Professional Documents
Culture Documents
NetApp, Inc.
495 East Java Drive
Sunnyvale, CA 94089 U.S.A.
Telephone: +1 (408) 822-6000
Fax: +1 (408) 822-4501
Support telephone: +1 (888) 4-NETAPP
Documentation comments: doccomments@netapp.com
Information Web: http://www.netapp.com
Part number: 210-04801_A0
February 2010
Table of Contents | 3
Contents
Copyright information ................................................................................. 9
Trademark information ............................................................................. 11
About this guide .......................................................................................... 13
Audience .................................................................................................................... 13
Terminology .............................................................................................................. 13
Keyboard and formatting conventions ...................................................................... 14
Special messages ....................................................................................................... 15
How to send your comments ..................................................................................... 15
What is new in this release ......................................................................... 17
Overview of new and changed features ..................................................................... 17
User interface changes ............................................................................................... 18
New and changed CLI commands ............................................................................. 20
What Performance Advisor is ................................................................... 23
Topology of Performance Advisor ............................................................................ 23
Components of the Performance Advisor topology ...................................... 23
Limitations of Performance Advisor ............................................................. 24
Authentication of a Performance Advisor instance ....................................... 24
Performance Advisor support for hierarchical groups ............................................... 24
Performance Advisor support for secure connections ............................................... 25
Administrator roles in Performance Advisor ............................................................. 25
Performance Advisor support for vFiler units ........................................................... 25
About NetApp Management Console ....................................................... 27
What NetApp Management Console is ..................................................................... 27
Applications that run in NetApp Management Console ............................................ 27
Licenses ..................................................................................................................... 28
Window layout and navigation .................................................................................. 29
Window customization .............................................................................................. 30
Data filtering .............................................................................................................. 31
Performance objects and counters ............................................................ 33
What performance objects are ................................................................................... 33
List of performance objects ....................................................................................... 33
What a performance counter is .................................................................................. 34
4 | Performance Advisor Administration Guide For Use with DataFabric Manager Server 4.0
Copyright information
Copyright © 1994–2010 NetApp, Inc. All rights reserved. Printed in the U.S.A.
No part of this document covered by copyright may be reproduced in any form or by any means—
graphic, electronic, or mechanical, including photocopying, recording, taping, or storage in an
electronic retrieval system—without prior written permission of the copyright owner.
Software derived from copyrighted NetApp material is subject to the following license and
disclaimer:
THIS SOFTWARE IS PROVIDED BY NETAPP "AS IS" AND WITHOUT ANY EXPRESS OR
IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE,
WHICH ARE HEREBY DISCLAIMED. IN NO EVENT SHALL NETAPP BE LIABLE FOR ANY
DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE
GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER
IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR
OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF
ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
NetApp reserves the right to change any products described herein at any time, and without notice.
NetApp assumes no responsibility or liability arising from the use of products described herein,
except as expressly agreed to in writing by NetApp. The use or purchase of this product does not
convey a license under any patent rights, trademark rights, or any other intellectual property rights of
NetApp.
The product described in this manual may be protected by one or more U.S.A. patents, foreign
patents, or pending applications.
RESTRICTED RIGHTS LEGEND: Use, duplication, or disclosure by the government is subject to
restrictions as set forth in subparagraph (c)(1)(ii) of the Rights in Technical Data and Computer
Software clause at DFARS 252.277-7103 (October 1988) and FAR 52-227-19 (June 1987).
Trademark information | 11
Trademark information
NetApp; the NetApp logo; the Network Appliance logo; Cryptainer; Cryptoshred; DataFabric; Data
ONTAP; Decru; Decru DataFort; FAServer; FilerView; FlexCache; FlexClone; FlexShare; FlexVol;
FPolicy; gFiler; Go further, faster; Manage ONTAP; MultiStore; NearStore; NetCache; NOW
(NetApp on the Web); ONTAPI; RAID-DP; SANscreen; SecureShare; Simulate ONTAP;
SnapCopy; SnapDrive; SnapLock; SnapManager; SnapMirror; SnapMover; SnapRestore;
SnapValidator; SnapVault; Spinnaker Networks; Spinnaker Networks logo; SpinAccess;
SpinCluster; SpinFlex; SpinFS; SpinHA; SpinMove; SpinServer; SpinStor; StoreVault; SyncMirror;
Topio; vFiler; VFM; and WAFL are registered trademarks of NetApp, Inc. in the U.S.A. and/or other
countries. Network Appliance, Snapshot, and The evolution of storage are trademarks of NetApp,
Inc. in the U.S.A. and/or other countries and registered trademarks in some other countries. The
StoreVault logo, ApplianceWatch, ApplianceWatch PRO, ASUP, AutoSupport, ComplianceClock,
DataFort, Data Motion, FlexScale, FlexSuite, Lifetime Key Management, LockVault, NOW,
MetroCluster, OpenKey, ReplicatorX, SecureAdmin, Shadow Tape, SnapDirector, SnapFilter,
SnapMigrator, SnapSuite, Tech OnTap, Virtual File Manager, VPolicy, and Web Filer are
trademarks of NetApp, Inc. in the U.S.A. and other countries. Get Successful and Select are service
marks of NetApp, Inc. in the U.S.A.
IBM, the IBM logo, and ibm.com are trademarks or registered trademarks of International Business
Machines Corporation in the United States, other countries, or both. A complete and current list of
other IBM trademarks is available on the Web at http://www.ibm.com/legal/copytrade.shtml.
Apple is a registered trademark and QuickTime is a trademark of Apple, Inc. in the U.S.A. and/or
other countries. Microsoft is a registered trademark and Windows Media is a trademark of Microsoft
Corporation in the U.S.A. and/or other countries. RealAudio, RealNetworks, RealPlayer,
RealSystem, RealText, and RealVideo are registered trademarks and RealMedia, RealProxy, and
SureStream are trademarks of RealNetworks, Inc. in the U.S.A. and/or other countries.
All other brands or products are trademarks or registered trademarks of their respective holders and
should be treated as such.
NetApp, Inc. is a licensee of the CompactFlash and CF Logo trademarks.
NetApp, Inc. NetCache is certified RealSystem compatible.
About this guide | 13
Next topics
Audience on page 13
Terminology on page 13
Keyboard and formatting conventions on page 14
Special messages on page 15
How to send your comments on page 15
Audience
This document is written with certain assumptions about your technical knowledge and experience.
This document is for system administrators and others interested in managing and monitoring storage
systems with DataFabric Manager and Performance Advisor.
This document is written with the assumption that you are familiar with the following technology:
• Data ONTAP operating system software
• The protocols you use for file sharing or transfers, such as NFS, CIFS, iSCSI, FCP, or HTTP
• The client-side operating systems (Linux or Windows)
Terminology
To understand the concepts in this document, you might need to know how certain terms are used.
Storage terms
array LUN Refers to storage that third-party storage arrays provide to storage systems
running Data ONTAP software. One array LUN is the equivalent of one disk on
a native disk shelf.
14 | Performance Advisor Administration Guide For Use with DataFabric Manager Server 4.0
Keyboard conventions
Enter, enter • Used to refer to the key that generates a carriage return; the key is named
Return on some keyboards.
• Used to mean pressing one or more keys on the keyboard and then pressing the
Enter key, or clicking in a field in a graphical interface and then typing
information into the field.
hyphen (-) Used to separate individual keys. For example, Ctrl-D means holding down the
Ctrl key while pressing the D key.
Formatting conventions
Monospaced font • Command names, option names, keywords, and daemon names.
• Information displayed on the system console or other computer monitors.
• Contents of files.
• File, path, and directory names.
Bold monospaced Words or characters you type. What you type is always shown in lowercase
font letters, unless your program is case-sensitive and uppercase letters are
necessary for it to work properly.
