWORK
ETHERNET NETWORK
PRIMARY
PATH
ALTERNATE PATH
NEXENTA
APPLIANCE:
node A
NEXENTA
APPLIANCE:
node B
SATA
OK
1000GB
7.2 Krpm
SATA
OK
SATA
OK
1000GB
7.2 Krpm
JBODs
1000GB
7.2 Krpm
1000GB
7.2 Krpm
1000GB
7.2 Krpm
1000GB
7.2 Krpm
SATA
OK
1000GB
7. 2Krpm
1000GB
7. 2Krpm
1000GB
7. 2Krpm
1000 GB
7.2K rpm
SATA
OK
1000 GB
7.2K rpm
1000 GB
7.2K rpm
1000 GB
7.2K rpm
1000 GB
7.2K rpm
1000GB
7.2 Krpm
SATA
OK
SATA
OK
1000GB
7.2 Krpm
SATA
OK
SATA
OK
Sun Storage
J4400
1000GB
7.2 Krpm
SATA
OK
SATA
OK
1000GB
7.2 Krpm
SATA
OK
SATA
OK
SATA
OK
1000 GB
7.2K rpm
SATA
OK
SATA
OK
SATA
OK
1000GB
7. 2Krpm
SATA
OK
SATA
OK
1000GB
7. 2Krpm
SATA
OK
SATA
OK
1000GB
7. 2Krpm
SATA
OK
SATA
OK
1000GB
7.2 Krpm
SATA
OK
1000GB
7.2 Krpm
SHARED VOLUME
Additional Resources
Nexenta has various professional services offerings to assist with managing
HA Cluster. Nexenta strongly encourages a services engagement to plan and
install the plugin. Nexenta also offers training courses on high availability
and other features.
4
High Availability Cluster User Guide
2
Configuring the HA Cluster
This section includes the following topics:
•
About Configuring the HA Cluster
•
Prerequisites
•
Resolving Hostname Conflicts
•
Defining Netmasks
•
Binding the Nodes using SSH
•
Configuring the HA Cluster
•
Adding a Shared Volume to the HA Cluster
•
Importing a Volume to the HA Cluster
•
Removing a Shared Volume
About Configuring the HA Cluster
You can configure and manage the HA Cluster through the appliance’s web
interface, the Nexenta Management View (NMV), or the Nexenta
Management Console (NMC).
!
Note:
This section applies to new installations of HA Cluster. When upgrading,
you save and restore the configuration from your previous cluster, so the
following sections do not apply.
Prerequisites
Before configuringthe HA Cluster, complete the following tasks:
•
Configure two NexentaStor appliances.
See NexentaStor Installation Guide.
•
Connect a shared storage to the NexentaStor appliances (SAS, iSCSI,
or FC).
High Availability Cluster User Guide
5
Configuring the HA Cluster
•
Install the HA Cluster plugin on both NexentaStor appliances.
See Plugins in the NexentaStor Installation Guide.
Resolving Hostname Conflicts
There is a name associated with a shared volume service that is referred to
as a virtual shared service hostname, or virtual IP address (VIP). The
network clients use the virtual hostname to connect to the shared volume .
You must make the appliances in the HA cluster group resolvable to each
other. This means they must be able to detect each other on the network
and communicate. To achieve this, you can either configure your DNS server
accordingly, or add records to the host tables on the NexentaStor
appliances.
The NexentaStor host table (file /etc/hosts) is a simple text file that
associates IP addresses with hostnames. Note that you can have only one
line per IP address.
For each host a single line should be present with the following information:
IP_address
пЃ¶
hostname
[aliases...]
To configure the shared cluster volume manually:
1. Log in to the NMC on one of the NexentaStor appliances.
2. Type the following to open the /etc/hosts file:
nmc:/$ setup appliance hosts
3. Add IP addresses for each node using vim commands:
Example:
Internet host table
::1
localhost
127.0.0.1
localhost
192.168.60.107 <nodeA nodeA.example.com> loghost
192.168.60.79 <nodeB nodeB.example.com>
192.168.60.22 <virtual_hostname>
4. Repeat Step 1 — Step 3 for nodeB.
Defining Netmasks
Similar to the host table, you must also define the netmask values for the
IP addresses.
