ucp doc 4.1 administrator's guide...contents preface v intendedaudience v productversion v...

Unified Compute Platform 4.1UCP DOC Administrator's Guide

MK-92UCP054-03

© 2014 - 2015 Hitachi Data Systems Corporation. All rights reserved.

No part of this publication may be reproduced or transmitted in any form or by any means, electronic ormechanical, including photocopying and recording, or stored in a database or retrieval system for anypurpose without the express written permission of Hitachi Data Systems Corporation (hereinafter referredto as “Hitachi Data Systems”).

Hitachi Data Systems reserves the right to make changes to this document at any time without notice andassumes no responsibility for its use. This document contains the most current information available at thetime of publication. When new or revised information becomes available, this entire document will beupdated and distributed to all registered users.

Some of the features described in this document might not be currently available. Refer to the most recentproduct announcement for information about feature and product availability, or contact Hitachi DataSystems at https://portal.hds.com.

Notice: Hitachi Data Systems products and services can be ordered only under the terms and conditions ofthe applicable Hitachi Data Systems agreements. The use of Hitachi Data Systems products is governed bythe terms of your agreements with Hitachi Data Systems.

By using this software, you agree that you are responsible for:

a) Acquiring the relevant consents as may be required under local privacy laws or otherwise fromemployees and other individuals to access relevant data; and

b) Ensuring that data continues to be held, retrieved, deleted, or otherwise processed in accordance withrelevant laws.

https://portal.hds.com/

Contents

Preface vIntended audience vProduct version vAccessing product documentation vRelated documents vComments viGetting help vi

Chapter 1: UCPDOCoverview 1Introduction to UCP DOC and UCP Disaster Recovery 1UCP Disaster Recovery overview 2What's new in UCP DOC 2

Chapter 2: Using the UCPDOCConsole 7Navigating through the UCP DOC Console 7

Downloading the UCP DOC CLI 8Accessing the UCP DOC Console 9

Permissions required to use the console 9Administering users and groups 9Administering sites 9

Monitoring site health 10How to... 11

Add a site 11Add a new user or group and grant them access 12Edit AMQP credentials 12Edit a site 12Edit an existing user or group to grant them access 13Remove a site 13

Contents iiiUCP DOC Administrator's Guide

Remove a user or group and their access 13View a site 13View a user or group and their access permissions 14

Chapter 3: Using UCPDisaster Recovery 15UCP Disaster Recovery volume replication 15

Replication technology and outage tolerance 17Synchronization status 18

Using UCP Disaster Recovery with vCenter 18Test volumes 21

UCP Disaster Recovery administration 23Administering site pairs 24Administering replication groups 24Best practices for setting up replication groups 24Failover scenarios 24API 26CLI 26

Administering volume replication 27Administering test volumes 27

How to... 27Create a site pair 27Create a test volume 28Expand a volume 28Pair and resynchronize a replication group 29Refresh site pair inventory 30Remove volume replication 30Remove a replication group 30Remove a site pair 31Remove a test volume 31Set up a replication group 32Set up volume replication 32

Chapter 4: Tracking UCP DOC jobs and events 35Displaying UCP DOC jobs 35Displaying UCP DOC events 36List of UCP DOC jobs 39List of UCP DOC events 41

iv ContentsUCP DOC Administrator's Guide

PrefaceThis guide explains the function and administration of Hitachi Unified ComputePlatform Director Operations Center (UCP DOC). The information containedis intended to help UCP administrators better understand how UCP DOC andHitachi UCP Disaster Recovery works, including concepts that are needed toconfigure and administer UCP DOC both with and without UCP DisasterRecovery. It is not a solution guide or a replacement for the release notes.

Intended audienceThis guide is intended for datacenter administrators who monitor the status ofmultiple UCP sites and who configure and administer UCP Disaster Recovery. Itassumes that they are familiar with UCP, the UCP sites they are administering,and the hardware used at a UCP site.

Product versionThis guide applies to release 4.1 of UCP.

Accessing product documentationIn this release, we are introducing the new UCP Documentation Portal(http://ucp.io/docs/4.1/). The documentation portal is a search-enabled onlineresource containing the complete set of product documentation for UCP 4000and 4000E systems running UCP Director Operations Center 4.1 software.

To ensure you are reading the most up-to-date UCP 4.1 product documentation,we highly recommend that you search the online portal first, before reviewingany offline documentation. Check this site for the most current documentation,including important updates that may have been made after the release of theproduct.

You can also download the most current UCP 4.1-related PDFs(http://ucp.io/docs/4.1/#PDFDownloads.htm).

Related documentsThe following documents contain additional information about UCP:

Preface vUCP DOC Administrator's Guide

http://ucp.io/docs/4.1/http://ucp.io/docs/4.1#PDFDownloads.htm

• UCP Pre-Installation Requirements and Configuration Guide — Containsinformation and procedures you need to be aware of for a successful UCPinstallation.

• UCP Network Architecture and Configuration Guide — Contains technicalreference information about the networking architecture of UCP systemsand provides procedures to help you configure the system to work in yournetwork environment.

• UCP Director Administrator's Guide — Contains technical and usageinformation for UCP and UCP Director. Describes how to administer UCPDirector through UCP Director web interface with both VMware vCenter andMicrosoft SCVMM.

• UCP Director API Reference Guide — Describes how to use the UCP DirectorAPI.

• UCP Director CLI Reference Guide — Describes how to use the UCP DirectorCLI.

• UCP Director Third-Party Copyrights and Licenses — Contains copyright andlicense information for the third-party software distributed with orembedded in UCP Director.

• UCP DOC Administrator's Guide — Contains technical and usage informationfor Unified Compute Platform Director Operations Center (UCP DOC).Describes how to administer UCP DOC through UCP DOC Console.

• UCP DOC API Reference Guide — Describes how to use the UCP DOC API.

• UCP DOC CLI Reference Guide — Describes how to use the UCP DOC CLI.

CommentsPlease send us your comments on this document:

[email protected]

Include the document title and number, including the revision (for example, -01), and refer to specific sections and paragraphs whenever possible. Allcomments become the property of Hitachi Data Systems.

Thank you!

Getting helpHitachi Data Systems Support Portal is the destination for technical support ofproducts and solutions sold by Hitachi Data Systems. To contact technicalsupport, log on to Hitachi Data Systems Support Connect for contactinformation: https://support.hds.com/en_us/contact-us.html.

Hitachi Data Systems Community is a global online community for HDScustomers, partners, independent software vendors, employees, andprospects. It is the destination to get answers, discover insights, and makeconnections. Join the conversation today! Go to community.hds.com, register,and complete your profile.

Comments

vi PrefaceUCP DOC Administrator's Guide

mailto:[email protected]://support.hds.com/en_us/contact-us.htmlhttps://support.hds.com/en_us/contact-us.htmlhttps://community.hds.com/http://community.hds.com/

UCP DOC overviewUCP Director Operations Center (DOC) is hosted on the UCPDatacenter virtualmachine (VM), which is hosted in the management block of one of the sites.Sites that are added to UCP DOC can be from one or more different physicallocations. For more information on component health and the managementblock, see UCP Administrator's Guide.

In addition to UCP DOC, the UCPDatacenter VM contains the followingsupporting software:

• AMQP server

• SQL express

To function correctly, the UCPDatacenter VM needs to have network access toeach site that is added, as well as the Microsoft Active Directory server used byUCP Director at the sites to authenticate users.

UCP DOC can be administered visually Using the UCP DOC Console on page 7,or programmatically through the UCP DOC API and UCP DOC CLI. For moreinformation, see the UCP DOC API Reference Guide or UCP DOC CLI ReferenceGuide.

• Introduction to UCP DOC and UCP Disaster Recovery below

• UCP Disaster Recovery overview on the next page

• What's new in UCP DOC on the next page

Introduction to UCP DOC and UCP Disaster RecoveryUCP DOC is a powerful tool that enables you to monitor and review the status ofone or more UCP installations, or sites, from a single window. By integrating itwith UCP Disaster Recovery, it also enables the administration and automationof volume replication between sites.

For more information, see UCP Disaster Recovery overview on the next page.