Special messages
This document might contain the following types of messages to alert you to conditions that you
need to be aware of.
Note: A note contains important information that helps you install or operate the system
efficiently.
Attention: An attention notice contains instructions that you must follow to avoid a system crash,
loss of data, or damage to the equipment.
Next topics
Overview of new and changed features on page 17
User interface changes on page 18
New and changed CLI commands on page 20
New features
Data collection You can view the counters for which data collection is enabled or disabled on a
configuration per storage system basis. The following are the other advantages that are
available with this feature:
• Disabling or enabling data collection for selected counters per storage
system
• Configuring sample rate and retention period for selected counter groups per
storage system
• Copying the data collection configuration to other storage systems
Advanced views You can use advanced views to establish a baseline for the data collected by
applying statistical parameters such as minimum, maximum, mean, and
percentile. Time filters enable you to retrieve the baseline data based on selected
business hours and week days. You can use the baseline data in advanced views
to set appropriate thresholds.
Data export This feature helps you to export the Performance Advisor data collected on the
DataFabric Manager server. The data is collected in CMA format. However, you
should use this feature only when NetApp requests the data from you.
Performance You can diagnose the storage system for any performance-related issue during a
diagnosis given period and get recommendations to resolve the issue. The event
correlation chart allows you to plot events of related objects. Also, the chart
plots the Operations Manager events, which can help you in diagnosing the
cause of the performance degradation.
18 | Performance Advisor Administration Guide For Use with DataFabric Manager Server 4.0
Client statistics You can collect the statistics related to CIFS and NFS client operations to find
the cause of latency. Also, you can use this feature to automatically collect per-
client statistics when a set of thresholds on a storage system is breached.
Changes
Flexible You can apply custom views on either a particular object or every instance of the
custom selected object. The following are the other advantages that are available with this
views feature:
• Creating copies of canned views and modifying them
• Displaying data for historical periods in custom views
• Comparing the data collected for a counter on multiple instances of an object
• Configuring bar chart preferences such as the number of objects and the kind of
chart
Command button The following user actions are consolidated in the Actions list:
changes in charts
• Add counters
What is new in this release | 19
• Show events
• Reset both axes
• Add thresholds
• Save the chart
• Print the chart
Improvements in • A new command button, Data Collection, is added to the Set Up Hosts
the Set Up Hosts window to configure data collection on storage systems. Starting from
window this release, data collection is disassociated from views. Hence, you can
perform tasks such as setting sample rates and retention periods by using
this command button. You can also edit and copy data collection
configuration using this button.
• The following new columns are added to the hosts list:
• Max Projected Space
• Used Space
• Percentage of Used Space
• A new tab, Data Collection, is added to the bottom-half of the Set Up
Hosts window. You can click this tab to display the maximum projected
space, used space, sample rate, and retention period of a counter group
for a selected host. Also, the window displays the data collection enabled
for counters belonging to a selected counter group.
Improvements in • Starting from this release, data collection is not part of custom views.
the Set Up Custom You cannot set sample rates and retention periods using this window.
Views window
• The following new command buttons are added:
• Copy
• Refresh
• The following new columns are added to the Views list:
• Applies to
• View Type
• You can apply custom views to one or more particular objects or for all
instances of objects by selected type.
• The details area displays the counter, the counter group, and the kind of
chart associated with the selected view.
20 | Performance Advisor Administration Guide For Use with DataFabric Manager Server 4.0
Next topics
Topology of Performance Advisor on page 23
Performance Advisor support for hierarchical groups on page 24
Performance Advisor support for secure connections on page 25
Administrator roles in Performance Advisor on page 25
Performance Advisor support for vFiler units on page 25
Next topics
Components of the Performance Advisor topology on page 23
Limitations of Performance Advisor on page 24
Authentication of a Performance Advisor instance on page 24
• Performance Advisor reads the performance data stored on the monitoring server and generates
performance-related views and charts. You must install Performance Advisor on a Windows or
Linux system, separate from the system on which the DataFabric Manager server is installed.
Note: Performance Advisor is available only on the host on which NetApp Management Console
is installed. You cannot view Performance Advisor remotely through a browser.
• The DataFabric Manager server monitors one or more storage systems or vFiler units.
The vFiler counters pertaining to its hosting storage systems are displayed for hosting storage
systems running Data ONTAP 7.1 or later.
Note: The vFilers, All report at Control Center > Home > Member Details > Virtual Systems
in Operations Manager lists your network’s vFiler units and their hosting storage systems.
See the Operations Manager Administration Guide for information about DataFabric Manager
support for the MultiStore feature.
Related information
Operations Manager Administration Guide - now.netapp.com/NOW/knowledge/docs/DFM_win/
dfm_index.shtml
About NetApp Management Console | 27
Next topics
What NetApp Management Console is on page 27
Applications that run in NetApp Management Console on page 27
NetApp Management Console Licenses on page 28
NetApp Management Console window layout and navigation on page 29
NetApp Management Console window customization on page 30
NetApp Management Console data filtering on page 31
Related information
Operations Manager Administration Guide - http://now.netapp.com/NOW/knowledge/docs/
DFM_win/dfm_index.shtml
Performance This application provides a single location from which you can view
Advisor comprehensive information about storage system and MultiStore vFiler unit
28 | Performance Advisor Administration Guide For Use with DataFabric Manager Server 4.0
performance and perform short-trend analysis. The application also helps you
identify the data infrastructure causes and potential causes of reduced
performance.
Performance Advisor is automatically enabled with the Operations Manager Core
license.
Protection This application provides a policy-based management tool to help you unify and
Manager automate backup and mirroring operations. The application uses a holistic
approach to data protection. It provides end-to-end, workflow-based design and
seamless integration of SnapVault, SnapMirror, and Open Systems SnapVault to
enable you to manage large-scale deployments easily.
The disaster recovery feature of the protection application enhances your data
protection services by enabling you to continue to provide data access to your
users, even in the event of mishap or disaster that disables or destroys the storage
systems in your primary data node. If the disaster recovery license is installed, you
can quickly enable your secondary storage systems to provide primary data
storage access to your users with little or no interruption, until your primary
storage systems are reenabled or replaced.
Provisioning This application helps you simplify and automate the tasks of provisioning and
Manager managing storage. The application provides policy-based provisioning and
conformance of storage in datasets. The application also enables you to manually
add volumes or qtrees to a dataset at any time, provides manual controls for space
and capacity management of existing storage and newly provisioned storage, and
allows you to migrate datasets and vFiler units to a new storage destination.
The deduplication feature of the provisioning application enhances your data
provisioning services by enabling you to eliminate duplicate data blocks to reduce
the amount of storage space used to store active data.
To enable the protection functions, you must install the Protection Manager license on the
DataFabric Manager server. The Protection Manager license also enables all the provisioning
functions provided by Provisioning Manager.
• Disaster recovery functions
To enable the disaster recovery functions, you must install the Protection Manager license and the
disaster recovery license.
• Provisioning functions
Provisioning functions are automatically enabled if you have the Protection Manager license
installed. If that license is not installed, you must install the Provisioning Manager license on the
DataFabric Manager server. This license enables only the provisioning features.
• Deduplication functions
To enable the deduplication functions, you must install the Provisioning Manager license and the
deduplication license.
Learning about The Get Started section of the user interface provides an overview of
Protection Protection Manager, a short tutorial that demonstrates how to set up
Manager protection, and a list of frequently asked questions (FAQs).
Learning about The Get Started section of the user interface provides an overview of disaster
disaster recovery recovery, a short tutorial that demonstrates how to set up disaster recovery,
and a list of frequently asked questions (FAQs).
Learning about The Get Started section of the user interface provides an overview of
Provisioning Provisioning Manager, a short tutorial that demonstrates how to set up
Manager provisioning, and a list of frequently asked questions (FAQs).
