Thursday Feb 05, 2015

Access Point counts in the OCDoctor

When you're trying to figure out questions of scaling in Ops Center, it's important to be able to tell exactly how much load the different parts of the infrastructure are handling. In Ops Center, the relevant number is the access point count. An access point is a connection between a Proxy Controller and a managed asset. We don't just count the number of assets directly because, if different parts of an asset are managed by different Proxy Controllers, that puts more load on the system.

There's a tool in the OCDoctor's toolbox directory that lets you count the access points in your environment. The AssetCount.sh script gives you the total number of access points managed by the Enterprise Controller, and gives additional information depending on which option you use:

  • The standard option shows the number of access points for each Proxy Controller - both the total and a detailed breakdown by asset category.
  • The machine option gives a list of the access points on each Proxy Controller in machine-readable format.
  • The agent option shows, for each Proxy Controller, how many assets are agent-managed, how many are agentless, and how many are SPs.

The Scaling and Performance Guide explains how to use the AssetCount.sh script. You can find it in the version 12.2.2 documentation library.

Thursday Jan 29, 2015

Database Versions

I've seen a few questions about what Oracle Database versions are supported with which versions of Ops Center, so I thought I'd clarify that point a bit.

 If you're using Ops Center 12.2.2, the latest version, and you have a customer-managed database, you can use Oracle Database 11.2.0.3 or 11.2.0.4. If you're using an embedded database, the upgrade to Ops Center 12.2.2 includes an upgrade of the embedded database to 11.2.0.4.

If you're using Ops Center 12.2.0 or 12.2.1, the supported database version is 11.2.0.3, for both embedded and customer-managed databases.

Thursday Jan 22, 2015

Enabling and Testing ASR

The Auto Service Request feature in Ops Center, when it's properly enabled, can automatically generate service requests based on qualified incidents in Ops Center, helping you get the systems up and running again quickly. The questions, then, are how do you enable it properly, and short of taking a crowbar to your new M7, how can you be certain it's working?

One of the new How-Tos that we've just released, Enabling and Testing Auto Service Request, can help you answer these questions. It talks about the prerequisites, explains how to add contact information and enable ASR, and shows you how to use a test trap to verify that ASRs are being created correctly.

This How-To is just one of the improvements in Ops Center 12.2.2. Take a look at the What's New for more information about these improvements, and the Deploy and Operate libraries for more how-tos.

Thursday Jan 15, 2015

Proxy Controller Backups

There were a number of new features introduced in Ops Center 12.2.2. One of the shiny ones is an expansion of the backup and recovery capabilities to include Proxy Controllers.

It's always been possible to create a backup of the Enterprise Controller and the co-located Proxy Controller using the ecadm backup command. Now, in version 12.2.2, there's a similar command - proxyadm backup - to create a backup of a Proxy Controller, including all of its asset data. You can save off this file and use it to restore the Proxy Controller to its backed-up state.

The only trick with OC backups is that the OS images directory isn't part of the backup file for size reasons, but they can be backed up separately.

The Backup and Recovery chapter of the Administration guide explains how to back up your Proxy Controllers and Enterprise Controller, and the What's New goes over many of the other new features in 12.2.2.

Thursday Jan 08, 2015

Sizing and Performance Guide

When you're planning an Ops Center deployment, planning to expand a datacenter where Ops Center is installed, or looking to optimize your Ops Center deployment, it's vital to have information about scaling and performance. You need to know whether your new systems will need another Proxy Controller, or whether the system you're planning on using for your Enterprise Controller is beefy enough.

We've added a Sizing and Performance Guide to the Ops Center library, to help you answer these sorts of questions for your environment:

  • The Resource Utilization chapter discusses the relative resource usage of common Ops Center uses, such as OS provisioning, update management, and virtualization.
  • The Scaling and Performance Guidelines chapter provides detailed information about the resources used by, and the scaling capabilities of, Ops Center components such as the Enterprise Controller, Proxy Controllers, database, virtualization controllers, and networks.
  • The Reference Systems chapter provides several reference system specifications for you to use in your planning.
  • The Report Service Configuration Properties appendix explains how to edit the reporting properties, or disable reporting entirely, to improve performance.