Chapter 1: UCP DOC overview 1UCP DOC Administrator's Guide

UCP Documentation Portal (http://ucp.io/docs/4.1/)

1

http://ucp.io/docs/4.1/

UCP Disaster Recovery overviewUCP Disaster Recovery automates the configuration of volume replication in aresource group between two sites in a site pair. Site pairs are created withinUCP Disaster Recovery by pairing two properly configured sites that areregistered in UCP DOC.

When two sites have been paired, you can use the site pair to replicate volumesfrom one site to another. This way, when changes are made to the first site, theprotected site, they are automatically copied to the second site, the recoverysite.

Because UCP DOC collects health monitoring information on the volumes thatyou pair, UCP Disaster Recovery will then record and monitor the specifichealth of the volumes being replicated.

Each UCP site abstracts the individual storage system. As a result, sites do notneed to use the same storage system model. Instead, UCP Disaster Recoveryonly requires each of the sites in a site pair to use a Hitachi enterprise-levelstorage system.

For more information, see Using UCP Disaster Recovery on page 15.

What's new in UCP DOCThis section describes the new features that were released with version 4.1 ofUCP Director Operations Center (DOC).

Disaster Recovery support for the Hitachi VSP G800 storagearray

In UCP 4.1 systems, the Hitachi Virtual Storage Platform (VSP) G800 storagearray, along with the G400 and G600 models, come with full support fordisaster recovery replication. They will replace the block storage array modelsHUS-VM and VSP that shipped with 4.0 and earlier systems. The G200 modeldoes not support disaster recovery replication.

Planned and unplanned failover functionalityNew functionality for bare-metal and Hyper-V workloads to support bothplanned and unplanned failover scenarios to a remote site.

For more information, see Failover scenarios on page 24.

Exporting UCP DOC logsUCP DOC provides new functionality to help administrators more efficientlycollect and download logs to their system. This feature collects the jobs list,event list and logs for UCP DOC and archives them to one file. It is especiallyuseful in cases where you need to transfer logs to the HDS support team fortheir assistance in remotely troubleshooting your UCP system.

2 Chapter 1: UCP DOC overviewUCP DOC Administrator's Guide


UCP Disaster Recovery overview


In the UCP DOC 4.1 release, you can only enable this feature using theUCP DOC APIs or CLIs. For a summary of the options you can use with thisfeature and when to use them, see the following table.

When to use... API CLIUse this when you need to specify thecondition of collected logs.

This CLI or API can be used to specifythe kind of logs you want to collect andthe time duration you want to use.When you submit the request throughthe API, you can get the ID of the jobfor collecting logs and archiving. Whenthe job is completed, you can find theID of the created log collection bundlein the job's related event. When youexecute the CLI, it blocks until the jobfor collecting logs and archivingcompletes, and then returns theinformation for the created logcollection bundle.

Generate logbundle

URI is“/api/logs/bundles,method is “POST”

New-UCPDatacenterLogbundle

Use this when you need to check thestatus of a log collection bundle.

After you have generated a log bundle,you need to specify the log bundle IDto download/remove it. This CLI or APIcan be used to check the status of thelog collection bundle you are about todownload/remove. You can list all ofthe log bundle information stored inthe system or get a single log bundleby specifying the ID.

Get list of loginformation

URI is“/api/logs/bundles,method is “GET”

OR

URI is“/api/logs/bundles/{log bundle ID}”,method is “GET”

Get-UCPDatacenterLogBundle

Use this when you need to downloadthe log bundle.

When “status” in log bundleinformation is “ReadyToDownload”,UCP DOC finishes making an archivedfile.

Download logbundle

URI is/api/download/logs/{log bundle ID},method is “GET”

Export-UCPDatacenterLogBundle

Use this when you need to remove thelog bundle file.

Remove log bundleRemove-UCPDatacenterLogBundle



What's new in UCP DOC


When to use... API CLIAfter getting the archived file, youshould remove the file existing on theUCPDatacenterVM. Since the size of logfile is large, this feature monitors thevolume of archived logs. When thevolume size will exceed the specifiedvalue, you can not invoke new jobs.

URI is/api/logs/bundles/{log bundle ID},method is“DELETE”

For more information about the log bundle APIs, see the UCP DOC APIReference Guide. For more information about the log bundle CLIs, see theUCP DOC CLI Reference Guide.

Support for SRM 6.0 with VMware 6.0SRM 6.0 is now supported with vSphere 6.0 appliances

Updated UCP 4.1 documentation on the WebIn this release, we are introducing the new UCP Documentation Portal(http://ucp.io/docs/4.1/). The documentation portal is a search-enabled onlineresource containing the complete set of product documentation for UCP 4000and 4000E systems running UCP Director Operations Center 4.1 software. Inaddition to offering rich search and online viewing capabilities, the portal hoststhe most up-to-date UCP 4.1 PDFs.

When you should use the portalTo ensure you are reading the most up-to-date UCP 4.1 product documentation,we highly recommend that you search the online portal first, before reviewingany offline documentation. We understand that with our product there will likelybe times when you are physically at the system and may not have internetaccess, yet need access to the latest content.

In these cases, we suggest that you first download the latest PDFs from theonline documentation portal onto a laptop or portable device so that you canbring them with you before physically walking to the system. To download themost current UCP 4.1-related PDFs, see the PDF Download page(http://ucp.io/docs/4.1/PDFDownloads.htm).

Additional benefitsIn addition, the portal also provides the following benefits:

• Search filters to enable more precisely scoped queries of UCP content.

For example, a search for "network switches" may show hundreds of topicsin search results. By selecting a search filter, say the 'API Reference'search filter, and then reapplying the same query would show you a muchmore refined list of results that are directly applicable to the filter scopeyou chose.




http://ucp.io/docs/4.1/http://ucp.io/docs/4.1#PDFDownloads.htmhttp://ucp.io/docs/4.1/

• Enables the UCP Documentation Team to collect detailed analysis, search-related metrics, and other extremely useful statistics that will help us tofurther improve the usability and discoverability of current and futureproduct documentation.





Using the UCP DOC ConsoleUCP DOC Console is the stand-alone web-based graphical interface that you canuse to administer UCP DOC and, when installed, UCP Disaster Recovery.

This section explains how to use UCP DOC Console to administer UCP DOC.

• Navigating through the UCP DOC Console below

• Accessing the UCP DOC Console on page 9

• Administering users and groups on page 9

• Administering sites on page 9

• How to... on page 11

Navigating through the UCP DOC ConsoleUCP DOC Console is divided into three functional areas, as follows:

• Info bar — A black bar at the top of UCP DOC Console that displays thefollowing:

¡ The Helpmenu, which contains links to product guides, the CLI installer,and the about page. The following product guides are included in theHelpmenu:

– UCP DOC Administrator's Guide

– UCP DOC API Reference Guide

– UCP DOC CLI Reference Guide

– UCP Administrator's Guide

For more information on downloading the CLI installer, see Downloadingthe UCP DOC CLI on the next page.

¡ The domain and username of the user that is logged into UCP DOCConsole.

¡ The date and time the page was loaded.

Chapter 2: Using the UCP DOC Console 7UCP DOC Administrator's Guide


2


• Navigation bar — A large bar located at the bottom of UCP DOC Console.

The navigation bar is divided into the following sections: Dashboard,Configuration, Disaster Recovery, and Jobs and Events. These sectionscontain related information and links, as follows:

¡ Dashboard— Displays the aggregate monitoring state of all componentsacross all sites administered by UCP DOC. For more information on themonitoring state, see Monitoring site health on page 10.

¡ Configuration — Contains links that enable you to add a UCP site, add auser or group, or edit the AMQP credentials.

¡ Disaster Recovery — Contains links to create a site pair or refresh sitepair inventory.

¡ Jobs and Events —Displays the most recent jobs and events that havebeen run by UCP DOC.

Click on the title of a section to display the corresponding page in the bodyarea.

• Body — Located between the info and navigation bars, the body is used todisplay the contents of the page that you are viewing. UCP DOC Console hasfour primary pages, as follows:

¡ Dashboard— Displays each site monitored by UCP DOC. Each sitedisplays the aggregate monitoring state of all components in the systemas well as the aggregate monitoring state of each component type. Formore information on aggregate component health and the Dashboardpage, see Monitoring site health on page 10.