Moving back and The Back and Forward arrow buttons toggle between your last and previous
forth between panes, even if the pane was in a different application.
panes
Toggling between • To use Protection Manager or Provisioning Manager, select Tasks >
applications
Manage Data or go to the Tasks Bar and click .
• To use Performance Advisor, select Tasks > Manage Performance or
go to the Tasks Bar and click .
Displaying help • To display the Help for all applications on NetApp Management Console,
click Help > Help Contents.
30 | Performance Advisor Administration Guide For Use with DataFabric Manager Server 4.0
• To display Help for the specific window that is currently displayed, click
Help > Help For This View or click .
Logging out • To log out, select File > Log Out or click the logout button.
• To log out and close NetApp Management Console, select File > Exit.
Viewing product To view information about the product licenses you have installed, click
license information Help > About > Licenses. Removed or added licenses are reflected in the
license list after you restart NetApp Management Console.
Applications that run in NetApp Management Console vary in the specifics of their window layout.
However, the windows are generally divided into two panes:
• Use the navigation pane on the left to move within an application.
• Use the content pane on the right to view and manage data. The content pane is generally divided
into two areas: lists of selectable data and details of the selected data. Click , when it is
displayed, for additional details.
Note: Specific applications might have additional navigation features not described in this section.
In addition, software version incompatibility may result in some reduced functionality, which may
cause some menu options or buttons to be disabled.
Hiding or You can hide the Tasks Bar to increase the space available for the content
redisplaying the pane. (The overall width of the window remains the same when the bar is
Tasks Bar hidden.) You use the View menu to hide or display the bar.
Selecting columns In content panes that include a list divided into columns, you can choose
to display which columns you want to display. To display or hide a column, click the
column-selection icon in the upper-right corner above the scroll bar ( ),
then click the name of a column heading from the selection list. Columns
already displayed are identified with a check mark.
Note: The default version of a list does not necessarily include all
available columns. You might find additional columns available in the
column selection list.
About NetApp Management Console | 31
Readjusting In content panes that include a list divided into columns, you can adjust the
column widths width of individual columns by horizontally dragging the vertical line
between column headings. If accommodating all the columns you want to
include results in column widths that obscure data, you can still see the
entire value of a column field by putting your cursor over the field. A tool tip
pops up, displaying the entire value of the field. You can maximize data
visibility across all columns by clicking the column selection icon ( ) and
choosing Pack All Columns.
Rearranging In content panes that include a list divided into columns, you can rearrange
column order the column order for a session by dragging and dropping column headings
horizontally within the table. The arrangement persists until you exit the
console.
Reapportioning a A splitter bar divides a content pane list area from the details area. You can
content pane move the splitter bar to increase or decrease the proportion of the content
pane that displays the list of selectable data.
Filtering by The Group selection list in the toolbar enables you to display only the data that
group pertains to objects in a selected group. This setting persists until you log out or
choose a different group.
Filtering by You can filter columns displaying site-specific values, such as storage system
regular names or dataset names, by regular expression. To filter a column by regular
expression expression, click the filter icon in the column heading ( ) and specify the
regular expression to match against values in the column field. Headings of
filtered columns are highlighted to remind you that some data is not currently
displayed.
NetApp Management Console uses Java regular expression syntax for filtering
data. For example:
• To view only items beginning with the letters "sch," type sch in the filter
field, which matches Schedule but not SnapMirror.
32 | Performance Advisor Administration Guide For Use with DataFabric Manager Server 4.0
Filtering by When a column displays a set of predefined possible values, you can choose to
column values display only the rows of data that include a selected value. To filter a column by a
specific, predefined value, click the filter icon in the column heading ( ) and
select one or more of the predefined values from the drop-down list. Headings of
filtered columns are highlighted to remind you that some data is not currently
displayed. Click "All" to select or clear all the values.
Filtering by
You can click in the upper-right corner of the list to select which columns you
column view
want displayed.
Sorting by You can click on the column header to change the sort order of the column
column values entries. When you click the column header, the sort arrow appears for that
column.
Performance objects and counters | 33
Next topics
What performance objects are on page 33
List of performance objects on page 33
What a performance counter is on page 34
List of performance counters on page 35
vFiler (applies to vFiler units and their Statistics that apply to vFiler units and their hosting storage system
hosting storage system)
For example, every storage system has an http_ops counter. However, if the storage system does
not have HTTP enabled, the counter contains only zero-value data. If SnapMirror is not licensed
on the storage system, the counters for the SnapMirror statistics contain no data.
Aggregate object
total_transfers Total number of disk operations serviced by the aggregate per second
user_reads Number of user read operations serviced by the aggregate per second
user_writes Number of user write operations serviced by the aggregate per second
cp_reads Number of read operations to the aggregate per second during consistency
point processing
user_read_blocks Number of blocks transferred for user read operations per second
user_write_blocks Number of blocks transferred for user write operations per second
cp_read_blocks Number of blocks transferred for consistency-point read operations per second
cifs object
cifs_ops Total number of CIFS operations
cifs_op_count Array of select CIFS operation counts
cifs_ops_pct Array of select CIFS operation counts as a percentage of total CIFS operations
cifs_latency Average latency for CIFS operations in milliseconds
cifs_latency_base Total observed CIFS operations to be used as a base counter for CIFS average
latency calculation
cifs_read_ops Total number of CIFS read operations per second
cifs_write_ops Total number of CIFS write operations per second
disk object
total_transfers Total number of disk operations serviced by the aggregate per second
36 | Performance Advisor Administration Guide For Use with DataFabric Manager Server 4.0
user_reads Number of user read operations serviced by the aggregate per second
user_writes Number of user write operations serviced by the aggregate per second
cp_reads Number of read operations to the aggregate per second during
consistency point processing
guaranteed_reads Number of disk read operations initiated each second for RAID
reconstruct or scrubbing activities
Note: This counter is available on storage systems running Data
ONTAP versions earlier than 7.1.
guarenteed_reads Number of disk read operations initiated each second for RAID
reconstruct or scrubbing activities
Note: This counter is available on storage systems running Data
ONTAP 7.1 or later.