пЃ¶
To configure the shared cluster volume manually, using NMC:
1. Type the following to open the /etc/netmasks file for nodeA:
nmc:/$ setup appliance netmasks
2. Add the netmask for each network address:
6
High Availability Cluster User Guide
Configuring the HA Cluster
Example:
192.168.1.0
192.168.13.0
192.168.0.0
255.255.255.0
255.255.255.0
255.255.0.0
3. Repeat Step 1 — Step 3 for nodeB.
Binding the Nodes using SSH
Before you configure the SSH bindings, complete the steps in Resolving
Hostname Conflicts. You must bind the two HA nodes together with the SSH
protocol so that they can communicate.
пЃ¶
To bind the two nodes, using NMV:
1. Click Settings > Network.
2. In the Network panel, click SSH-Bind.
3. Type the following for the remote server:
•
Host name
•
User name
The user must have administrative priviledges.
•
Password
4. Click Bind.
5. Repeat Step 1 — Step 4 on the other node.
пЃ¶
To bind the two nodes, using NMC:
1. Log in to one of the NexentaStor appliances
2. Type the following:
nmc:/$ setup network ssh-bind
3. Type the host name of the NexentaStor appliance that you want to
bind.
4. Type the root password.
5. Repeat Step 1 — Step 4 on the other node.
6. To verify that you set up the bindings correctly, type:
nmc:/$ show network ssh-bindings
High Availability Cluster User Guide
7
Configuring the HA Cluster
Configuring the HA Cluster
Before you configure the HA Cluster, verify that you completed the steps in
Resolving Hostname Conflicts and Binding the Nodes using SSH. You need
to configure multiple options for the HA cluster before you can use it
successfully.
You cannot assign a NexentaStor appliance to more than one HA Cluster.
!
Note:
пЃ¶
To configure an HA cluster, using NMV:
1. On the first node, select Settings > HA Cluster.
2. In the Cluster Settings panel, click Initialize.
3. Type or change the Cluster name.
4. Optionally, type a description.
5. Select Enable Network Monitoring
The Cluster monitors the network for nodes.
6. Optionally, select Configure Serial heartbeat.
•
!
Select the serial ports.
Specify serial ports, if the NexentaStor appliances are connected using the
serial ports.
Note:
7. Click OK.
8. Click Configure.
9. Click Yes.
пЃ¶
To configure an HA cluster, using NMC:
1. Type:
nmc:/$ create group rsf-cluster
2. Follow the onscreen instructions.
3. Verify that you created the group, type:
nmc:/$ show group rsf-cluster
Adding a Shared Volume to the HA Cluster
After you configure the HA Cluster, you must add a shared volume to the
volume service. If you do not add a shared volume to the Cluster when you
create it, you can add it later.
пЃ¶
To add a volume to HAC, using NMV:
1. In the Cluster Settings panel, click Volumes.
8
High Availability Cluster User Guide
Configuring the HA Cluster
2. Select a volume from the dropdown menu.
3. In the Failover Hostname field, type the failover hostname found
in the /etc/hosts file.
If the failover hostname does not exist, create one, assign the
failover host to an available IP address, and add it to the /etc/hosts
file.
Example:
192.168.60.22
failover-host
4. Select Primary appliance.
5. Select Heartbeat devices.
6. Select failover interfaces.
7. Click Add this volume to the cluster.
8. Click Confirm.
пЃ¶
To create a shared service, using NMC:
1. Type:
nmc:/$ setup group rsf-cluster <cluster_group_name>
shared-volume add
2. Follow the onscreen instructions.
Importing a Volume to the HA Cluster
You may need to import a shared volume if it is exported. The volume may
be exported when you upgrade the NexentaStor appliance or it can be
imported on another NexentaStor appliance.
пЃ¶
To import a volume, using NMV:
1. Click Data Management > Data Sets.
2. In the All Volumes panel, click Import.
3. Click Import for the relevant volume.
пЃ¶
To import a volume, using NMC:
пЃµ
Type:
nmc:/$ setup volume import
Removing a Shared Volume
You can remove a shared volume from the HA Cluster control any time. The
remove operation does not delete any data on the volume. The volume
remains imported on the active HA Cluster node. Hovewer, the volume
becomes unavailable through the failover hostname.