¡ Configuration — Used to add and remove users, groups, and sites. Formore information on the Configuration page, see How to... on page 11.

¡ Disaster Recovery — Used to automate volume replication between twosites.

Note: UCP Disaster Recovery is optional. If it is not installed, the DisasterRecovery section of the navigation bar will indicate that it is not installed and youwill not be able to view the Disaster Recovery page.

¡ Jobs and Events — Displays the status of all UCP DOC jobs. For moreinformation on the Jobs and Events page, see Tracking UCP DOC jobsand events on page 35.

Downloading the UCP DOC CLI

To access the UCP DOC CLI, you will need to download the UCP DOC CLIinstaller and install it from within UCP Director web interface. To download theUCP DOC CLI installer, from the Helpmenu, click on the Download CLI link. Formore information on the UCP DOC CLI, see UCP DOC CLI Reference Guide.

8 Chapter 2: Using the UCP DOC ConsoleUCP DOC Administrator's Guide


Navigating through the UCP DOC Console


Accessing the UCP DOC ConsoleTo access the UCP DOC Console, while using an account with sufficientprivileges, navigate to the following URL:

https://UCPDATACENTER/ui

Where UCPDATACENTER is the IP address of the UCPDatacenter VM.

For more information about UCP DOC privileges, see Permissions required touse the console below. For more information about accounts, see Administeringusers and groups below.

Permissions required to use the console

Access to UCP DOC is role-based. To connect to UCP DOC Console, your ADaccount needs to:

• Have an appropriate role applied to it

• Be added to a group that has the role applied to it in UCP DOC.

UCP DOC supports the following roles:

• Administrator — Able to administer all portions of UCP DOC and UCPDisaster Recovery.

• Viewer — Enables read-only access to UCP DOC.

Administering users and groupsThe Users/Groups tab of the Configuration page is used to display, add, edit,and remove user accounts and groups in UCP DOC. For more information, seethe following procedures.

• Add a new user or group and grant them access on page 12

• Edit an existing user or group to grant them access on page 13

• Remove a user or group and their access on page 13

• View a user or group and their access permissions on page 14

Administering sitesWhen a UCP site is registered with UCP Director Operations Center (DOC), UCPDOC will begin to collect monitoring information on that site. The healthinformation for each site is then aggregated together by component type foreach site that is added to UCP DOC.

By aggregating the health information for each component type at a site, UCPDOC is able to track the health of the storage system as well as all of theservers and switches at a site. For more information, see Monitoring site healthon the next page.



Accessing the UCP DOC Console


Before you can use UCP DOC to monitor a site or use UCP Disaster Recovery toconfigure a site for disaster recovery, you need to add that site to UCP DOC.

Use the following procedures when you need to administer sites.

• Add a site on the facing page

• Edit a site on page 12

• Remove a site on page 13

• View a site on page 13

Monitoring site health

The Dashboard page is used to display the status of sites that have been addedto UCP DOC.

To display the Dashboard page, from the navigation bar, click on the title of theDashboard section.

For each site, in addition to the Serial Number and UCPManagement URLfields, the following site properties are shown:

• Monitoring State — An aggregate of all component monitoring states that isequal to the most concerning monitoring state of any individual componentat the site. From most concerning to least concerning, values can be: Error,Warning, Unknown, and Healthy. If a site is unreachable, the MonitoringState of the site will not be available and will appear as unreachable.

For example, if all components are Healthy, then the monitoring state willbe Healthy, but if even one component is in aWarning or Error state, thenthe entire site will appear asWarning or Error.

• Component health — Below the overall site properties, the aggregatemonitoring state of each component type at the site is shown in a table. Foreach component type, the table shows the total number in each monitoringstate, as follows:

¡ Error ( ) — UCP Director is not able to communicate with theindicators or the indicators are not functioning properly.

¡ Warning ( ) — UCP Director is able to communicate with the indicatorsbut there is a issue. In the case of storage, server, or chassis resources,this means that the corresponding element manager has detected awarning.

¡ Unknown ( ) — UCP Director is not able to determine the status of theindicator.

¡ OK ( ) — UCP Director is able to communicate with the indicators andthe indicators are functioning properly.



Administering sites


For more information on administering a site, including UCP monitoring statesand component types, see UCP Administrator's Guide.

How to...User/group administration and site access is administered on the Sites andUsers/Groups tabs of the Configuration page. To access the Configurationpage, from the navigation bar, click on the Configuration link.

This section covers the following procedures.

• Add a site below

• Add a new user or group and grant them access on the next page

• Edit AMQP credentials on the next page

• Edit a site on the next page

• Edit an existing user or group to grant them access on page 13

• Remove a site on page 13

• Remove a user or group and their access on page 13

• View a site on page 13

• View a user or group and their access permissions on page 14

Add a site

Before you can use UCP DOC to monitor a site or use UCP Disaster Recovery toconfigure a site for disaster recovery, you need to add that site to UCP DOC.

To add a site to UCP DOC:

1. From the Sites tab on the Configuration page, click on the Add UCP Siteicon ( ).

2. On the Add UCP Site window:

¡ In the UCP Site Name field, type a unique name that will be used tolabel the site in UCP DOC.

¡ In the UCPManagement URL field, type the URL of the UCPManagementVM of the site.

¡ In the Username field, type the username of a user with administrativeaccess to the vCenter and UCPManagement VMs in the following format:User@Domain.

¡ In the Password field, type the password that corresponds to thespecified account in the Username field.

3. Click on the OK button.



How to...


Add a new user or group and grant them access

Use the following procedure to add a user or group and grant it access to theUCP DOC console.

1. From the Users/Groups tab on the Configuration page, click on the AddUser/Group icon ( ).

2. In the User/Group field, type the name of the user or group to add to UCPDOC. The username must be entered in the following format:

Domain\Username

3. As appropriate, select either the Administrator or Viewer role.


Edit AMQP credentials

Advanced Message Queuing Protocol (AMQP) is a service that is run on theUCPManagement VM of each site. UCP DOC subscribes to the AMQP messagingservice at each site that is added to UCP DOC to receive disaster recovery-events.

For security reasons, the credentials used to connect to AMQP can be changed,but all sites must use the same credentials.

To configure AMQP credentials, after they have been configured on the AMQPservice:

1. In the Configuration section of the navigation bar, click on the AMQPCredentials button.

2. On the AMQP Credentials dialog:

¡ In the Username/Group field, type the username or group withadministrative access to the AMQP server.

¡ In the Password field, type the password that corresponds to thespecified account in the Username field.


Edit a site

To edit a site in UCP DOC:

1. From the Sites tab on the Configuration page, click on the Edit icon ( )next to the site that you want to edit.

2. On the Edit UCP window, make the appropriate changes as explained in Adda site on the previous page.




How to...


Note: The site name of the site is shown at the top of the edit window.

Edit an existing user or group to grant them access

Use the following procedure to edit a user or group and grant them access tothe UCP DOC console.

1. From the Users/Groups tab on the Configuration page, click on the Editicon ( ) next to the user that you want to edit.

2. Select either Administrator or Viewer access for the user.


Remove a site

A site can not be removed if it is part of a site pair.

When a site is removed, all jobs and monitoring information for that site will beretained. If the site is added back to UCP DOC, the saved jobs and monitoringinformation will be re-associated with the site.

To remove a site from UCP DOC:

1. From the Sites tab on the Configuration page click on the Remove icon () next to the site that you want to remove.

2. Click on the Yes button.

Remove a user or group and their access

Use the following procedure to remove a user or group and their access to theUCP DOC console.

1. From the Users/Groups tab on the Configuration page, click on theRemove icon ( ) next to the user that you want to remove.


View a site

The Sites tab of the Configuration page is used to display, add, edit, andremove sites in UCP DOC.

To display the Sites tab of the Configuration page, from the navigation bar:

1. Click on the title of the Configuration section.

2. Click on the Sites tab at the top of the page.



How to...


For each site, the following properties are displayed:

• Serial Number — The serial number of the site.

• UCPManagement URL — The URL of the UCPManagement VM at the site.