fcp object
fcp_ops FCP operations per second
fcp_write_data FCP bytes written per second
fcp_read_data FCP bytes read per second
fcp_read_latency Average latency of read operations observed over all LUNs in the system
accessed over FCP in milliseconds
fcp_read_ops Total number of read operations observed over all LUNs in the system
accessed over FCP
fcp_write_latency Average latency of write operations observed over all LUNs in the system
accessed over FCP in milliseconds
fcp_write_ops Total number of write operations observed over all LUNs in the system
accessed over FCP
fcp_latency Average latency for FCP operations in milliseconds
38 | Performance Advisor Administration Guide For Use with DataFabric Manager Server 4.0
ifnet object
recv_packets Packets received per second
recv_errors Errors per second while receiving packets
send_packets Packets sent per second
send_errors Errors per second while sending packets
collisions Collisions per second on CSMA interfaces
recv_data Total bytes received per second
send_data Total bytes sent per second
recv_mcasts Packets received per second via multicast
send_mcasts Packets sent per second via multicast
recv_drop_packets Receive packets dropped per second
iSCSI object
iscsi_ops iSCSI operations per second
iscsi_write_data iSCSI bytes written per second
iscsi_read_data iSCSI bytes read per second
iscsi_read_latency Average latency of read operations observed over all LUNs in the system
accessed over iSCSI in milliseconds
iscsi_read_ops Total number of read operations observed over all LUNs in the system
accessed over iSCSI
iscsi_write_latency Average latency of write operations observed over all LUNs in the system
accessed over iSCSI in milliseconds
iscsi_write_ops Total number of write operations observed over all LUNs in the system
accessed over iSCSI
iscsi_latency Average latency for ISCSI operations in milliseconds
LUN object
read_ops Read operations per second
write_ops Write operations per second
other_ops Other operations per second
read_data Read bytes per second
write_data Write bytes per second
Performance objects and counters | 39
nfsv3 object
nfsv3_ops Total number of NFSv3 operations
nfsv3_op_count Array of select NFSv3 operation counts
nfsv3_op_latency_base Array of select NFSv3 operation counts for latency calculation
nfsv3_op_percent Array of select NFSv3 operations as a percentage of total NFSv3
operations
nfsv3_op_latency Array of latencies of select NFSv3 operations
nfsv3_avg_op_latency Average latency of the NFSv3 operations
nfsv3_read_size_histo Histogram of NFSv3 read sizes
nfsv3_write_size_histo Histogram of NFSv3 write sizes
nfsv3_read_latency Average latency for NFSv3 read operations in milliseconds
nfsv3_read_ops Total observed NFSv3 read operations
nfsv3_write_latency Average latency for NFSv3 write operations in milliseconds
nfsv3_write_ops Total observed NFSv3 write operations
nfsv4 object
nfsv4_ops Number of total NFSv4 operations per second
nfsv4_op_latency Array of latencies of select NFSv4 operations
nfsv4_read_ops Number of NFSv4 read operations per second
nfsv4_avg_latency Average latency of an NFSv4 request
nfsv4_write_ops Number of NFSv4 write operations per second
40 | Performance Advisor Administration Guide For Use with DataFabric Manager Server 4.0
Processor object
sk_switches Number of sk switches
hard_switches Number of context switches
domain_busy Array of processor time in percentage spent in various domains
processor_busy Percentage of elapsed time that the processor is executing non-idle
processes
processor_elapsed_time Wall-clock time since boot, used for calculating processor utilization
qtree object
Note: To minimize negative performance impact on storage systems that it monitors, Performance
Advisor does not automatically generate default performance views for its qtree performance
objects; however, you can use the following counters to define performance views for individual
qtrees.
System object
nfsv3_ops NFS operations per second
nfsv4_ops NFS operations per second
cifs_ops CIFS operations per second
http_ops HTTP operations per second
dafs_ops DAFS operations per second
fcp_ops FCP operations per second
iscsi_ops iSCSI operations per second
net_data_recv Network KB received per second
net_data_sent Network KB sent per second
disk_data_read Disk KB read per second
disk_data_written Disk KB written per second
cpu_busy Percentage of time one or more processors is busy in the system
Note: For storage systems running Data ONTAP 7.2 or earlier, the
cpu_busy counter is the amount of time that any one CPU is busy. This
results in a value for cpu_busy that is inflated. For storage systems
Performance objects and counters | 41
running Data ONTAP 7.2.1 or later, the cpu_busy counter is the greater
of either average CPU utilization or the busiest domain.
Target object
cifs_ops CIFS operations per second
read_ops Read operations per second
write_ops Write operations per second
other_ops Other operations per second
read_data Read bytes per second
write_data Write bytes per second
queue_full SCSI queue full responses per second
protocol_errors FCP protocol errors per second
invalid_xmit_words Invalid words transmitted per second
crc_errors CRC errors per second
lr_sent LRs sent per second
lr_received LRs received per second
discarded_frames Number of frames discarded per second
nos_received NOSs received per second
ols_received OLSs received per second
queue_depth_hba queue depth for the HBA
avg_latency_hba Average latency in milliseconds for all operations on the HBA
42 | Performance Advisor Administration Guide For Use with DataFabric Manager Server 4.0
vFiler object
vfiler_cpu_busy Percentage of CPU capacity utilized per second by the vFiler unit’s hosting
storage system
vfiler_net_data_recv Network data received in kilobytes per second
vfiler_net_data_sent Network data sent in kilobytes per second
vfiler_read_ops Read operations per second on the vFiler unit
vfiler_write_ops Write operations per second on the vFiler unit
vfiler_misc_ops Miscellaneous operations per second on the vFiler unit
vfiler_read_bytes Data read in kilobytes per second on the vFiler unit
vfiler_write_bytes Data written in kilobytes per second on the vFiler unit
Volume object
total_ops Number of operations per second serviced by the volume
avg_latency Average latency in milliseconds for all operations on the volume
read_ops Number of reads per second to the volume
read_data Bytes read per second from the volume
read_latency Average time for reads to the volume
write_ops Number of writes per second to the volume
write_data Bytes written per second (b_per_sec) to the volume
write_latency Average time for writes to the volume
other_ops Number of other operations per second to the volume
other_latency Average time for other operations to the volume
internal_msgs Number of internal backdoor messages per second to the volume
read_blocks Number of blocks read per second from the volume
write_blocks Number of blocks written per second to the volume
nfs_read_ops Number of NFS reads per second to the volume
Performance objects and counters | 43
nfs_read_data Bytes read per second via NFS from the volume
nfs_read_latency Average time for NFS reads to the volume
nfs_write_ops Number of NFS writes per second to the volume
nfs_write_data Bytes written per second via NFS to the volume
nfs_write_latency Average time for NFS writes to the volume
nfs_other_ops Number of other NFS operations per second to the volume
nfs_other_latency Average time for other NFS operations to the volume
cifs_read_ops Number of CIFS reads per second to the volume
cifs_read_data Bytes read per second via CIFS from the volume
cifs_read_latency Average time for CIFS reads to the volume
cifs_write_ops Number of CIFS writes per second to the volume
cifs_write_data Bytes written per second via CIFS to the volume
cifs_write_latency Average time for CIFS writes to the volume
cifs_other_ops Number of other CIFS operations per second to the volume
cifs_other_latency Average time for other CIFS operations to the volume
san_read_ops Number of block protocol reads per second to the volume
san_read_data Bytes read per second via block protocol from the volume
san_read_latency Average time for block protocol reads to the volume
san_write_ops Number of block protocol writes per second to the volume
san_write_data Bytes written per second via block protocol to the volume
san_write_latency Average time for block protocol writes to the volume
san_other_ops Number of other block protocol operations per second to the volume
san_other_latency Average time for other block protocol operations to the volume
What access permission is | 45
Next topics
RBAC for configuration of views on page 45
RBAC for changing storage system settings on page 46
RBAC for thresholds on page 46
RBAC for threshold templates on page 47
RBAC for configuration of events on page 48
RBAC for configuration of alarms on page 48
See the performance data of a Database Read Managed object associated with the
counter in a view counter instance in the view
Modify a custom view: Remove a Database Read Managed object associated with the
counter counter instance in the view
Note: To view real-time counter data in a view, you require the Database Read capability.
46 | Performance Advisor Administration Guide For Use with DataFabric Manager Server 4.0
• Change credentials
• Transport protocol of a storage
system
Create or edit thresholds on the Database Write On each instance of the object
counters of the following objects:
• vFiler unit
• Aggregate
• Volume
• Qtree
• LUN
• Disk
• Network interface
What access permission is | 47
View thresholds on the counters of Database Read On each instance of the object
the following objects:
• vFiler unit
• Aggregate
• Volume
• Qtree
• LUN
• Disk
• Network interface
• System
• NFSv3
• CIFS
• FCP
• iSCSI
• System
• NFSv3
• CIFS
• FCP
• iSCSI
Listing all alarms Alarm Read Group associated with the alarm
What access permission is | 49
Performance Advisor and server setup | 51
Next topics
Enabling the performance-monitoring server on page 51
Secure connections for Performance Advisor on page 52
Steps
1. In Operations Manager, select Control Center > Setup > Options > Performance Advisor to
display the Performance Advisor options.
2. If the application is disabled, in Performance Advisor Options, select Performance Advisor
Enabled > Enabled.
By default, the DataFabric Manager server enables the performance-monitoring server.