High Availability Cluster User Guide
9
Configuring the HA Cluster
пЃ¶
To remove a shared volume, using NMV:
1. In the Cluster Settings panel, click Volumes.
2. Select the Remove a volume tab.
3. Select a volume.
4. Click
пЃ¶
.
To remove a shared volume, using NMC:
1. Type:
nmc:/$ setup group rsf-cluster <cluster> shared-volume
<volume> remove
System response:
Remove shared volume <volume> and restart HA Cluster
<cluster> ?
(y/n)
2. Confirm the operation by typing y.
System response:
Removing shared volume <volume>, please wait ...
Sep 3 13:47:56 myhost RSF-1[4879]: [ID 702911
local0.alert] RSF-1 cold restart: All services stopped.
Waiting for remove operation to complete ... done.
10
High Availability Cluster User Guide
3
Heartbeat and Network
Interfaces
This section includes the following topics:
•
About Heartbeat and Network Interfaces
•
Heartbeat Mechanism
•
Modifying Heartbeat Interfaces
About Heartbeat and Network Interfaces
A NexentaStor appliance in the HA Cluster constantly monitors the state and
status of the other appliance in the Cluster through heartbeats. Because HA
Cluster servers must determine that an appliance (member of the cluster)
has failed before taking over its services, you configure the cluster to use
several communication channels through which to exchange heartbeats.
Heartbeat Mechanism
The loss of all heartbeat channels represents a failure. If an appliance
wrongly detects a failure, it may attempt to start a service that is already
running on another server, leading to so-called split brain syndrome. This
can result in confusion and data corruption. Multiple, redundant heartbeats
prevent this from occurring.
In NexentaStor, VDEV labels for devices in the shared volume perform the
heartbeat function. If a shared volume consists of a few disks, NexentaStor
uses VDEV labels for one or more disks for the heartbeat mechanism. You
can specify which disks.
The heartbeat mechanism uses sectors 512 and 518 in the blank 8K space
of the VDEV label on each of the shared disks.
Though the quorum disk option still remains in the configuration file,
Nexenta recommends using the shared VDEV labels.
High Availability Cluster User Guide
11
Heartbeat and Network Interfaces
HA Cluster supports the following types of heartbeat communication
channels:
•
Disk Heartbeat
Accessible and writable from all appliances in the cluster or VDEV
labels of the devices in the shared volume.
•
Network Interfaces
The preferred heartbeat connection is a dedicated "cross-over"
connection between the nodes. You can also use or add any other
interface type (simple, IPMP, aggregate) for additional resiliency.
•
Serial Links
Heartbeat conection using a null-modem serial cable plugged into
both NexentaStor nodes.
Modifying Heartbeat Interfaces
When you define a cluster, you define the heartbeat properties. However,
you can modify them later, if needed.
A NexentaStor appliance cannot belong to more than one cluster.
!
Note:
пЃ¶
To change heartbeat properties, using NMV:
1. Click Settings > HA Cluster.
2. In Cluster Settings, click Heartbeats.
Adding volume heartbeats
1. Click the Volume Heartbeats tab,
2. Right click on a disk and select the vdev.
3. Click Save Settings.
Adding network heartbeats:
1. Click the Appliance Heartbeats tab,
2. Right click on Network heartbeats and select Add a network
hearbeat.
3. In the Create network heartbeat dialog, type the IP address or a
hostname available on a remote NexentaStor appliance.
4. Optionally, click Test.
5. Click OK.
6. Click Save Settings.
12
High Availability Cluster User Guide
Heartbeat and Network Interfaces
Adding serial heartbeats:
1. Click the Appliance Heartbeats tab,
2. Right click on Serial heartbeats and select Add a serial hearbeat.
3. In the Select serial ports for heartbeat dialog, select serial ports
on the local and remote NexentaStor appliances.