View a user or group and their access permissions

Use the following procedure to view a user or group and their access to theUCP DOC console.

1. From the navigation bar, click on the title of the Configuration section.

2. Click on the Users/Groups tab at the top of the page.

For each user, the following properties are displayed:

• User/Group— The domain and user or group ID.

• Role — The role that the user or group has been assigned to.



How to...


Using UCP Disaster Recovery When using vCenter, UCP Disaster Recovery automates the complex task ofmanually using CCI commands to configure volume replication between twosites in a site pair. This enables volumes from the storage system at one site,the primary site, to be automatically replicated to the storage system at asecond site, the recovery site.

This section explains how UCP Disaster Recovery functions, and how to use UCPDOC Console to administer UCP Disaster Recovery.

• UCP Disaster Recovery volume replication below

• Using UCP Disaster Recovery with vCenter on page 18

• UCP Disaster Recovery administration on page 23

• How to... on page 27

UCP Disaster Recovery volume replicationBecause the storage system is abstracted by UCP Director, sites do not need touse the same storage system model. Instead, UCP Disaster Recovery requiresthat both sites use supported enterprise-class storage systems.

In addition, UCP Disaster Recovery requires that both UCP sites, as well as UCPDatacenter Operations Center, all have identical software versions. Forexample, version 4.1 on both sites and 4.1 on UCP Datacenter OperationsCenter.

A site pair is a logical construct that is used to pair two sites together. Each sitecan only be a member of one site pair, and a site pair is only able to containtwo sites. After sites have been paired together, replication groups need to becreated to contain the volumes that will be replicated.

Replication groups are logical collections of volumes that are replicated fromone site to another. It is at the replication group level that the primary andrecovery sites are defined. As a result, two sites are able to perform reciprocalbackups for each other through the configuration of multiple resource groups.

Chapter 3: Using UCP Disaster Recovery 15UCP DOC Administrator's Guide


3


Because replication can be bi-directional if more than one replication group isadded, volumes are identified by their role in the replication group.

• Protected volume — The volume at the protected site. This is the primarydata source.

• Recovery volume — A volume at the recovery site that contains thereplicated data.

• Test volume — A test volume can be added to recovery groups that containonly ESXi hosts. Test volumes are used to test recovery scripts andprocesses without affecting the recovery volume. For more information ontest volumes, see Test volumes on page 21.

When volume replication has been set up, changes to the protected volume arereplicated to the recovery volume. The following diagram shows thecomponents and connections involved to replicate volumes between two sites.

16 Chapter 3: Using UCP Disaster Recovery UCP DOC Administrator's Guide


UCP Disaster Recovery volume replication


After volume replication has been setup, UCP Disaster Recovery protects thevolumes that are being replicated. UCP Director does this by blocking anytasks, such as detaching a volume from a host, that might compromise volumereplication. This is done to prevent accidentally interrupting volume replication.

Because UCP DOC collects health monitoring information on the volumes thatyou pair, UCP Disaster Recovery will then record and monitor the specifichealth of the volumes being replicated.

Replication technology and outage tolerance

UCP Disaster Recovery supports both synchronous and asynchronousreplication technology, depending on the needs of the site.

• Synchronous — Hitachi TrueCopy Synchronous Remote Replication

Synchronous replication is designed for fast, robust, real-time replicationwithin most metropolitan distances. When using synchronous replication,data is written to both sites at the same time and the data that is written atthe primary site is not confirmed as successful until the write at the backupsite has been confirmed as well.

• Asynchronous — Hitachi Universal Replicator

Asynchronous replication is designed for scenarios where synchronousreplication is not possible, such as when the distance is excessive or whenconnection latency is a concern. When using asynchronous replication,instead of synchronously writing data to both the primary and recoveryvolumes, changes to replicated data are tracked in journals until writes canbe confirmed.

UCP uses a single journal stored in a single HDP pool as the journal source.The journal at a site is then used to contain all journal volumes at that site.When a journal volume is used, changes to the protected volume arecollected in the journal volume at the protected site. The data is thentransfered to the journal volume at the recovery site, where it can bewritten to the recovery volume.

For the G1000 storage system, the size of the journal volumes can bevariable. For other enterprise Hitachi storage systems, the size of thejournal volumes is determined by the maximum outage tolerance betweenthe two sites. The greater the outage tolerance, the larger the journalvolumes need to be to accommodate data changes.

The outage tolerance is the number of minutes worth of data changes duringan outage that are OK to be held in journal volumes before they arereplicated. Because of this, larger journal volumes help prevent the loss ofreplicated data if two sites are temporarily unable to communicate or iflatency causes a backlog of changes.



UCP Disaster Recovery volume replication


Synchronization status

UCP Disaster Recovery tracks the status of the volumes that it synchronizesand reports them as follows:

• Paired — Indicates that an active replication link exists between theprotected volume and its replicated volume or between the replicatedvolume and its test volume.

• Copy — Indicates that the primary side volume is actively copying data tothe secondary (replicated) side volume or between the replicated volumeand its test volume.

• Simplex — When using copy on write volumes, indicates a failure inreplication. This occurs by default before data starts being replicated aftervolumes have been paired, but can also happen after a failover. When inthis state, the test volume will require a pair resynchronization. For moreinformation on copy on write volumes, see Test volumes on page 21.

• Pair Suspend Split — Indicates that replication between the protectedvolume and recovery volume or between the recovery volume and a testvolume is paused, typically via user intervention.

• PairSuspendError — Indicates that replication between the protectedvolume and recovery volume or between the recovery volume and a testvolume is paused, typically when an error occurs. This can happen if adisaster has occurred at either site or if the connection between thereplicated pairs is down.

• SuspendForSwapping — Indicates that replication has been suspended andfailed over to the secondary side. This mode is intended to convert thesecondary side to become writable, as well as protect the secondary sidefrom being written over by the primary side in the case that replication isrestored.

• Pair with Journal Full Warning — This state is a warning to the user that ajournal being used for replication is at 80% capacity. This status will beseen when the replication link is in asynchronous Hitachi UniversalReplication mode.

• Pair Suspend with Journal Full Error — Indicates that replication betweenthe protected volume and recovery volume or between the recovery volumeand a test volume is paused due to the journal buffer being full.

Using UCP Disaster Recovery with vCenterAfter configuring volume replication:

• For replication groups that contain non-ESXi hosts, you can configure aseparate automated failover solution that makes use of the replicatedvolumes.



Using UCP Disaster Recovery with vCenter


• For replication groups that contain only ESXi hosts, UCP Disaster Recoverycan automate the creation of test volumes. You can then use test volumeswith VMware Site Recovery Manager (SRM) for the purpose of testingrecovery plans.

To enable the creation of test volumes, the REPL VM is added to themanagement block at each site. The REPL VM hosts the following services:

• VMware Site Recovery Manager (SRM).

• Hitachi Storage Replication Adapter (SRA) — Integrated with SRM to issueCCI commands to enable the testing of recovery plans.

• Hitachi Command Control Interface (CCI).

An SRM test plan can be configured to use a test volume. If the replicationgroup does not use test volumes, the SRM test plan can be configured to use thereplicated volumes for. Executing a test plan enables you to test the automatedfailover of the primary site. Automated failover enables a backup site tofunction as a primary site if the primary site becomes unreachable.

When using UCP Disaster Recovery with SRM, SRA is integrated with SRM tobring volume information into SRM and enables it to communicate with thestorage system. This facilitates SRM in issuing CCI commands to test recoveryplans and enables SRM to execute a failover in the case of a disaster.





The following diagram shows how the components involved in volume recoveryare integrated with the components that are involved in volume replication.

Because UCP Disaster Recovery augments SRM instead of replacing it, a firmunderstanding of SRM is important to properly configure both. Configurationchoices in SRM, UCP Disaster Recovery, and UCP Director can affect thesettings that you need to specify in the others. For example:

• Creating SRM protection groups that contain the same hosts and volumes asthe resource groups that you create in UCP Disaster Recovery.





• According to VMware recommendations, VMs that support a singleworkload, such as a web server and a database server, should have theirvolumes protected together. VMware recommends placing their volumes inthe same SRM protection group, which means that they should also be in thesame replication group.