3. Click Update.
52 | Performance Advisor Administration Guide For Use with DataFabric Manager Server 4.0
Next topics
Setting up transport between the performance-monitoring server and Performance
Advisor on page 52
Setting up transport protocols between Performance Advisor and monitored devices on page 53
Setting up appliance-level transport between Performance Advisor and monitored
devices on page 53
List of options for appliance-level setting on page 54
Steps
1. In Operations Manager, select Control Center > Setup > Options > Performance Advisor to
display the Performance Advisor options.
2. Select among the following options to accept or modify the settings as appropriate to your
environment:
• Server HTTP Port—Sets the port used by the performance-monitoring server to receive HTTP
connections from the application. The default value is 8088.
• Server HTTP Enabled—Enables or disables HTTP as a transport protocol used by the
application. The default value is Yes.
• Server HTTPS Port—Sets the port used by the performance-monitoring server to receive
HTTPS connections from the application. The default value is 8488.
• Server HTTPS Enabled—Enables or disables HTTP as a transport protocol used by the
application. The default value is Yes.
3. Click Update.
Performance Advisor and server setup | 53
Steps
1. In Operations Manager, select Control Center > Setup > Options > Performance Advisor to
display the Performance Advisor options.
2. Select among the following Performance Advisor Transport options to accept or modify the
settings as appropriate to your environment:
Options Description
httpOnly Specifies HTTP only as the global transport protocol between the performance-monitoring
server and all monitored devices. This is the default setting
httpsOk Enables both HTTP and HTTPS as global transport protocols between the performance-
monitoring server and all monitored devices.
Note: When the transport protocol for the host is HTTPS, you must not set Performance
Advisor Transport to HTTP.
3. Click Update.
Steps
4. Click Update.
Next topics
Starting Performance Advisor on page 55
Stopping Performance Advisor on page 56
Unavailable storage systems or vFiler units in Performance Advisor on page 56
Unavailable menu options in the Performance Advisor interface on page 56
Zero-value data in a Performance Advisor chart on page 56
Absence of data in a Performance Advisor chart on page 57
Display of new storage systems or vFiler units in Performance Advisor on page 57
Steps
1. Click Start > All Programs > NetApp > Management Console.
Alternatively, start NetApp Management Console on Windows by clicking the NetApp
Management Console icon on the desktop.
2. Enter the fully qualified domain name, user name, and the password for the DataFabric Manager
server you want to log in to.
The login screen remembers the user name, domain name, transport, and port number from your
most recent successful login. If you are logging in as the same user as before, you need to enter
only your password.
3. Select HTTP or HTTPS transport.
The default port is selected automatically. If you log in using any other port, you must set the
serverHTTPSPort option to the matching port.
4. Click Connect.
5. Click the Manage Performance icon to display the top-level groups configured for the DataFabric
Manager server that you have logged in to.
56 | Performance Advisor Administration Guide For Use with DataFabric Manager Server 4.0
Step
Performance The Monitor dashboard in Performance Advisor displays information about the
dashboard current overall performance status of your systems, such as top performance
events, top storage systems by the total number of operations, top storage systems
by network throughput, and top storage systems by CPU utilization. You can
modify the default settings to configure any view on the global group as a
dashboard.
To access an application dashboard, you must have appropriate privileges. Items that you do not have
privileges for are not displayed in the dashboard. If you encounter access problems, contact the
administrator who maintains your DataFabric Manager roles and privileges.
Next topics
Dashboard panel descriptions on page 59
What events are on page 60
How to know when an event occurs on page 60
You can view more details about the Performance Advisor dashboard by clicking in the
dashboard panel title bar.
Top Performance Displays the five events with the highest severity levels. Items are ordered
Events first by severity, then by time of the events. More detail about each event is
provided in Monitor > Events.
Top Storage Displays a bar chart of the top five DataFabric Manager storage systems,
Systems by sorted by the highest network throughput. The number above each bar chart
Network displays the exact value of the throughput per second from that system. The
Throughput vertical axis displays the megabytes of throughput per second. The
horizontal axis displays the name of the storage system.
60 | Performance Advisor Administration Guide For Use with DataFabric Manager Server 4.0
Top Storage Displays a bar chart of the top five DataFabric Manager storage systems,
Systems by Total sorted by the highest total operations. The vertical axis displays the total
OPs operations per second for that storage system. The horizontal axis displays
the storage system name.
Top Storage Displays a bar chart of the top five DataFabric Manager storage systems,
Systems by CPU sorted by the highest average CPU utilization. The number above each bar
Utilization chart displays the exact value of the throughput per second from that system.
The vertical axis displays the percentage CPU usage of the storage system.
The horizontal axis displays the name of the storage system.
Next topics
Monitoring events on page 61
Responding to and acknowledging events on page 61
Deleting events on page 62
Overview of the NetApp Management Console dashboards | 61
Monitoring events
You can view a list of all events that occurred and view detailed information about any event. You
can also view deleted events that are marked for deletion.
Steps
1. For licensed protection and provisioning applications, from the navigation pane,
click Notifications > Events. For Performance Advisor, from the navigation pane, click Monitor
> Events.
2. (Optional) You can customize the Events window in any of the following ways:
• Select an event to view details about that event.
• Click in a column header to control which event entries you want displayed.
• Click the sort arrows in a column header to change the sort order of the entries in that column.
•
Click in the upper-right corner of the list to select which columns are displayed.
• Drag the bottom of the events list area up or down to resize that area.
3. (Optional) You can view events that are marked for deletion by displaying hidden columns and
removing a text filter, as follows:
a.
Click in the upper-right corner of the list and select the Deleted By and Deleted Time
columns.
b. Click in the Deleted By column header and delete the default search string.
The default filtering of deleted events is stopped and the deleted event entries are displayed in
the list.
Steps
1. For licensed applications, from the navigation pane, click Notifications > Events. For
Performance Advisor, from the navigation pane, click Monitor > Events.
2. In the Events window, select an event to view the details about it.
3. Select an event and click Acknowledge to indicate that you are taking responsibility for
managing that event.
4. In the Acknowledge Events dialog box, click Yes to acknowledge the selected event.
Your user name and the time are entered in the Events list for the selected event.
5. Find the cause of the event and take corrective action.
Deleting events
You can delete an event from the event list. Typically, you delete an event only after you
acknowledge it and resolve the problem. However, you can also delete unacknowledged events if
they are not important.
Steps
1. For licensed applications, from the navigation pane, click Notifications > Events. For
Performance Advisor, from the navigation pane, click Monitor > Events.
2. In the Events window, select an event and click Delete.
3. In the Delete Events dialog box, click Yes to delete the selected event.
4. Verify that the event is deleted by viewing the Events window.
You can view events that are marked for deletion by clicking in the corner of the list.
Views in Performance Advisor | 63
Next topics
What the performance view window is on page 63
Performance-monitoring with custom views on page 73
Analysis of real-time data on page 77
Canned views at various levels on page 79
Next topics
What default views are on page 64
What canned views are on page 64
What custom views are on page 64
What advanced views are on page 65
What real-time data is on page 66
What historical data is on page 66
Hierarchy of objects on page 66
Analyzing historical data on page 66
What synchronizing time axes is on page 67
Viewing performance data on page 67
Displaying tabbed views on page 67
Selecting a performance view on page 67
Closing a performance view on page 68
Data in performance views on page 68
Changing the time period in a chart on page 69
Resetting both axes on page 69
Synchronizing time axes on page 69
Undoing synchronization of time axes on page 70
How to use the legend to interpret performance data on page 70
How to view thresholds on page 70
Creating advanced views on page 70
64 | Performance Advisor Administration Guide For Use with DataFabric Manager Server 4.0
Because custom views offer so much flexibility, it is advisable to consider the following objectives
before creating a custom view:
Views in Performance Advisor | 65
• Access real-time data, other than what appears in the default view
• Group counters into combinations that are not offered by default views
• Use a chart type different from the charts in the default view
You can temporarily select a different chart type in a default view through the right-click menu.