4. Click OK.
5. Click Save Settings.
пЃ¶
To change heartbeat properties, using NMC:
1. Type:
nmc:/$ setup group rsf-cluster <cluster_name>
hb_properties
System response:
•
Enable inter-appliance heartbeat through primary interfaces?:
Yes | No
•
Enable inter-appliance heartbeat through serial ports?: Yes |
No
•
Proceed: Yes | No
2. Follow the onscreen instructions.
High Availability Cluster User Guide
13
Heartbeat and Network Interfaces
This page intentionally left blank
14
High Availability Cluster User Guide
4
Configuring Storage Failover
This section includes the following topics:
•
About Configuring Storage Failover
•
Cluster Configuration Data
•
Mapping Information
•
NFS/CIFS Failover
•
Configuring iSCSI Targets for Failover
•
Configuring Fibre Channel Targets for Failover
About Configuring Storage Failover
HA Cluster detects storage system failures and transfers ownership of
shared volumes to the alternate NexentaStor appliance. HA Cluster ensures
service continuity in the presence of service level exceptional events,
including power outage, disk failures, appliance running out of memory or
crashing, etc.
Cluster Configuration Data
When you configure SCSI targets for either FC or iSCSI in a cluster
environment, make sure that you are consistent with configurations and
mappings across the cluster members. HA Cluster automatically propagates
all SCSI Target operations. However, if the alternate node is not available or
not configured at the time of the configuration change, problems can occur.
By default, the operation results in a warning to the User that the remote
update failed.
You can also set HA Cluster to synchronous mode. In this case, the action
fails completely if the remote update fails.
пЃ¶
To set the synchronous mode, using NMC:
1. Type:
nmc:/$ setup appliance nms property
rsf_config_update_synchronous
High Availability Cluster User Guide
15
Configuring Storage Failover
System response:
View or modify NMS property
�rsf_config_update_synchronous'. RSF-1 Appliance
configuration update mode. 1 - Strict Synchronous update
across the cluster, 0 - Asynchronous update, if synchronous
update fails. Navigate with arrow keys (or hjkl), or CtrlC to exit.
2. Select an appropriate value:
•
1 — Strict Synchronous update across the cluster
•
0 — Asynchronous update, if synchronous update fails.
To protect local configuration information that did not migrate, periodically
save this configuration to a remote site (perhaps the alternate node) and
then use NMC commands to restore it in the event of a failover.
The restore command restores previously saved configuration data that
includes:
•
Target groups
•
Host groups (stmf.config)
•
Targets
•
Initiators
•
Target portal groups (iscsi.conf)
Mapping Information
Use SCSI Target to map zvols from the cluster nodes to client systems. It is
critical that the cluster nodes contain the same mapping information.
Mapping information is specific to the volume and is stored with the volume
itself.
You can perform manual maintenance tasks on the mapping information
using the mapmgr command.
NFS/CIFS Failover
You can use HA Cluster to ensure the availability of NFS shares to users.
However, note that HA Cluster does not detect the failure of the NFS server
software.
NFS/CIFS settings are volume-level properties that migrate between nodes
automatically upon failover. However, settings such as idmap may need to
be defined on both nodes
HA Cluster does not detect CIFS server failures.
16
High Availability Cluster User Guide
Configuring Storage Failover
Configuring iSCSI Targets for Failover
You can use HA Cluster to failover iSCSI volumes from one cluster node to
another. The target IQN moves as part of the failover.
Setting up iSCSI failover involves setting up a zvol in the shared volume.
!
Note that you perform the process of creating a zvol and sharing it through
iSCSI separately from the HA Cluster configuration.
Note:
If you create iSCSI zvols before marking the zvol’s volume as a shared
cluster volume, then when you share the cluster volume as an active iSCSI
session, it may experience some delays. Depending on the network,
application environment and active workload, you may also see command
level failures or disconnects during this period.
When you add a shared volume to a cluster which has zvols created as back
up storage for iSCSI targets, it is vital that you configure all client iSCSI
initiators, regardless of the operating system, to access those targets using
the shared logical hostname that is specified when the volume service was
created, rather than a real hostname associated with one of the appliances.
Note that the cluster manages all aspects of the shared logical hostname
configuration. Therefore, do not configure the shared logical hostname
manually. Furthermore, unless the shared volume service is running, the
shared logical hostname is not present on the network, however, you can
verify it with the ICMP ping command.
пЃ¶
To configure iSCSI targets on the active appliance, using NMV:
1. Click Data Management > SCSI Target.
2. In the zvols panel, click create.
3. Make the zvol block size > 200MB.
HAC automatically migrates the newly created zvol to the other
appliance on failover. Therefore, you do not have to duplicate it
manually.