• Because SRM is able to perform a failover recovery of ESXi hosts, but nothosts running other operating systems, volumes attached to ESXi hostsshould not be added to the same replication group as non-ESXi hosts.

• Initiating a failover recovery of non-ESXi servers requires either a manualprocedure or a third-party solution.

• All volume pairs in a replication group are synchronized together. Individualpairs are not able to be synchronized apart from the replication group.

• An SRM protection group can be included in more than one protection plan.For example, one protection plan may power up a minimum number of VMs,while another protection plan may power up all VMs. It is important to haveenough hosts at the recovery site to support the plan with the most VMs in itto avoid a resource-intensive recovery plan from being executed in SRMwithout enough hosts at the recovery site to support it.

• After a failover recovery, the recovery site will become the primary site.The original primary site, however, will not automatically convert to beingthe new recovery site. Manual intervention may be necessary after afailover recovery is performed.

• There should be a corresponding SRM protection group for each UCP DOCreplication group.

Test volumes

If a replication group contains volumes that are only attached to ESXi hosts,then a test volume can be created for the replication group. When creating atest volume, you can create either a full test volume or a copy on write testvolume.

Synchronization happens at the replication group level. As a result, all testvolumes in a replication group must be of the same type. Because of this, whenadding new volume pairs to an established replication group, it is important tobe aware of what type of test volume the replication group is using because thenew volume pair will use the same type.

During replication, both test and recovery volumes are read-only. When therelationship between a test volume and the recovery volume is suspended, thetest volume will become read/write until the test is over. Afterward, when thepairing relationship is restored and the test volume is synchronized, it willbecome read only again. This helps ensure that the test volume is kept in syncwith the recovery and protected volumes.





Using SRM without a test volumeWhen storage system capacity needs to be conserved, and where it isacceptable to perform testing on the actual replicated volume, it is possible toconfigure a replicated volume to not have a test volume when configuring areplication group.

When a replication group is configured without a test volume, if a test isinitiated, then the relationship between the primary and recovery volume issuspended. SRM will then test the recovery plan against the recovery volume.After the test, you will need to manually resynchronize the replication group inUCP Disaster Recovery to re-establish the relationship between the primaryand recovery volumes and synchronize any changes that were made afterreplication was suspended.

Using SRM with a test volumeSometimes, it is not acceptable to suspend replication to a recovery volumewhen you want to test a recovery plan in SRM. When this is the case, you canuse a test volume.

Because the data in the test volume is a representation of the data in therecovery volume, it can be used for tests instead of the recovery volume. Thisenables tests to be performed at the same time that replication is beingperformed. To do this, UCP Disaster Recovery suspends the relationshipbetween the recovery volume and and the test volume to test a recovery plan.

Test volumes are defined when creating and configuring replication groups, andare attached to recovery hosts or clusters along with the recovery volumes asthey are defined. Because of this, recovery hosts and clusters will see twice thevolumes of a protected host.

There are two types of test volumes that you can use, a full test volume and acopy on write test volume.

Full test volume

A full test volume, also known as a ShadowImage, is an exact replica of therecovery volume and ensures full redundancy at all times. Because it is a fullreplica, however, the recovery site will need twice the total provisioned spaceof the primary site, not just the written blocks. As a result, there are threecopies of all volumes, providing greater redundancy.

When performing a test against a test volume, SRM will first suspendreplication between the recovery volume and the test volume. Because SRM istesting the recovery plan against the test volume, a successful test implies thatboth the recovery volume and the test volume would be able to function in thecase of a failover. Also, after the test is finished, the recovery and test volumewill need to be resynchronized.

Copy on write test volume

Where space should be conserved, but a test volume is required, a copy onwrite test volume can be used. Copy on write test volumes are Hitachi virtualvolumes, or v-vols, that are created in a Hitachi Thin Image (HTI) pool.





Copy on write test volumes offer the ability to test SRM without suspendingreplication between primary volume and recovery volume, while also using apercentage of the size of a full test volume. Instead of being an exact replica ofa recovery volume, a copy on write test volume is a virtual representation ofthe recovery volume that contains pointers to the actual data in the recoveryvolume. As a result, the host that is connected to both the recovery and testvolume sees both volumes as complete copies of the data, even though theactual data is only present in the recovery volume.

Because of this, while you can use a copy on write test volume to test an SRMrecovery plan, it is not able to provide the heightened level of data protectionthat a full test volume provides if something happens to the recovery volume.

UCP Disaster Recovery requires 20% of your HTI pool capacity to be emptywhen creating new copy on write test volumes to accommodate their variablecapacity. This preserves a buffer of space to accommodate the generation ofcopy on write test volumes for existing volume pairs.

When a copy on write test volume is first created, it is a clean map of all blocksin the volume that it represents and is shown in simplex state.

Changes that are made to the protected volume will be copied to both therecovery volume and the HTI pool that the copy on write test volume is locatedin, and pointers on the copy on write test volume are updated accordingly. Thecopy on write test volume will be in paired state when this is taking place.

When an SRM protection plan is tested, it will treat the copy on write testvolume as a normal read\write volume. When the relationship between therecovery volume and test volume is suspended, changes to the protectedvolume will continue to be written to the recovery volume and HTI pool, but thepointers in the copy on write volume are not updated.

When finished testing the SRM recovery plan, you will need to synchronize thereplication group in UCP Disaster Recovery to re-establish the relationshipbetween the primary, recovery, and test volumes and synchronize any changesthat were made after replication was suspended.

UCP Disaster Recovery administrationUCP Disaster Recovery is administered on the Disaster Recovery page in UCPDOC Console.

The Disaster Recovery page lists each site pair that has been added to UCPDisaster Recovery at the top of a page. When a site pair is selected, theDisaster Recovery page will list each replication group that is part of the sitepair as well as the volumes that have been added to that replication group.

To display the Disaster Recovery page, log into UCP DOC Console and click onthe Disaster Recovery link in the navigation bar at the bottom of the page.

• Administering site pairs on the next page

• Administering replication groups on the next page

• Administering volume replication on page 27



UCP Disaster Recovery administration


• Administering test volumes on page 27

Administering site pairs

In addition to displaying site pairs, the Disaster Recovery page is used to add,edit, and remove site pairs in UCP Disaster Recovery.

• Create a site pair on page 27

• Remove a site pair on page 31

• Refresh site pair inventory on page 30

Administering replication groups

Replication groups contain the volumes that are replicated between sites in asite pair. They are displayed within site pairs on the Disaster Recovery page.As a result, the corresponding site pair needs to exist before the replicationgroup can be created.

• Set up a replication group on page 32

• Pair and resynchronize a replication group on page 29

• Remove a replication group on page 30

Best practices for setting up replication groups

When setting up your replication groups, we recommend the following bestpractices:

• Include all data stores and volumes associated with a single workload intothe same replication group

• Optimize your disaster recover configuration to stay below 4000 replicatedvolumes

Although the UCP DOC software supports all maximum limits andconstraints that are driven by the underlying hardware, all softwareoperations such as, refresh, create, delete, etc., are optimized to support4000 replicated volumes.

Failover scenarios

In the event of a disaster or during routine maintenance, UCP provides theability to protect bare-metal/windows hypervisor workloads and allow endusers to failover a system to a remote site. Available failover scenarios arediscussed in this section.





Planned failover scenarioPerforming a failover on a replication group will allow the volumes at therecovery site to become writable. A planned failover will also switch thedirection of replication, so the volumes that were initially protected will be usedas the recovery volumes after the failover has completed.

A planned failover can be performed on a replication group once the data hasbeen successfully copied. A planned failover requires both UCP environments tobe functioning properly, and that all hosts associated with the replication groupbe powered off on the protected site. If the fiber link between sites is down orthe protected site is down, an unplanned failover is required.

For planned failover-related CLI references, see the UCP DOC CLI ReferenceGuide.

For planned failover-related API references, see the UCP DOC API ReferenceGuide.

Note: A planned failover can be performed on a bare-metal or Hyper-V replicationgroup. For managing failover of ESXi hosts, see "Using UCP Disaster Recovery withvCenter" on page 18.