However, the view returns to the default chart type each time you invoke it from the navigation
pane.
Note: Although you can combine any counters in a single chart, if you combine counters with
different units, the chart shows multiple vertical (“y”) axes with the respective unit labels of the
counters. For example, combining the disk_data_read counter (KB per second) with the
processor_busy counter (percentage) results in a chart with multiple vertical (“y”) axes, where
each axis displays the respective unit label of the counter.
Related information
Boost.Accumulators - www.boost.org/doc/libs/1_38_0/doc/html/accumulators.html
66 | Performance Advisor Administration Guide For Use with DataFabric Manager Server 4.0
Hierarchy of objects
You can view the hierarchy of logical and physical objects in a navigation tree from the navigation
pane.
By default, only the hierarchy of logical objects appears in the navigation pane. From the navigation
pane, click View > Physical to view the hierarchy of physical objects.
Steps
Steps
Steps
Step
1. In the Summary View window, select the view from the Available Views list.
68 | Performance Advisor Administration Guide For Use with DataFabric Manager Server 4.0
Step
Step
1. Click the buttons located next to the Show Threshold box to view the performance data.
• Click l< to move to the beginning of data collection.
• Click > to move forward in the time scale.
• Click < to move backward in the time scale.
• Click >l to move to the ending of the time scale.
Step
Steps
Steps
Steps
Steps
Steps
Steps
Steps
Steps
• SAN
After the diagnosis, Performance Advisor provides observations and recommendations for each
category.
In addition, events are plotted in performance views to mark values that are indicative of
performance degradation. You can correlate events generated for an object and its related objects
during any period of time.
The DataFabric Manager server indicates the impact of performance on a storage system by
generating events such as disks:some-failed. Such events, which are not triggered due to a breach of
user-defined thresholds, are also plotted in the performance views.
Steps
Next topics
When you must create a custom view on page 74
How flexible the custom views are on page 74
Counter data across views on page 74
74 | Performance Advisor Administration Guide For Use with DataFabric Manager Server 4.0
Steps
Steps
Steps
Steps
Steps
Next topics
Methods for preserving performance data on page 77
Saving a chart on page 78
Exporting a view to a .csv file on page 78
Limitations to exporting performance data on page 79
• Exporting a view
Saving a chart
You can save the chart of the performance view with details in your directory.
Steps
Steps
1. From the navigation pane, click the group or the object that you want to view and for which you
want to export data.
2. Click the name of the view you want to display.
3. From the File menu, select Export.
4. Select and specify the date range, the counters, and the date format.
For advanced views, you can specify the metric data that you want to export.
5. Click Browse.
6. Specify the location and the name of the file.
7. Select CSV in the Files of Type field to export the data to CSV format.
8. Click Save.
9. Click OK.
Views in Performance Advisor | 79
Result
The application saves a performance view file with a .csv extension to a location that you specify.
Note: You can view the .csv output file format in Microsoft Excel. You can also view this file
using common text editors, because the data is in plain-text format.
Next topics
Canned views at the global or group level on page 80
Canned views at the storage system level on page 82
Canned views at the vFiler level on page 87
Canned views at the processor level on page 88
Canned views at the network interface level on page 88
Canned views at the dataset level on page 89
80 | Performance Advisor Administration Guide For Use with DataFabric Manager Server 4.0
Top CPUs
This view shows the top five CPUs in a group. The following panels are available under this view:
• cpu_busy
This panel displays the cpu_busy counter.
• recv_packets
This panel displays the recv_packets counter.
• send_data
This panel displays the send_data counter.
• send_errors
This panel displays the send_errors counter.
• send_packets
This panel displays the send_packets counter.
Top Aggregates
This view shows the details of the top five aggregates in the storage system.
The following panels are available under this view:
• total_transfers
This panel displays the aggregate:total_transfers counter.
• user_reads
This panel displays the aggregate:user_reads counter.
• cp_reads
This panel displays the aggregate:cp_reads counter.
• user_writes
This panel displays the aggregate:user_writes counter.
Top LUNs
This view shows the details of the top five LUNs in the storage system.
The following panels are available under this view:
• read_ops
This panel displays the lun:read_ops counter.
• other_ops
This panel displays the lun:other_ops counter.
• write_ops
This panel displays the lun:write_ops counter.
Top Volumes
This view shows the details of the top five volumes in the storage system.
The following panels are available under this view:
• read_ops
This panel displays the volume:read_ops counter.
• other_ops
This panel displays the volume:total_ops counter.
Views in Performance Advisor | 87
• write_ops
This panel displays the volume:write_ops counter.
• vfiler_net_data_recv
This panel displays the vfiler_net_data_recv counter.
• vfiler_net_data_sent
This panel displays the vfiler_net_data_sent counter.
• Throughput in Pkts/Sec
This panel displays the ifnet:recv_packets, ifnet:send_packets, and ifnet:total_packets counters.
The ifnet:total_packets counter is calculated using the following formula:
ifnet:total_packets = ifnet:recv_packets + ifnet:send_packets
• Throughput in Bytes/Sec
This panel displays the ifnet:recv_data, ifnet:send_data, and ifnet:total_data counters.
The ifnet:total_data counter is calculated using the following formula:
ifnet:total_data = ifnet:recv_data + ifnet:send_data
• Packet Errors
This panel displays the ifnet:recv_errors, ifnet:send_errors, and ifnet:total_errors counters.
The ifnet:recv_errors counter is calculated using the following formula:
ifnet:total_errors = ifnet:recv_errors + ifnet:send_errors
• Multicast Packets
This panel displays the ifnet:send_mcasts, ifnet:recv_mcasts, and ifnet:total_mcasts counters.
The ifnet:total_mcasts counter is calculated using the following formula:
ifnet:total_mcasts = ifnet:send_mcasts + ifnet:recv_mcasts
• volume:<volume_name>:nfs_read_ops
• volume:<volume_name>:nfs_write_ops
• volume:<volume_name>:nfs_other_ops
• volume:<volume_name>:nfs_ops
• CIFS Iops by Optype
This panel displays the following counters:
• volume:<volume_name>:cifs_read_ops
• volume:<volume_name>:cifs_write_ops
• volume:<volume_name>:cifs_other_ops
• volume:<volume_name>:cifs_ops
• SAN Iops by Optype
This panel displays the following counters:
• volume:<volume_name>:san_read_ops
• volume:<volume_name>:san_write_ops
• volume:<volume_name>:san_other_ops
• volume:<volume_name>:san_ops
• User Ops v/s Sys Ops
This panel displays the following counters:
• volume_user_ops
• volume_sys_ops
The above counters are calculated using the following formula:
• volume_user_ops = sum(volume:<instance>:nfs_read_ops + nfs_write_ops + nfs_other_ops +
cifs_read_ops + cifs_write_ops + cifs_other_ops + san_read_ops + san_write_ops +
san_other_ops)
• volume_sys_ops = volume:<instance>:total_ops - volume_user_ops
• Flexcache Iops by Optype
The above counters are calculated using the following formula:
• volume:<volume_name>:flexcache_read_ops
• volume:<volume_name>:flexcache_write_ ops
• volume:<volume_name>:flexcache_other_ ops
• write_align_histo
• Unaligned WAFL Ops
This panel displays the following counters:
• unaligned_reads
• unaligned_writes
• Partial blocks
This panel displays the following counters:
• read_partial_blocks
• write_partial_blocks
Target Basic
This view provides the basic details at the target level.
The following panels are available under this view:
• read_ops
This panel displays the read_ops counter.
• write_ops
This panel displays the write_ops counter.