4. From the iSCSI pane, click iSCSI > Target Portal Groups and
define a target portal group.
!
Note:
It is critical that the IPv4 portal address is the shared logical hostname
specified when the volume service was created, instead of a real hostname
associated with one of the appliances.
HAC automatically replicates the newly created target portal group to the
other appliance.
пЃ¶
To create an iSCSI target and add it to the target portal group, using
NMV:
1. Click iSCSI > Targets.
This limits zvol visibility from client initiators to the target portal
group. The newly created iSCSI target is automatically replicated to
the other appliance.
High Availability Cluster User Guide
17
Configuring Storage Failover
2. Type a name and an alias.
The newly created iSCSI target displays in the Targets page.
пЃ¶
To create a LUN mapping to the zvol, using NMV:
1. From the SCSI Target pane, click Mappings.
This creates a LUN mapping to the zvol for use as backup storage for
the iSCSI target. The newly created LUN mapping is automatically
migrated to the other appliance on failover.
2. On the client, configure the iSCSI initiator to use both the IQN of the
iSCSI target created and the shared logical hostname associated
with both the volume service and the target portal group to access
the zvol through iSCSI.
Failover time varies depending on the environment. As an example,
initiating failover for a pool containing six zvols, the observed failover time
is 32 seconds. Nodes may stall while the failover occurs, but otherwise
recover quickly.
See Also:
•
“Managing SCSI Targets” in the NexentaStor User Guide
•
SCSI Target FC User Guide
Configuring Fibre Channel Targets for Failover
As a prerequisite for configuring Fibre Channel targets, change the HBA port
modes of both appliances from Initiator mode to Target mode.
пЃ¶
To change the HBA port mode, using NMV:
1. Click Data Management > SCSI Target Plus > Fibre Channel >
Ports.
2. Select Target from the Mode dropdown menu.
3. Once you change the HBA port modes of both appliances from
Initiator mode to Target mode, reboot both appliances so the Target
mode changes can take effect.
Miscellaneous Options for Fibre Channel
When using Fibre Channel, you can enable ALUA (Asymmetric Logical Unit
Access) mode. In Fibre Channel, devices are identified and accessed by their
WWN. Since the WWN is from the HBA that is accessed, HAC needs a
method for mapping and presenting LUNs through the FC HBAs in both
nodes. ALUA allows HAC to map and present LUNs through the FC HBAs in
both nodes, with the node that owns the zvol and volume actively serving
data, and the other HBA identifying itself as a "standby path" for each LUN.
The proxy daemon that runs on the nodes ensures that when you map a zvol
on one node, that HAC automatically maps a standby path on the standby
node.
18
High Availability Cluster User Guide
Configuring Storage Failover
пЃ¶
To enable ALUA mode:
1. In the Cluster settings panel, click Advanced.
2. Click Miscellaneous Options tab.
3. Select Enable ALUA mode.
See Also:
•
SCSI Target FC User Guide
High Availability Cluster User Guide
19
Configuring Storage Failover
This page intentionally left blank
20
High Availability Cluster User Guide
5
Advanced Setup
This section includes the following topics:
•
About Advanced Setup
•
Setting Failover Mode
•
Adding a Virtual IP Address and Hostname
•
Implementing Miscellaneous Options
About Advanced Setup
This section describes advanced functions of HA Cluster, such as setting the
failover mode, adding virtual hostnames and volumes, and other
miscellaneous options.
Setting Failover Mode
The failover mode defines whether or not an appliance attempts to start a
service when it is not running. There are separate failover mode settings for
each appliance that can run a service.
You can set the failover to the following modes:
•
Setting Manual Failover Mode
•
Setting Automatic Failover Mode
Setting Manual Failover Mode
In manual mode, the HA Cluster service does not initiate the failover
when it detects a failure. However, it generates warnings when the
parallel appliance is not available. If the appliance cannot obtain a
definitive answer about the state of the service, or the service is not
running anywhere else, the appropriate timeout must expire before you
High Availability Cluster User Guide
21
Advanced Setup
can take any action. The primary service failover modes are typically set
to automatic to ensure that an appliance starts its primary service(s) on
boot up. You may want to set the manual failover mode for maintenance.