You can find this document, along with the rest of the documentation, in the Ops Center 12.2.2 documentation library.

Thursday Dec 18, 2014

Upgrading to version 12.2.2

Now that version 12.2.2 of Ops Center is out, you might be interested in upgrading.

We've put together a guide that walks you through the upgrade process. Each type of environment - HA or single Enterprise Controller - has separate chapters explaining how to do the upgrade through the UI or from the command line.

Something to note is that, if you're using an embedded database, the upgrade includes an upgrade of the DB to Oracle Database 11.2.0.4. If you're using an embedded database, you'll have to log into MOS and download the new DB bundles. The doc explains this step too, but if you've done upgrades in the past it'll be new to you. If you're using a customer-managed database, 11.2.0.4 is also now officially supported.

Thursday Dec 11, 2014

Ops Center 12.2.2 now available

Ops Center version 12.2.2 is now available.

The new version includes a number of enhancements, including:

  • Performance improvements
  • An upgrade of the embedded database to version 11.2.0.4
  • Improved discovery of non-global zones and logical domains
  • Backup and recovery for Proxy Controllers
  • Improvements for Oracle VM Server for SPARC
  • Overlapping network support for Oracle SuperCluster
  • Improvements to Auto Service Request (ASR)
Keep watching this blog for more information about the new content and information about how to upgrade.

Thursday Dec 04, 2014

Command Line Interface

I've seen a couple of questions recently about the Ops Center CLI - what it can do and where it's documented - so I thought I'd give you a brief rundown.

The Ops Center command-line interface lets you perform many of the same tasks that you can perform through the UI, including asset discovery and management, OS and firmware provisioning, and administrative tasks like managing roles and jobs. Not every UI action has an equivalent CLI action, but there are several tables in the CLI guide that explain what it can and cannot do.

The CLI also gives you the ability to script some Ops Center actions, letting you automate regular tasks and run them without using the UI.

For more information about the CLI, take a look at the CLI guide.

Thursday Nov 20, 2014

Restoring an Enterprise Controller on a new system

An Ops Center user was recently planning a deployment, and they were wondering about disaster recovery. Apart from Ops Center's High Availability option, they asked, what recovery options are available if an EC system completely dies?

One other option that you can use is Ops Center's backup and restore capability. If you've made a backup of your Enterprise Controller, it's possible to restore that Enterprise Controller on a new system. There are a few tricks with doing so, though:

  • The new system has to have the same version of Ops Center, including all upgrades and IDRs.
  • If the IP address of the new system is different, you have to manually change the Proxy Controllers to point to the new IP.
  • Ops Center libraries are not part of the backup, so if they're on nfs storage, they need to be accessible to the new system as well.
  • The /var/opt/sun/xvm/images/os directory, containing OS images, needs to be backed up and restored separately, because it's big.

The Backup and Restore chapter explains how to restore the Enterprise Controller on a new system.

Tuesday Nov 18, 2014

Search on the New Library

The search option in the new docs.oracle.com  layout has been frustrating a lot of people. When you first try it out, it seems like many of the options that used to be there - like searching within a specific library - are gone.

The good news is that you can still narrow your search to a specific product category. When you do a search, you can select a product category on the left to narrow your search:


You can then click down below, where it says "6 products", to narrow your search to the Ops Center 12.2.1 library:

If you do a search from inside one of the books in a library, this is done automatically.

Thursday Nov 13, 2014

New Library Design

As you may have noticed, we've revamped the design for docs.oracle.com. While the new look and feel lets us present information for a lot of products in a clearer (and snazzier) format, it also means that the path to the Ops Center docs from the main page is different.

The new library front page has a set of icons for the various categories of documentation. To get to Ops Center, you'll click on the Enterprise Manager category.


This brings up the main page for Enterprise Manager software, which has a tab for Ops Center. You'll click that tab:

Then, once the tab is displayed, you'll click the version of Ops Center that you want:


It's worth noting that the urls for the Ops Center libraries haven't changed, so if you have them bookmarked you don't need to change anything. You can also still get to the library through the help button at the top of the Ops Center UI. This only affects how you get to those libraries through the landing page.

Thursday Nov 06, 2014

Incident Icons

Ops Center packs a lot of information into a browser window. The upside is that you have a lot of information available, but the downside is that some of the information is presented using icons that can be confusing at first look.

The incident icons are one example of this sort of concentrated information. If you select an asset in Ops Center, you'll see a set of icons in the top-right section of its display:


These icons indicate the number of critical, warning, and info-level incidents on the selected asset. You can click any of the icons to go to the incidents tab:


There's also another set of icons in the top left that give you a broader picture of the incidents in your environment:

The first two icons show you how many unassigned critical and warning-level incidents there are in your environment. The middle icon shows the number of relayed incidents. The right two icons show how many critical and warning-level incidents have been assigned to you (that is, whoever is currently logged in.) Clicking on any of the icons takes you to the relevant section of the message center.

The Managing Incidents how-to has more information about these icons and about how you can manage incidents in Ops Center.

Thursday Oct 30, 2014

Running Scripts Using an Operational Profile

I saw an interesting question recently about operational profiles. An Ops Center user wanted to run a script with several variables on several managed systems, with the variables having different values on each system, and they wanted to know the best way to do it.

This can all be done through an operational profile (and an operational plan, which runs the profile). First, you'll select Operational Profiles in the Plan Management section of the UI, then you'll click Create Profile.

You put your script in here, using whatever variables you need. You also specify a type - Remote Shell scripts can be run on any system with an agent, using root permissions, while EC Shell scripts are only run on the Enterprise Controller system with the logged-in user's credentials.

When you click next, there's a screen to define the variables used. You specify the variables, but leave the values blank. Then, when you run the operational plan on a system, you'll be prompted to supply the values for each variable.


Take a look at the Plans and Profiles chapter of the Feature Reference Guide for more information.

Thursday Oct 23, 2014

Managing user access in multiple sites

An Ops Center user who's setting up their environment sent in a question about their users:

"I'm looking to manage three different data centers from one Enterprise Controller instance. However, the three data centers have different administrators. How can I make sure that each administrator can see and manage only the resources that they're supposed to?"

The answer here is that you can use Ops Center's asset groups, combined with its fine-grained roles capabilities, to control which users can see and do what.

First, you create a new asset group in the Assets section of the UI. In this example, I'm creating a group for one of the three data centers:


Once you've created the group, you can add the correct assets to it, by selecting the assets and clicking Add Asset to Group:


Now that you have the assets for one of the datacenters grouped together, you add that admin to Ops Center:


Then you'll select that user and click the Manage User Roles icon. When the wizard comes up, you make sure they have the correct roles, then deselect the "Use the default role associations" checkbox:


When you click next, you select which groups the roles should apply to. So, for this user, we can apply their Asset Admin role only to the Data Center A group:


And there you have it. Rinse and repeat for other groups and users, and each user will be able to see and manage only the correct assets. For more information, check out the Asset Management and User and Role Management chapters.

Thursday Oct 16, 2014

How-To: Creating and Managing Network Domains

Efficient networking is crucial for a virtual datacenter. (I suppose it's crucial for any kind of datacenter, but I'm focusing here.) In Ops Center, you can keep your vDC networks running using network domains. Network domains manage public networks with defined resources as well as ad-hoc private networks.

To create a network domain through Ops Center, you need to identify at least one fabric to support it, then run the Create Network Domain wizard. This wizard lets you create new dynamic private networks for the new domain and associate existing networks with it.

Once you've created a network domain, you can associate it with a server pool by selecting the server pool and running the Associate Network Domain wizard. This wizard lets you pick a network domain, then pick a physical interface on each server in the server pool for the network domain to use. Once you've done this, the server pool can use public and private networks within the network domain.

The Creating and Managing Network Domains how-to provides more background information and walks you through both of these wizards.

About

This blog discusses issues encountered in Ops Center and highlights the ways in which the documentation can help you

Search

Archives
« May 2015
SunMonTueWedThuFriSat
     
1
2
3
4
5
6
8
9
10
11
12
13
15
16
17
18
19
20
22
23
24
25
26
27
28
29
30
31
      
Today