• other_ops
96 | Performance Advisor Administration Guide For Use with DataFabric Manager Server 4.0
Next topics
What alarms are on page 97
Overview of alarm management tasks on page 98
How hosts become visible to Performance Advisor on page 103
What thresholds are on page 106
Alarm management This list describes the alarm management tasks that you perform most
tasks often.
Responding to When you receive an alarm, you should acknowledge the event and resolve
alarms the condition that triggered the alarm. If the repeat notification feature is
enabled and the alarm condition persists, you continue to receive
notifications until you acknowledge the event.
Viewing the list of View the list of all currently configured data protection alarms when you
alarms want an overview of what is configured for any selected group. You can
control the sort order of any column in the list. You can also control which
alarms are displayed by filtering the information in any column in the list.
98 | Performance Advisor Administration Guide For Use with DataFabric Manager Server 4.0
Viewing the View the configuration details of any selected alarm when you want to
configuration values know specific information or you are considering modifying the
for any selected configuration.
alarm
Editing the alarm Modify the configuration of an alarm when you need to accommodate
changes in the environment. For example, you might need to update a
recipient e-mail address, add another recipient pager address, add a
recipient script name, or change the hours during which the alarm is active.
Testing an alarm Test a new or modified alarm by having Protection Manager send a test
message to all the recipients configured for the selected alarm.
Disabling or Disable an alarm when you need to stop its functioning for a while but want
enabling an alarm to retain the alarm configuration. For example, if you have an alarm
configured for the SnapVault Backup Failed event and you scheduled
downtime for maintenance on that host, you might disable the alarm during
the planned downtime. When the host is up again, you can enable the
alarm.
Adding an alarm Configure a new alarm when you want notification of a specific event or
any event of a specific severity type for a group.
Deleting an alarm Delete an alarm only when you are sure that you no longer need notification
of the event or severity type occurring for the group.
Next topics
Monitoring alarms on page 99
Editing alarms on page 99
Testing alarms on page 100
Enabling or disabling alarms on page 100
Adding alarms on page 101
Deleting alarms on page 102
Responding to alarms on page 102
Alarms, hosts, and thresholds | 99
Monitoring alarms
You can view a list of all the alarms configured for your application and the configuration values for
each one.
Steps
1. For licensed applications, from the navigation pane, click Notifications > Alarms. For
Performance Advisor, from the navigation pane, click Set Up > Alarms.
2. (Optional) You can customize the Alarms window in any of the following ways:
• Select an alarm to see the configured property values for that alarm.
• Click in a column header to control which alarm entries you want displayed. The color of
the column header changes to indicate that you are filtering the entries in that column.
Some column filters display a drop-down list to select from and other column filters display a
search field in which you can enter text to select.
• Click the sort arrows in a column header to change the sort order of the entries in that column.
•
Click in the upper-right corner of the list to select which columns are displayed.
• Drag the bottom of the alarms list area up or down to resize that area.
Editing alarms
You can change the configuration of an existing alarm.
Steps
1. For licensed applications, from the navigation pane, click Notifications > Alarms. For
Performance Advisor, from the navigation pane, click Set Up > Alarms.
2. Select an alarm in the Alarms window and click Edit.
3. Use the tabs in the Properties sheet (Group, Event Type, Recipient, and Details) to change the
configured property values.
4. Click Apply to commit your changes.
100 | Performance Advisor Administration Guide For Use with DataFabric Manager Server 4.0
Testing alarms
You can test an alarm to check its configuration, after creating or editing the alarm.
Steps
1. For licensed applications, from the navigation pane, click Notifications > Alarms. For
Performance Advisor, from the navigation pane, click Set Up > Alarms.
2. Select an alarm and click Test.
You can select any alarm regardless of whether it is enabled or disabled.
3. Click OK in the Confirm Alarm(s) Test window to begin the test.
A test event notification is sent to each configured recipient. If a script is configured, the test
notification runs the script.
4. Verify the alarm recipient configuration by checking that each recipient received a test message.
If a script recipient is configured, check that the script was successfully started.
and you scheduled down time for maintenance on a host that is in the resource pool for the group
associated with that alarm, you might disable the alarm during the planned down time. When the host
is up again, you can enable the alarm to resume its functioning.
Steps
1. For licensed applications, from the navigation pane, click Notifications > Alarms. For
Performance Advisor, from the navigation pane, click Set Up > Alarms.
2. In the alarms list, locate the entry for the alarm you want to enable or disable.
3. Clear the Enabled check box to disable the associated alarm, or select the Enabled check box to
enable it.
The alarm is immediately disabled or enabled.
4. Verify your change by viewing the results that are displayed in the Alarms window.
Adding alarms
You can add an alarm when you want immediate notification that a specified event or event class or
event of a specified severity level occurred.
Steps
1. For licensed applications, from the navigation pane, click Notifications > Alarms. For
Performance Advisor, from the navigation pane, click the Set Up Alarms window.
2. Click Add to start the Add Alarm wizard.
3. On each page of the wizard, enter the appropriate information.
4. Click Finish to commit your choices and close the Add Alarm wizard.
5. Verify the creation and configuration of the alarm by viewing the results that are displayed in the
Alarms window.
102 | Performance Advisor Administration Guide For Use with DataFabric Manager Server 4.0
Deleting alarms
You can delete an alarm when you no longer need immediate notification of an event or severity
type.
Steps
1. For licensed applications, from the navigation pane, click Notifications > Alarms. For
Performance Advisor, from the navigation pane, click Set Up > Alarms.
2. Select one or more alarms in the Alarms window and click Delete.
3. Click OK in the Confirm Alarm(s) Delete dialog box to delete the selected alarms.
The alarm configurations are deleted and the alarms are removed from the Alarms list.
4. Verify your deletion by viewing the results that are displayed in the Alarms window.
Responding to alarms
When you receive an alarm, you can acknowledge the event and resolve the condition that triggered
the alarm.
Steps
1. For licensed applications, from the navigation pane, click Notifications > Events. For
Performance Advisor, from the navigation pane, click Monitor > Events.
2. Select an event whose details you want to view.
Alarms, hosts, and thresholds | 103
3. Click Acknowledge to indicate that you are taking responsibility for managing that event.
4. Click Yes in the Acknowledge Events dialog box to acknowledge the selected event.
Your user name and the time are entered in the Events list for the selected event.
5. Find the cause of the event and take corrective action.
Next topics
List of fields in the hosts table on page 103
Editing host credentials on page 104
Configuration of data collection on page 104
Copying the data collection configuration on page 105
Effect of increasing sample rates on page 105
Setting retention periods and sample rates on page 105
Field Description
Data Collection Enabled Check box to enable or disable data collection from
the host
Login Credential Describes whether the user was authenticated on the
host
Status The status of the host
Steps
Steps
Steps
Next topics
What resource properties are on page 107
Viewing threshold values on page 107
When to use thresholds in combination on page 107
What the Details tab is on page 107
Adding thresholds on page 108
Editing thresholds using the Summary tab on page 108
Editing thresholds using charts on page 109
Alarms, hosts, and thresholds | 107
Steps
Next topics
What logical objects are on page 108
What physical objects are on page 108
Adding thresholds
You can add thresholds for the performance counters on storage systems using the Set Up
Thresholds window. Alternatively, you can right-click the object in the logical or physical hierarchy
to add thresholds.
Steps
Steps
Steps
Steps
7. Click Apply.
8. Click OK.
Deleting thresholds
You can delete thresholds by using the Summary tab in the Set Up Thresholds window.
Steps
Creating alarms
You can create alarms by using the Add Alarm wizard.
Steps
apply to the object. For example, you can create a threshold template that contains a threshold for the
cp_writes counter greater than 10 per second for 5 minutes. You can apply these values in a template
to objects to generate events.