!
Note:
Setting a service to manual mode when the service is already running does
not stop that service, it only prevents the service from starting on that
appliance.
пЃ¶
To set the failover mode to manual, using NMV:
1. Click Advanced Setup > Cluster Operations > Set all Manual.
2. Click Yes to confirm.
!
Note:
Before HAC performs an operation, it saves the state of the services in the
cluster, which you can later re-apply to the cluster using the restore
button. Once HAC restores the service state, HAC clears the saved state.
пЃ¶
To set the failover mode to manual, using NMC:
пЃµ
Type:
nmc:/$ setup group rsf-cluster <cluster_name> sharedvolume <volume_name> manual
Setting Automatic Failover Mode
In automatic mode, the appliance attempts to start the service when it
detects that there is no available parallel appliance running in the cluster.
AUtomati failover mode is the default setting.
пЃ¶
To set the failover mode to automatic, using NMV:
1. Click Advanced Setup > Cluster Operations > Set all Automatic
2. Click Yes to confirm.
пЃ¶
To set the failover mode to automatic, using NMC:
пЃµ
Type:
nmc:/$ setup group rsf-cluster <cluster_name> sharedvolume <volume_name> automatic
пЃ¶
To stop all services in the Cluster, using NMV:
1. Click Stop All Services.
2. Click Yes to confirm.
Adding a Virtual IP Address and Hostname
You can add a VIP, or shared hostname, when you create the Cluster. You
can also add additional VIPs later. Additional VIPs provide the access to a
shared volume using a different path.
пЃ¶
To add a virtual IP address, using NMV:
1. In the Cluster Settings panel, click Advanced.
22
High Availability Cluster User Guide
Advanced Setup
2. Click Additional Virtual Hostnames.
3. Select a shared volume from the drop-down list.
4. Click Add a new virtual hostname.
5. Type the virtual hostname and netmask.
6. Select the interface for each node.
7. Click Add.
8. If prompted, type the IP address of the failover node.
!
Note:
Type the IP address that is not in use and that is accessible from both
nodes of the HA Cluster. You can add the hostname and IP address pair to
the NexentaStor host tables.
See Resolving Hostname Conflicts.
9. Click Add.
10.Click Save Settings.
11.Click OK to confirm the modifications.
пЃ¶
To add a virtual IP address, using NMC:
1. Type:
nmc:/$ setup group rsf-cluster <HA Cluster> vips add
2. Select the HA Cluster service.
3. Type a virtual hostname.
4. If you type the IP address or hostname, that one or more HA Cluster
nodes cannot resolve, NexentaStor prompts you to modify the local
host tables.
•
If you want to modify the local host tables:
a. Type y.
b. Type the IP address and host name.
•
Alternatively, you can configure the DNS server settings.
a. Type n.
b. Log in to your DNS server and add the host name and IP address
pair to the DNS settings.
c. Repeat Step 1 — Step 3 and Step 5 — Step 10.
5. Select a network interaface for this node.
Nexenta recommends that you configure additional network
interfaces rather than specifying the primary network interface.
6. Select network interaface for the remote node.
7. Type the failover netmask.
8. Confirm the settings by typing y.
System response:
Stop adding VIPs?
(y/n)
High Availability Cluster User Guide
23
Advanced Setup
9. Type y to finish adding VIPs.
10.Alternatively, type n to add more VIps and repeat Step 1 — Step 9.
Implementing Miscellaneous Options
Miscellaneous components add an advanced level of control for fine-tuning
the HA cluster.
пЃ¶
To enable miscellaneous components:
пЃµ
Select the desired component(s):
•
Enable ALUA (Asymmetric Logical Unit Access) mode
Makes a SCSI target available from both nodes, even though it is
physically present on only one node. The target with the node is
considered active. The other node is considered passive.
When used in conjunction with client side multi-pathing, ALUA
ensures target rescanning is not required on failover, and also
that the path to the standby node is valid even when not in use,
(because the multi-path client continuously checks the state of
the standby node and the path to the standby target).