Unplanned failover scenarioAn unplanned failover serves the same purpose as a planned failover, but canbe used in a scenario where the primary site is not functioning. During anunplanned failover, an attempt will be made to reverse the replication groupdirection. If this is not possible, it will make the recovery site’s volumeswritable but will not reverse the replication direction.

In this scenario, the status of the replication group will becomeSuspendForSwapping. In a SuspendedForSwapping status, the newly writablevolume will not accept changes from the protected site. This protects therecovery site from data corruption in the scenario where the protected sitecomes back online. For more information about this and other statuses, seeSynchronization status on page 18.

For unplanned failover-related CLI references, see the UCP DOC CLI ReferenceGuide.

For unplanned failover-related API references, see the UCP DOC API ReferenceGuide.

Note: An unplanned failover can be performed on a bare-metal or Hyper-Vreplication group. For managing failover of ESXi hosts, see "Using UCP DisasterRecovery with vCenter" on page 18.

Failover scenario where DOC is unavailableFor scenarios where DOC resides on the primary site and that site experiencesan outage, it is still possible to perform a failover on a replication group. Thiscommand does not perform the validation that is normally performed by DOC,and is only intended to be used in scenarios where DOC is unavailable.





Note: This command is subject to change in future releases.

Important: This command should not be used for SRM protected volumes.

API

To perform a failover from the API, a post request tohttps://ucpmanagement.ucp.local/replication/horcms/0/copygroups/P-Test/horctakeover can be performed.

Where the copygroup name is P-Test. The name of a copygroup can bedetermined by prepending a “P-“ to the replication group name that was used tocreate the replication group.

CLI

To perform this function from the command line, we must first ignore self-signed certs. This is accomplished with the following function:

function Ignore-SelfSignedCerts {add-type -TypeDefinition @"using System.Net;using System.Security.Cryptography.X509Certificates;public class TrustAllCertsPolicy : ICertificatePolicy {public bool CheckValidationResult(ServicePoint srvPoint, X509Certificate certificate,WebRequest request, int certificateProblem){ return true; }}"@[System.Net.ServicePointManager]::CertificatePolicy = New-ObjectTrustAllCertsPolicy}#function

Once the Ignore-SelfSignedCerts function has been run, an Invoke-RestMethodcommand can be used to trigger the failover.

ExampleInvoke-RestMethodhttps://ucpmanagement.ucp.local/replication/horcms/0/copygroups/P-Test/horctakeover -Credential $creds -Method Post –Body 30 –ContentTypeapplication/json

Where the copygroup name is P-Test, and $creds are the credentials to UCPDirector.





Administering volume replication

Replicated volumes are displayed and administered within replication groups onthe Disaster Recovery page.

• Set up volume replication on page 32

• Expand a volume on the next page

• Remove volume replication on page 30

Administering test volumes

All test volumes within a replication group must be of the same type. As aresult, test volumes are administered at the replication group level.

• Create a test volume on the next page

• Remove a test volume on page 31

How to...This section covers the following procedures.

• Create a site pair below

• Create a test volume on the next page

• Expand a volume on the next page

• Pair and resynchronize a replication group on page 29

• Refresh site pair inventory on page 30

• Remove volume replication on page 30

• Remove a replication group on page 30

• Remove a site pair on page 31

• Remove a test volume on page 31

• Set up a replication group on page 32

• Set up volume replication on page 32

Create a site pair

In addition to displaying site pairs, the Disaster Recovery page is used to add,edit, and remove site pairs in UCP Disaster Recovery.

When creating a site pair, UCP Disaster Recovery verifies the:

• Ethernet connectivity between each site.

• Replication path between the storage systems.



How to...


To create a site pair:

1. From the Disaster Recovery page, click on the Create Site Pair icon ( ).

2. On the Create Site Pair screen:

¡ In the Name field, type a name for the site pair.

¡ In the First Site field, select the first site of the site pair.

¡ In the Second Site field, select the second site of the site pair.

¡ From the Replication Technology section, select the type of replicationto perform, either Synchronous (TC) or Asynchronous (HUR).

¡ If Asynchronous (HUR) is selected, in the Maximum Outage Tolerancefield, enter the maximum outage tolerance.


Create a test volume

To add test volumes to a replication group that does not have any test volumes:

1. From the Disaster Recovery page, click on the site pair that contains thesites you want to create test volumes in.

2. Click on the site that contains the replication group you want to create testvolumes in.

3. Next to the replication group that you want to create test volumes in, clickon the Create Test Volume icon ( ).

4. In the Test Volume Pool section:

1. Select the test volume type to use, as follows: Full Copy or Copy onWrite.

2. Select the pool that the test volume will be created in:

– To have UCP Disaster Recovery automatically select the pool, clickon the Default option.

– To manually select the pool, click on the Existing option and selectthe pool in the Pool field.

5. Click on the Next button.

6. In the Summary section, review the protection configuration, and then clickon the Finish button.

Expand a volume

To expand a protected volume, UCP Disaster Recovery will:

1. Verify the replication path between the storage systems.



How to...


2. Suspend replication between all protected, recovery, and test volumes inthe replication group.

3. Expand the selected protected, recovery, and test volumes.

4. Resume replication between all protected and recovery volumes in thereplication group.

Note: When a volume is being expanded, it will not appear in the replication groupuntil the volume is expanded and replication is reestablished or recreated. If this isthe only volume in the replication group, the replication group will not appear untilthe volume is expanded and replication is recreated.

Note: During volume expansion, no change is made to the journals used forasynchronous replication in either storage system.

To expand a volume:

1. From the Disaster Recovery page, click on the replication group to expandthe volume in.

2. In the volumes table, click on the Expand Replicated Volume icon ( ) nextto the volume to expand.

3. Enter the new size of the volume in the New Volume Size field.


Pair and resynchronize a replication group

When pairing and resynchronizing a replication group, UCP Disaster Recoveryattempts to pair all of the replicated volume pairs in the replication group. Thisincludes:

1. Initiating replication between the protected and recovery volumes.

2. Initiating replication between the recovery and test volumes.

3. Creating the protected and recovery journals if using asynchronousreplication in a simplex state.

This can be used to recover a volume in simplex or suspended status after thecorresponding failure has been corrected.

To pair and resynchronize a replication group:

1. From the Disaster Recovery page, click on the site pair that contains thereplication group you want to pair and resynchronize.



How to...


2. Click on the Pair and Resynchronize Replication Group icon ( ) beside thereplication group that you want to pair and resynchronize.

3. Select the protected site from the Select Protected Site field if needed.


Refresh site pair inventory

To refresh site pair inventory:

1. In the Disaster Recovery section of the navigation bar, click on the RefreshSite Pair Inventory button.

2. When prompted, click on the Yes button.

Remove volume replication

When removing a volume, UCP Disaster Recovery:

1. Stops replication between the recovery volume and the test volume

2. Stops replication between the protected volume and the recovery volume.

The individual protected, recovery, and test volumes are not detached anddeleted. When removing a protected volume from a replication group thatcontains only one protected volume, UCP Disaster Recovery will also removethe replication group.

To remove replication from a volume:

1. From the Disaster Recovery page, click on the replication group to removevolume replication from.

2. In the volumes table, click on the Remove Replicated Volume icon ( )next to the volume to remove.


Remove a replication group

When removing a replication group, UCP Disaster Recovery:

1. Stops replication between the recovery and test volumes.

2. Stops replication between the protected and recovery volumes.

3. Removes the journals at the recovery and protected sites if using.asynchronous replication. While the journal will be removed, the journalvolume will not be automatically deleted.



How to...


Note: Removing a replication group will not detach and delete the test andrecovery volumes. To detach and delete the volumes, use UCP Director webinterface to access the appropriate site.

To remove a replication group:

1. From the Disaster Recovery page, click on the site pair that you want toremove the replication group from.

2. Click on the Remove Replication Group icon ( ).


When you remove the last replicated volume from a replication group, UCPDisaster Recovery will automatically remove the corresponding replicationgroup. For more information on removing a volume, see Administering volumereplication on page 27.

Remove a site pair

When you remove a site pair from UCP Disaster Recovery:

• All associated replication groups are removed.

• Replication between the two sites for all volumes is disabled.

• Individual volumes at each site are not removed.

To remove a site pair:

1. From the Disaster Recovery page, click on the Remove Site Pair icon ( ).


Remove a test volume

To remove test volumes from all protected volumes in a replication group:

1. From the Disaster Recovery page, click on the site pair that contains thesites you want to remove test volumes from.

2. Click on the site that contains the replication group you want to remove testvolumes from.

3. Next to the replication group that you want to remove test volumes from,click on the remove icon ( ).

4. Click on the Finish button.



How to...


Set up a replication group

Before you set up replication groups, we recommend that you first review Bestpractices for setting up replication groups on page 24.

When setting up a replication group, UCP Disaster Recovery:

1. Creates and adds replicated volume pairs to the replication group.

2. Creates journals at the protected and recovery sites if using asynchronousreplication.

3. Sets up replication for the replicated volume pair.

Replication groups are not created independently. Instead, when setting upvolume replication, you create a replication group by adding volume replicationto a new replication group. For more information on setting up volumereplication, follow the steps in Administering volume replication on page 27and, in the Configure Replication Group section, select the Create New option.

Set up volume replication

When setting up volume replication, UCP Disaster Recovery:

1. Verifies the replication path between the storage systems.

2. Creates the recovery volume or volumes and attaches them to the recoveryhost or cluster.

3. Creates the test volume or volumes on the storage system, if applicable.

4. For asynchronous replication:

¡ Creates new journals at the protected and recovery sites if creating anew replication group.

¡ Expands existing journals, if necessary, for an existing replicationgroup.

5. Initiates replication between the protected and recovery volumes.

6. Initiates replication between the recovery and test volumes.

7. Pairs and resynchronizes all simplex or suspended volumes in thereplication group that the volume is added to if adding volume replication toan existing replication group.

To create a new replication group or add volume replication to an existingreplication group, follow these steps:

1. On the Disaster Recovery page, click on the Configure as Protected icon () next to the site that you want to setup replication in.



How to...


2. In the Configure Replication Group section, select the replication groupthat you want to add the replicated volume too.

¡ To use an existing replication group, select the Add to Existing option,and then select the group from the Select from Existing field.

¡ To create a new replication group, select the Create New option, andthen type the name of the replication group in the Enter a Name field.

3. In the Protected Resource section:

1. In the Select Source field, select the host or cluster that is connected tothe volume you want to protect.

2. In the Select Volume field, select the volumes in the protected site thatyou want to protect.

4. In the Recovery Resource section:

¡ In the Select Target field, select the host or cluster that will beconnected to the recovery volume.

¡ Select the pool that the volume will be created in:


– To manually select the pool, click on the Existing option and selectthe pool in the Select from Existing field.


6. In the Test Volume Configuration section:

1. Select the test volume type to use, as follows: Full Copy, Copy onWrite, or None.

2. If creating a Full Copy or Copy on Write test volume, select the poolthat the test volume will be created in:


– To manually select the pool, click on the Existing option and selectthe pool in the Select from Existing field.


8. In the Summary section, review the protection configuration, and then clickon the Finish button.



How to...


Tracking UCP DOC jobs and eventsUCP DOC tracks all actions as jobs. This includes both actions that take placewithin UCP DOC as well as actions that UCP DOC triggers at a remote site. Italso tracks all events that are triggered by jobs or sites that affect UCP DOC orUCP Disaster Recovery.

This section describes how jobs and events are displayed in UCP DOC Console.

• Displaying UCP DOC jobs below

• Displaying UCP DOC events on the next page

• List of UCP DOC jobs on page 39

• List of UCP DOC events on page 41

Displaying UCP DOC jobsActions performed within UCP DOC are tracked as jobs. Jobs are used to trackboth actions that take are triggered by UCP DOC, like adding or removingusers, as well as actions that take place at remote sites as part of UCP DisasterRecovery functionality, such as resynchronizing storage systems. UCP DOCretains all jobs indefinitely.

The Jobs tab of the Jobs and Events page is used to display all jobs that aretracked by UCP DOC.

To display the Jobs tab of the Jobs and Events page, from the navigation bar:

1. Click on the title of the Jobs and Events section.

2. Click on the Jobs tab at the top of the page.

The Jobs tab contains two tables, as follows:

• The top table lists all of the jobs that are tracked by UCP DOC.

• The bottom table is the Related Events table. It is used to display all eventsthat are related to a selected job in the top table. If there are no relatedevents, then the Related Events table will be blank. For more informationon events, see List of UCP DOC events on page 41.

Chapter 4: Tracking UCP DOC jobs and events 35UCP DOC Administrator's Guide


4


For each job tracked by UCP DOC, the following properties are displayed:

• status — The status of the job. Job status is indicated as follows:¡

Success ( ) — The job completed successfully.

¡

Error ( ) — The job failed.

¡

Running ( ) — The job is currently running.

• Job ID— The UCP DOC ID of the job.

• Description — A plain-text description of the job.

• User — The AD domain and user ID of the user who initiated the job.

• Progress — The progress of the job is listed in this column, either Failed,Completed, or a percentage of the current progress.

• Start Date Time — The date and time the job started.

• End Date Time — The date and time the job finished.

• Target Type — The resource type the job was executed against.

• Target ID— The ID of the resource the job was executed against.

• Target Global ID— The fully qualified ID of the target. This includes the sitethat the target is associated with.

For a list of all jobs that can be generated by UCP DOC, see List ofUCP DOC jobs on page 39.

Displaying UCP DOC eventsEach job that is run can trigger one or more events, which are then trackedwithin UCP DOC. UCP DOC retains all events indefinitely.

The Events tab of the Jobs and Events page is used to display all events thathave been triggered by a job. Events related to a specific job are displayed onthe Related Events table on the Jobs tab when a job is selected.

To display the Events tab of the Jobs and Events page, from the navigationbar:

1. Click on the title of the Jobs and Events section.

2. Click on the Jobs tab at the top of the page.

For each event tracked by UCP DOC, the following properties are displayed:

• severity — The severity of the event. Event severity is indicated as follows:¡

Informational ( ) — Informational only. No action needs to be taken.

36 Chapter 4: Tracking UCP DOC jobs and eventsUCP DOC Administrator's Guide


Displaying UCP DOC events


¡

Warning ( ) — Abnormal but non-critical behavior has occurred.Actions may need to be taken to identify and correct the issue.

¡

Error ( ) — The target is not functioning correctly. Action likely needsto be taken to correct the issue.

• Event ID— The UCP DOC ID of the event.

• Description — A plain-text description of the event. Events that aretriggered by a job on a remote site are prefaced with:

Remote UCP job name:

• User — The AD domain and user ID of the user who initiated the triggeringjob.

• Date Time — The date and time the event was triggered.

• Target Type — The resource type the triggering job was executed against.

• Target ID— The ID of the resource the triggering job was executed against.

• Target Global ID— The fully qualified ID of the target the triggering jobwas executed against. This includes the site that the target is associatedwith.

For a list of all events that can be generated by UCP DOC, see List ofUCP DOC events on page 41

Chapter 4: Tracking UCP DOC jobs and events 37UCP DOC Administrator's Guide


Displaying UCP DOC events


38 Chapter 4: Tracking UCP DOC jobs and eventsUCP DOC Administrator's Guide



List of UCP DOC jobsThe following is a list of all jobs that are tracked by UCP DOC.

• Add replicated volume pair to replication group

• Add user to UCP Director Operations Center

• Create replication group for site pair

• Create site pair for UCP Director Operations Center

• Create test volume(s) for replication group

• Delete user from UCP Director Operations Center

• Deploying CentOS on all blades

• Deploy management virtual machine

• Expand replicated volume pair

• Register UCP Instance with UCP Director Operations Center

• Unregister UCP Instance from UCP Director Operations Center

• Update UCP Instance registered with UCP Director Operations Center

• Onboard UCP appliance

• Pair and resync a replication group

• Refresh site pair inventory

• Remove replicated volume pair from replication group

• Remove the replication group

• Remove site pair for UCP Director Operations Center

• Remove the test volume pair

• Update Active Directory configuration

• Update blade running configuration

• Update chassis available configuration

• Update chassis running configuration

39UCP DOC Administrator's Guide


A


• Update converged switch running configuration

• Update configuration of UCP Director Operations Center

• Update HUS storage available configuration

• Update HUSVM storage available configuration

• Update HUS VM storage running configuration

• Update management server network running configuration

• Update management server running configuration

• Update service VM network running configuration

• Update syslog available configuration

• Update user in UCP Director Operations Center

• Update management server switch port group running configuration

• Validate blade

• Validate chassis

• Validate HUS storage

• Validate HUS VM storage

• Validate management server

• Validate management server network

• Validate network Ethernet switch



List of UCP DOC jobs


List of UCP DOC eventsThe following table lists all events that are tracked by UCP DOC.

Event Severity

Unable to connect to the service on {url}. Make sure the service URL is correct, firewall andnetwork settings on this computer and on the server computer are configured properly, andthe appropriate services have been started on the server.

Error

Site pair: {sitePairId} for site pair name: {displayName} was created and added toinventory.

Info

Cannot access or read from database. Error

Destination host unreachable: {host}. Error

URL: {url} format is invalid. Error

Unable to connect to the inventory service at {url}. Error

Site pair: {sitePairId} for site pair: {displayName} was removed from inventory. Info

Replication group: {replicationGroupId} was not found for site pair: {sitePairId}. Error

Failed to reach UCP site: {ucpSiteId}. Error

Site pairs site: {siteIds} already exists in inventory. Error

Site pair: {sitePairId} was not found. Error

Failed to connect to UCP instance with IP Address: {ipAddress}. Error

UCP instance at URL {url} with serial number {serialNumber} is already registered andcannot be registered again.

Error

Cannot unregister UCP instance with IP Address: {ipAddress} as it is part of a DR site pair:{displayName}.

Error

Successfully registered UCP instance with URL: {url}. Info

Failed to register UCP instance with URL: {url}. Error

Successfully unregistered UCP instance with URL: {url}. Info

Failed to unregister UCP instance with URL: {url}. Error

Failed to update UCP instance with ID: {id}. Error

Ucp instace at URL {url} has serial number {newSerialNumber}. This is different that theregistered serial number {oldSerialNumber}. This update is not allowed.

Error

Successfully updated UCP instance {id} with URL: {url}. Info



B


Event Severity

UCP user: {username} is not a UCP administrator for UCP instance at: {url}. Error

UCP version {instanceVersion} for UCP instance at {url} is not supported. UCP DirectorOperations Center supports UCP version {minimumVersion} onwards.

Error

User {user} is not authorized to access orchestrator service at {url}. Error

Unexpected failure occurred while connecting to server at {url}. Status code:{statusCode}. Please contact your help desk or system administrator.

Error

UCP instance at URL {url} is associate with a site pair, so it cannot be unregistered. Error

Failed to pair and resync replication group: {replicationGroupId}. Error

Pair and resync of replication group: {replicationGroupId} was successful. Info

Primary site: {PrimarySiteId} was not found. Error

No path exists between sites: {siteIds}. Error

Successfully added user '{userName}'. Info

Failed to add user with name '{userName}'. Error

User with name '{userName}' already exists. Error

Failed to update user with name '{userName}'. Error

Failed to delete user with name '{userName}'. Error

Cannot delete all administrators. Error

Successfully updated user with name '{userName}'. Assigned to roles: '{roleList}'. Info

Successfully deleted user with name '{userName}'. Info

Refresh of site pair inventory was successful. Info

Test volumes pairs: {volumePairIds} do not belong to the same site and storage system. Error

The remote job: {jobId} for UCP at site: {siteId} has failed with the following errormessage: {eventMessage}.

Error

Remote resource not found. url: {url}, response: {responseBody}. Error

Replicated volume: {volumeId} was successfully added to replication group:{replicationGroupName}.

Info

Failed to reach UCP Site: {ucpSiteId} in site pair: {sitePairId}. Error

Remote UCP job name: '{jobName}' with id: {jobId} completed at site: {siteId}. Info

Remote UCP job name: '{jobName}' with id: {jobId} started at site: {siteId}. Info

Created replication group: {replicationGroupId}. Info

Replicated volume pair: {replicatedVolumePairId} was not found. Error

Replicated volume pair: {replicationVolumePairId} was successfully removed fromreplication group: {replicationGroupName}.

Info

Expand size: {size} successful for replicated volume pair id: {replicatedVolumePairId}. Info

Pair state: {state} invalid for expanding replicated volume pair id:{replicatedVolumePairId}.

Error



List of UCP DOC events


Event Severity

Expand size: {size} is invalid for replicated volume pair id: {replicatedVolumePairId}. Error

Remove replication group: {replicationGroupId} is successful. Info

Failed to connect Monitor State Queue at host {hostName}. Error

Successfully created journal: {journalId} for replication group: {replicationGroupId}. Info

Successfully added standard volume pairs for primary volumes: {primaryVolumeIds} intoreplication group: {replicationGroupId}.

Info

Successfully created test replication group for: {testCopyGroupId} for replication group:{replicationGroupId}.

Info

Successfully added test volume pairs for secondary volumes: {secondaryVolumeIds} intoreplication group: {replicationGroupId}.

Info

Successfully removed journals from replication group: {replicationGroupId}. Info

Successfully expanded journals for replication group: {replicationGroupId}. Info

Successfully refreshed inventory for site pair: {sitePairId}. Info

Successfully removed standard volume pair {standardvolumePairId} from replication group:{replicationGroupId}.

Info

Successfully removed test volume pair {testVolumePairId} from replication group:{replicationGroupId}.

Info

Successfully updated UCP Datacenter configuration. Info

Failed to send/receive test message to AMQP. Error

Failed to update UCP Datacenter configuration. Error

The remote UCP at site: {siteId} has encountered an error: {eventMessage} Error

An unexpected error occuredwhile trying to communicate with the remote UCP at site:{siteId}.

Error

The UCP Site: {ucpSiteId} is in use in site pair: {sitePairId}. Error

Site pair name: {sitePairName} is currently in use at site pair: {sitePairId}. Select a uniquesite pair name and try again.

Error

Could not find user/group '{userName}' in the directory. Error

Could not resolve remote Cci server hostname: {cciServername} from site: {siteId}. Error

The horcm instance: {horcmId} could not be found on site: {siteId}. Error

Unable to allocate journal volumes on site: {ucpInstanceId}. Error

Successfully removed primary configuration from replication group:{replicationGroupName} at sites: {siteIds}.

Info

Successfully removed test configuration from replication group: {replicationGroupName} atsite: {siteId}.

Info

Left journal not found for replication group: {replicationGroupId} on site pair: {sitePairId}.Please make sure that free journal volumes exist for your site.

Error



List of UCP DOC events


Event Severity

Right journal not found for replication group: {replicationGroupId} on site pair: {sitePairId}.Please make sure that free journal volumes exist for your site.

Error

Remote UCP job name: '{jobName}' with id: {jobId} failed at site: {siteId}. Warning

The requested primary site id {requestedPrimarySiteId} does not match the currentprimary site id {currentPrimarySiteId}.

Error

The display name '{displayName}' provided for the UCP instance: {ucpId} is already used byanother registered UCP instance. This operation is not allowed.

Error

The service URL '{serviceUrl}' provided for the UCP instance: {ucpId} is already used byanother registered UCP Instance. This operation is not allowed.

Error

The physical replication links from site: {sourceSiteId} to site: {targetSiteId} are in an errorstate. Verify that the physical replication links between the storage arrays at both sites arein a healthy state and try again.

Error

The replication link id: {pathId} on storage port: {portName} in path group:{pathGroupId} at site: {siteId} is in an error state. For optimal replication performance,verify that the physical replication links between the storage arrays at sites: {siteIds} are ina healthy state.

Warning

Replication group: {replicationGroupId} contains replicated volume pairs that are replicatedin oposite directions. This is not supported. All replicated volume pairs in the same replicationgroupmust be replicated in the same direction.

Error

At least one of the test replicated volume pairs in the r

ucp doc 4.1 administrator's guide...contents preface v intendedaudience v productversion v...

Documents