You can create and view templates using the Set Up Threshold Templates window, which is
divided into two halves. The top half of the Set Up Threshold Templates window displays the
template table, and the bottom half lists the events, counters, and resource properties. The object tab
in the bottom-half of the page displays the objects that are applied to the selected template.
Next topics
When to use threshold templates on page 111
Creating threshold templates on page 111
Editing threshold templates on page 112
Steps
Steps
Glossary
lower threshold The type of threshold set for an event generation when the counter value falls
and remains below the lower threshold value for longer than the Threshold
Interval specified.
managed object A managed object represents any object that has an identity and a name in the
DataFabric Manager object table. A managed object is an object that is
contained within a DataFabric Manager group. Volumes, aggregates, qtrees,
and LUNs are examples of managed objects.
NetApp NetApp Management Console is the client platform for the Java-based NetApp
Management Manageability Software applications. NetApp Management Console runs on a
Console Windows or Linux workstation, separate from the system on which the
DataFabric Manager server is installed.
object Typically there is an object associated with each hardware or software
subsystem within Data ONTAP. Examples of hardware objects are processor,
disk, NVRAM, and networking card objects. FCP, iSCSI, CIFS, and NFS are
the examples of software protocol objects. WAFL, RAID and target are the
examples of internal objects specific to Data ONTAP. Virtual objects like the
system object capture key statistics across all the other objects in one single
place. Examples of system objects are avg_processor_busy, nfs_ops, cifs_ops,
and net_data_recv.
Performance The Performance Advisor component installed on the NetApp Management
Advisor Console platform enables you to monitor the performance of storage systems
and vFiler units.
The user interface of Performance Advisor contains only performance-
monitoring information. This Performance Advisor interface is distinct from
Operations Manager, which contains other DataFabric Manager information.
performance- The Performance Advisor component that is enabled on the DataFabric
monitoring Manager server to collect and archive performance data at regular intervals
server from the monitored storage systems and vFiler units. Performance Advisor
gathers sets of data that the performance-monitoring server has collected to
generate its graphical charts and views.
performance A collection of one or more counters accessible through Performance Advisor
view for a group of storage systems or subsystems through one or more charts.
Performance Advisor displays canned views that are built in to the system, and
custom views.
physical objects Object types that represent the physical resources in a storage system such as
disk, aggregates, memory, network interfaces, resource pools, and RAID
groups are known as Physical Objects.
physical The hierarchy that displays the physical objects and instances.
hierarchy
Glossary | 115
real-time data Data that is passed through Performance Advisor for display but is not stored.
Real-time data is collected only as long as the performance view window
displaying it is open on Performance Advisor. Real-time data is accessible only
to the client by which it is retrieved.
Real-time data is suitable for diagnosing immediate performance issues.
resource pool A managed object in DataFabric Manager, containing storage provisioning
resources like storage systems, aggregates, and spare disks.
retention period The amount of previously collected data for a counter that is accessible to the
user. You can set the retention period based on units of a minute, hour, day,
week, month, or year.
sample rate The rate at which data is collected for a counter. You can set the sample rate
based on units of a minute.
storage set Containers that are used for delegation, replication, and in some cases,
substorage provisioning. The only container of merit in a storage set is a
volume (flexible or traditional). A storage set contains a group of volumes
whereas a volume should be in at most one storage set.
storage system An appliance that is attached to a computer network and is used for data
storage. FAS appliances and NearStore systems are examples of storage
systems.
templates and A template is a view definition that applies to object instances known to
canned templates Performance Advisor.
A set of pre-configured templates in Performance Advisor is known as Canned
Template.
threshold The amount of time in seconds for which an event generation is suppressed
interval before Performance Advisor decides that a counter has crossed a specified
threshold and an event needs to be generated. The same interval is also be used
to generate a normal event.
unmanaged Objects apart from the managed objects belong to the class of unmanaged
object objects. An unmanaged object does not have a unique identity in the
DataFabric Manager table.
upper threshold The type of threshold set for an event generation when the counter value
exceeds and remains above the higher threshold value for longer than the
Threshold Interval specified.
vFiler unit One or more virtual storage systems that can be configured on a single physical
storage system licensed for the MultiStore feature. DataFabric Manager 3.4 and
later enables monitoring and management of vFiler units.
view A collection of related panels represented together and displayed by the
Performance Advisor client.
Index | 117
Index
A selecting columns to display 30
Tasks bar, hiding and redisplaying 30
access permissions counters
overview 45 about 34
Add Alarm wizard shared across views 74
task 101 Custom view
administrator roles multiple axis 74
requirements for Performance Advisor tasks 25 purpose 73
advanced views when to create 74
creating 70 custom views
alarms about 64
adding 101 creating 75
deleting 102 deleting 76
disabling 100 editing 76
enabling 100 flexible 74
list of monitoring and managing tasks for 98 opening 76
modifying 99
monitoring 99
pausing 100
D
responding to 102 dashboards
testing 100 described 59
viewing 99 panels, descriptions of 59
applications in NetApp Management Console 27 dashboards, performance
descriptions of panels 59
C dashboards, protection
descriptions of 59
canned views 64 dashboards, provisioning
charts descriptions of 59
bar chart 68 data collection
changing the time period 69 configuration 104
changing the type 71 copying configuration 105
polyline 68 DataFabric Manager
resetting axes 69 client platform for 27
saving 78
scatter 68
client platforms 27
E
console events
adjusting column widths 30 definition of 60
applications supported 27 deleting from event log 62
applications, toggling 29 determining when they occur 60
filtering data display 31 responding when an event occurs 61
Get Started feature 29 viewing the list of 61
Help, accessing 29
logging out 29
moving backward and forward 29 F
rearranging column order 30
filtering data display 31
118 | Performance Advisor Administration Guide For Use with DataFabric Manager Server 4.0
overview 23
requirement for starting 51
G secure connections for 52
Get Started feature 29 starting 55
stopping 56
performance counters
H list of 35, 37–42
Performance dashboard 59
Help performance data
accessing 29 exporting 78
hierarchical groups methods for preserving 77
support for 24 viewing 67
host credentials performance issues
editing 104 diagnosing 73
diagnosis 72
L performance monitoring server
location in the network topology 23
licenses performance objects
NetApp Management Console applications 27, 28 about 33
limitations, exporting data 79 list of 33
limitations, Performance Advisor 24 performance views
logging out 29 working with data in 68
logs performance-monitoring server
for viewing the list of events 61 enabling 51
M R
monitoring real-time data
dashboard panels, overview of 59 viewing 75
dashboards, overview of 59 regular expressions, filtering examples 31
events 61
S
N
sample buffer, setting 105
NetApp Management Console sample rates
about 27 Increasing 105
applications supported 27 Secure connections
defined 27 support for 25
licenses 28 SSL
See also console certificate for 25
storage systems
CPU utilization 77
P CPU utilization limitations 24
Performance Advisor display unavailable 56
accessing 55 refreshing the list 57
application described 27 synchronize
hosts in 103 charts across all views 69
location in the network topology 23 charts in a view 69
log in to 24 undo 70
Index | 119
T views
advanced 65
Tasks bar, hiding and redisplaying 30 aggregate 94
thresholds blank charts 57
about 106 canned 79
add 71 chart axes 74
adding 108 dataset 89
deleting 110 disk 94
editing 108 displaying 66
logical hierarchy 108 event correlation 72
physical hierarchy 108 global 80, 81
view 70 group 80, 81
Thresholds how to select 67
adding 111 LUN 93
adding alarms 110 navigation 66
Details tab 107 network interface 88
editing 109, 112 processor 88
transport protocols 53 qtree 92
resource pool 89
U storage system 82–86
target 95
unavailable menu options 56 vFiler unit 87
volume 90, 91
V
vFiler units Z
display unavailable 56 zero-value data
refreshing the list 57 display of 56
support for 25