24
High Availability Cluster User Guide
6
System Operations
This section includes the following topics:
•
About System Operations
•
Viewing the HA Cluster Status
•
Manually Triggering a Failover
•
Verifying Shared Volume Status
•
Viewing Support Logs
About System Operations
There are a variety of commands and GUI screens to help you with daily
cluster operations. There is a set of cluster-specific commands to
supplement NMC.
Viewing the HA Cluster Status
You can view the status of the HA Cluster and heartbeats at any time.
пЃ¶
To view the HA Cluster configuration, using NMV:
1. In the Cluster Settings panel, click Status.
2. Click the tabs to view Cluster Status and Heartbeat Status.
пЃ¶
To view the HA Cluster configuration, using NMC:
1. Type:
nmc:/$ show group rsf-cluster <cluster_name>
Example:
nmc:/$ show group rsf-cluster HA-Cluster
System response:
PROPERTY
name
appliances
VALUE
: HA-Cluster
: [NodeA NodeB]
High Availability Cluster User Guide
27
System Operations
machinesigs
:
{"NodeA":"XXXXXXXXXX","NodeB":"YYYYYYYYYY"}
hbipifs
: NodeA:NodeB: NodeB:NodeA:
netmon
: 1
info
: Nexenta HA-Cluster
generation
: 1
refresh_timestamp
: 1375745271.30001
type
: rsf-cluster
creation
: Aug 5 16:27:51 2013
SHARED VOLUME: ha-vol
svc-ha-vol-shared-vol-name : ha-vol
svc-ha-vol-ipdevs
: ha-vol NodeA:e1000g0
NodeB:e1000g0
svc-ha-vol-ipdevs-IPv6
:
svc-ha-vol-attached-vols
:
svc-ha-vol-main-node
: NodeA
svc-ha-vol-inittimeout
: 20
svc-ha-vol-runtimeout
: 8
svc-ha-vol-mhdc-disable
: n
svc-ha-vol-monitor
:
{"NodeA":{"monitor":"","ipdevs":{"e1000g0":""}},"NodeB":{
"monitor":"","ipdevs":{"e1000g0":""}}}
svc-ha-vol-resdisks
: NodeA:c1t3d0 NodeB:c1t1d0
HA CLUSTER STATUS: HA-Cluster
NodeA:
ha-vol
running
auto
e1000g0
20 8
NodeB:
ha-vol
stopped
auto
e1000g0
20 8
unblocked
ha-vol
unblocked
ha-vol
Manually Triggering a Failover
You can manually trigger a failover between systems when needed.
Performing a failover from the current appliance to the specified appliance
causes the volume sharing service to stop on the current appliance, and the
opposite actions take place on the passive appliance. Additionally, the
volume exports to the other node.
пЃ¶
To manually trigger a failover, using NMC:
пЃµ
Type:
nmc:/$ setup group rsf-cluster <cluster_name> failover
Verifying Shared Volume Status
Verify the status on the shared volume service using NMV or NMC.
28
High Availability Cluster User Guide
System Operations
пЃ¶
To view the status of a shared volume, using NMV:
1. In the Cluster Settings panel, click Status.
пЃ¶
To view the status of a shared volume, using NMC:
пЃµ
Type:
[email protected]:/$ show group rsf-cluster
System response:
HA CLUSTER STATUS: HA-Cluster
nodeA:
vol1-114
stopped
manual unblocked 10.3.60.134 e1000g0
nodeB:
vol1-114
running
auto
unblocked 10.3.60.134 e1000g0
20 8
20
8
Viewing Support Logs
Gather the information about HA Cluster event or errors from the HA Cluster
log file.
пЃ¶
To view support logs, using NMV:
пЃµ
пЃ¶
Click View Log.
To view support logs, using NMC:
пЃµ
Type:
nmc:/$ show group rsf-cluster <cluster name> log
High Availability Cluster User Guide
29
System Operations
This page intentionally left blank
30
High Availability Cluster User Guide
7
Testing and Troubleshooting
This section includes the following topics:
•
Repairing a Broken Cluster Service
•
Replacing a Faulted Node
Repairing a Broken Cluster Service
NexentaStor tracks various appliance components, and their state. If and
when failover occurs (or any service changes to a broken state),
NexentaStor sends an email to the administrator describing the event.
!
During the NexentaStor installation, you set up SMTP configuration and
test so that you can receive emails from the appliance.
Note:
There are two broken states:
•
Broken_Safe
A problem occurred while starting the service on the server, but it
was stopped safely and you can run it elsewhere.
•
Broken_Unsafe
A fatal problem occurred while starting or stopping the service on the
server. The service cannot run on any other server in the cluster until
it is repaired.
!
Warning:
Manually verify and troubleshoot the volume before marking the state as
repaired. Failure to do so could result in cross-mounting of the volume and
lead to data corruption.
пЃ¶
To repair a shared volume which is in broken state, using NMC:
nmc:/$ setup group rsf-cluster shared-volume repair
<cluster_name> <volume_name>
This initiates and runs the repair process.
High Availability Cluster User Guide
31
Testing and Troubleshooting
пЃ¶
To mark a service as repaired, using NMV:
1. Click Settings > HA Cluster.
2. In the Action column, set the action to repaired.
3. Click Confirm.
Replacing a Faulted Node
NexentaStor provides advanced capabilities to restore a node in a cluster, in
case the state changes to out of service. There is no need to delete the
cluster group on another node and reconfigure it and all of the cluster
services.
пЃ¶
To replace a faulted node, using NMC:
1. Type:
nmc:/$ setup group rsf-cluster <group_name> replace_node
After executing the command, the system asks you to choose which node
to exclude from the cluster and which new node to use instead. The system
checks the host parameters of the new node and if they match the
requirements of the cluster group, it replaces the old one.
!
Note:
32
Before performing a replace node operation, you must set up the identical
configuration on the new or restored hardware, which HA Cluster uses to
replace the old faulted node. The new node must be bind to the existing
node using SSH. Otherwise, the operation fails. You must also make the
serial port heartbeats configuration the same.
High Availability Cluster User Guide
Index
A
adding
hostname 22
hostname address 22
shared volume 8
virtual IP address 22
additional resources 4
B
binding
nodes 7
C
cluster
configuration data 15
configuration data
cluster 15
configuring
fibre channel
failover 18
HA cluster 5, 8
iSCSI Targets
failover 17
storage failover 15
D
defining
netmasks 6
E
exclusive access
to storage 2
F
failover
configuring iSCSI
Targets 17
NFS/CIFS 16
service 3
storage 2
triggering
manually 28
failover mode
setting 21
fibre channel
failover configuring 18
misc options 18
G
guide
audience v
H
HA cluster
configuring 5, 8
heartbeat interfaces
modifying 12
heartbeat mechanism 11
hostname
adding 22
hostname conflicts
resolving 6
I
implementing
misc options 24
M
mapping information 16
misc options
implementing 24
modifying
heartbeat interfaces 12
failover mode 21
shared disk 12
shared volume
adding to cluster 8
status 28
SSH
binding the nodes 7
storage
failover 2
failover configuring 15
support
contact v
system operations 27
T
triggering
failover 28
V
virtual IP address
adding 22
N
netmasks
defining 6
network architecture 3
sample 3
nexentastor
appliances 1
NFS/CIFS failover 16
nodes
binding 7
R
resolving hostname
conflicts 6
RSF-1 cluster service 1
S
scsi-2 pgr 3
service
failover 3
setting
High Availability Cluster User Guide
33
Index
This page intentionally left blank
34
High Availability Cluster User Guide
Global Headquarters
455 El Camino Real
Santa Clara, California 95050
Nexenta EMEA Headquarters
Camerastraat 8
1322 BC Almere
Netherlands
Houston Office
2203 Timberloch Pl. Ste. 112
The Woodlands, TX 77380
Nexenta Systems Italy
Via Vespucci 8B
26900 Lodi
Italy
Nexenta Systems China
Room 806, Hanhai Culture Building,
Chaoyang District,
Beijing, China 100020
Nexenta Systems Japan
гЂ’ 102-0083
Chiyodaku koujimachi 2-10-3
Tokyo, Japan
Nexenta Systems Korea Chusik Hoesa
3001, 30F World Trade Center
511 YoungDongDa-Ro
GangNam-Gu, 135-729
Seoul, Korea
3000-hac-v3.1.x-000031-A
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement