Infrastructure Management Software Blog

Universal Log Management to Improve Troubleshooting

Both Operations Manager and SiteScope include log file management capabilities, But ArcSight Logger’s additional functionality makes the Operations Bridge team even more productive in searching log files for recurring patterns that affect system availability and performance.

Discovery: Loading data into the Run-time Service Model

This is the second post in a series on the Run-time Service Model. It covers all the different discovery methods we use to load information about configuration items (CIs) into the Run-time Service Model.

OH, IL, WI, IN, MI Operations Center Technical Roadshow - April 20th to April 29th - Don't miss it!

Ever wish you could talk face-to-face with more technical people about Operations Center and Network Management Center products? Don’t really have the time or budget to travel very far to do so?  Well, here is a great opportunity to meet and talk with technical experts on products like Operations Manager and NNMi – right in your background.

Vivit will be hosting a series of six (6) one-day sessions, where there will be a nice mix between presentations and Q&A sessions around these products.  The sessions will be held in the following states on the following days:

- (Columbus) Ohio – April 20, 2010

- (Orrville) Ohio – April 21, 2010

- (Dearborn) Michigan – April 22, 2010

- Wisconsin – April 27, 2010

- (Chicago) Illinois – April 28, 2010

 - (Fishers) Indiana – April 29, 2010

Feel free to contact me if you have any further questions about this roadshow at

Labels: agent| agentless| agentless monitoring| agents| automating operations management| automation| BES| BlackBerry Enterprise Server| CMDB| consolidate events| consolidated event| Consolidated Event and Performance Management| consolidated event management| Consolidated Management| correlate events| DDM| Discovery and Dependency Mapping| event console| event consolidation| event correlation| event management| Hewlett Packard| HP Network Node Manager| HP OMi| HP OpenView| HP Operations Center| HP Operations Manager| infrastructure management| infrastructure monitoring| IT dashboard| IT infrastructure management| IT infrastructure monitoring| IT management| manager of managers| managing IT| managing IT infrastructure| managing IT operations| monitoring| Network Management| Network Node Manager| NNM| NNMi| Norm Follett| OM| OMi| OML| OMU| OMU 9.0| OMW| OpenView| OpenView Operations| Operations Center| Operations Manager| Operations Manager i| Operations Manager on Linux| Operations Manager on Unix| Operations Manager on Windows| performance| Performance Agent| performance management| Performance Manager| performance monitoring| SiteScope| Smart Plug-in| Sonja Hickey| SPI| TBEC| Topology Based Event Correlation| topology-based event correlation| virtual server| virtual servers| virtual systems management| virtualization management| Virtualization SPI| virtualization sprawl| virtualization strategy| virtualizationation| virtualized environment| virtualized environments| Virtualized Infrastructure| Vivit

Learn how Independence Blue Cross reduced IT Operations costs

Join HP Software and Solutions for a live InformationWeek webcast with special guests Maryann Phillip, Director of Service Delivery at Independence Blue Cross (IBC), and Ken Herold, Practice Manager & Principal Architect with Melillo Consulting.

Hear first-hand how IBC is using HP Operations Center products like Operations Manager, Performance Manager, and DDM in addition to agentless and agent-based data collection to:

  • achieve profitable growth through enabling technologies

  • reduce costs by achieving a competitive cost structure

  • manage medical costs better -- through operational stability & improvements

Register today and learn how you can streamline and make YOUR processes more efficient.

Smart Plug-Ins (SPIs) and Agent-Based/Agentless Data Collection Explained: What you need to manage your IT environment

Have questions about HP Operations Center SPIs like the infrastructure SPI or the SPI for virtualization? Wondering what agents you need to get? And how many? The following post is a good high-level summary of what you need and where you need it. Read on …

In general, an HP Operations Manager solution consists basically of two things: an Operations Manager server along with data collection technologies. Data collection technologies, at a very high level, are either agent-based or agentless. The purpose of this post is to explain the latter two: agent-based data collection and agentless data collection and what you need to implement a solution comprised of both. The following two figures provide high-level architectural representations of the following discussion.












Agent-based data collection explained

HP Operations Manager agents collect, aggregate, and correlate monitoring information to manage data and events collected and aggregated from multiple sources. The agents can suppress irrelevant and duplicate events and correlate the remaining relevant events to produce actionable and enriched management information. In addition, dependencies and propagation rules show the cause of an incident, which assist in reducing mean-time-to-recovery and downtime. Agents are installed on each managed system or node, regardless if it is a physical or a virtual machine, and have the following additional capabilities:

· Allow the addition and customization of monitoring sources not included in out-of-the-box monitoring policies.

· Collect and analyze performance data from operating systems and installed applications and use historical patterns to establish performance baselines.

· Autonomously perform automated corrective actions (in isolation from the Operations Manager server) and manage by exception (forward only actionable events to the Operations Manager server through the use of intelligent filtering, duplicate suppression, and correlation techniques).

· Set up HTTPS communication with the Operations Manager server – even in outbound-only communications configurations.

· Support monitoring data center technologies such as virtualization and clusters.

Agentless data collection explained

Agentless data collection, through the use of HP SiteScope monitoring probes, complements agent-based data collection by providing flexibility in how information is gathered from the IT environment. Like agent-based data collection, agentless monitoring is performed on both physical and virtual systems and has the following capabilities:

· Gathers detailed performance data for infrastructure targets without installing an agent on the managed node.

· Provides easy monitoring of the IT infrastructure.

· Has an intuitive user interface.

· Allows actions to be initiated automatically when a monitor’s status changes.

· Provides solution templates that enable quick deployment of monitoring probes, which include specialized monitors, default metrics, proactive tests, and best practices for an application or monitoring component.

· Has the ability to monitor previously unmanaged or hard-to-manage systems and devices through easy-to-use customization tools.

Infrastructure Smart Plug-Ins – what are they and where do they fit in?

Infrastructure Smart Plug-ins supplement agents by collecting data at the infrastructure or managed systems level. They provide out-of-the-box, packaged, and intelligent management and are comprised of the following three SPIs:

· The “system” SPI discovers operation system and platform resources, generates alerts on system diagnostic events, monitors system services and processes, and monitors resource utilization.

· The “cluster” SPI automatically discovers and represents cluster nodes and configured resource groups in a clustered environment, monitors cluster services and processes, and enables monitoring of clustered applications - even as they move “on-the-fly” between cluster servers.

· The “virtualization” SPI -which is supported on the most common virtualization hypervisors - discovers and monitors virtualization platforms (both host and virtual machines) and provides graphs and reports on resource utilization.

Application SPIs – are these different than infrastructure SPIs?

In a sense, yes and no. Yes because they basically perform the same functions as infrastructure SPIs in terms of collecting, aggregating, and correlating monitoring information. No in terms of what data they are responsible for. Infrastructure SPIs, as mentioned before, do this at the system level, whereas application SPIs do this at the application level. The following picture builds on the previous one, but more clearly depicts where application and infrastructure SPIs reside:



And does HP have SPIs! We have SPIs for databases (Oracle, Informix, Microsoft SQL Server, IBM DB2), web application servers (IBM WebSphere, JBoss, Oracle WebLogic), storage (HP Storage Area Manager, Veritas NetBackup and Volume Manager), and ERP/CRM (PeopleSoft, SAP, Siebel) products. Not to mention lots of SPIs developed by HP partners around Cisco, Novell NetWare, and Documentum products.  

New Licensing of Agents and SPIs!

Yes, we’ve changed our licensing structure for both SPIs and Agents. They are now instance-based, meaning you have one per operating system or application instance. Plus, we’ve got this great new “Operating System Instance Advanced License”, which includes the following:

o Operations Manager agents

o “System” Smart Plug-In

o “Cluster” Smart Plug-In

o 15 agentless monitoring probes/points

If monitoring a virtualized environment, a “virtualization” SPI - although an infrastructure SPI like the “system” and “cluster” SPIs - is purchased separately from the Operating System Instance Advanced License. One virtualization SPI is required for each monitored Virtual Server host.

I hope that this has helped explained agents and SPIs. If you have any questions about this post or instrumentation in general, please feel free to comment on this post.

For HP Operations Center, Sonja Hickey.

Get the latest updates on our Twitter feed @HPITOps

Join the HP OpenView & Operations Management group on LinkedIn.

The full stack (OMW, SiteScope, OMi, NNM, Service Desk, CMDB)

As I was getting ready to leave yesterday, a colleague stopped by my desk and asked “do you want to be a hero?” That certainly peaked my interest. It turned out we had a customer downstairs in our executive briefing center that wanted some clarification about all the pieces of our stack fit together.

The customer was the CTO of a major IT firm in the Asia-Pacific region. They manage approximately 4,000 servers using OMW 8.1. They use both agents and SPIs, as well as SiteScope agentless monitoring. In addition, they monitor the faults and performance of their network using NNM, and roll those events into their Operations Manager console. In addition, they use Service Desk 4.5 along with a CMDB (configuration management database) that tracks all the configuration items and relationships among them across their enterprise. A *very* rough schematic  of what they have appears in the diagram below in red.

 Our discussions were divided into two main areas:
1. What they are doing today and what they should be doing?
2. What can they do in the future?

Current Situation
The first question was about best practices. Were they using the software correctly to manage their infrastructure? The answer is a resounding yes. They use OMW as the central event management console, collecting data from agents, SiteScope (agentless monitoring), and NNM for network events.

And, they integrate their service desk with OMW, opening and closing tickets, and tracking changes to the IT infrastructure in their CMDB. They implemented the CMDB about two years ago, in conjunction with their Service Desk implementation.

Next Steps
The next questions focused on what should they be doing or what can they do next to improve their IT management.

We started with a discussion about OMi. The customer was confused about how OMi fits with OMW - the first question was whether it replaces OMW, whether they receive OMi as part of an upgrade (entitlement), and finally, what specific value OMi provides since they currently use OMW as the centralized event consolidation tool.

As readers of this blog know, OMi is a separate product that adds on to Operations Manager. (See green box at top of the above diagram). Its main value is that it leverage the system topology information in the CMDB to greatly speed the time to repair IT problems, especially in complex environments. We have many resources to learn more about OMi, including:
Product overview
High-level webinar on OMi
Deep-dive technical webinar on OMi
Answers to technical questions on OMi

The next topic was automation. We talked about how companies use Operations Orchestration (OO)  to automate their IT processes (runbooks). OO uses events in OM to trigger its process flows. The good news was that this customer has spent the past two years documenting and improving their IT processes. They already know what processes occur frequently and how much manual effort they require. This may be the next logical step for them as it leverages their existing IT infrastructure and processes. EMA recently write a white paper on how process automation augments event consolidation.

Migration Challenges
One issue that arose was that the CMDB connected to their service desk is not the latest UCMDB that OMi uses for its topology-based event correlation (TBEC). The customer has two options here.
1. Leave the existing CMDB in place and let OMi create an operational data store that contains the configuration information it needs. The advantage of this approach is that it leaves the current management infrastructure intact and just adds OMi on top. OMi uses the SPIs to auto-discover the IT infrastructure and relationships among the elements. OMi’s data store is self-contained and requires minimal external input.
2. Migrate the existing CMDB associated with Service Desk to the latest version of UCMDB. The advantage of this approach is that the customer ends up with a single CMDB. They can migrate their existing data using a tool such as ICM (information consolidation manager) from Netscope.

For organizations already integrating their events into a single Operations Manager console, you are on the right track. If you already use a CMDB to track your IT infrastructure, you are very far along the IT management maturity curve, even more so if you use some means of automatic discovery to keep it current.

To take things to the next level, you have two options: focus on further event correlation and reduction with OMi or automate your existing IT processes with Operations Orchestration. You can pursue these in series or in parallel, depending on your priorities. Both will deliver a tangible return on investment and fast payback period.

For HP Operations Center, Peter Spielvogel.

Get the latest updates on our Twitter feed @HPITOps

Join the HP OpenView & Operations Management group onLinkedIn.

Controlling SiteScope from Operations Manager

I have been getting many questions from both customers and colleagues about how Operations Manager and SiteScope work together. This is a very timely topic as we have some new capabilities connecting SiteScope and Operations Manager.

Since many readers will not be attending my talk about this topic at Software Universe, I’ll preview the information here. (Alex Ryals and I will be focusing on customer success stories in our presentation, so it is still very much worth attending, even if you already know the product integration part.)

The main role of Operations Manger is to serve as an enterprise event console or operations bridge, consolidating events from various domain managers for servers, storage, and networks, from both HP and other vendors. It accomplishes this using agents that run on each managed node, monitoring availability and performance. These agents send information to the Operations Manager server based on user-defined policies. The agents can also act autonomously, performing corrective actions without communicating with the server. This is very useful for minimizing network traffic, or even assuring operation if a connection between the server and managed node gets interrupted.

SiteScope complements this mission by monitoring servers and the applications running on them using agent-less technologies. SiteScope too monitors both HP and other hardware. In some cases, enterprises have some servers on which administrators either cannot or will not install agents. In other cases, customers will monitor servers using a combination of both agent-based and agent-less technology. One common example is for monitoring email environments running Microsoft Exchange, Active Directory and all the supporting infrastructure.

So, how do Operations Manager and SiteScope fit together?

  • SiteScope forwards its events into Operations Manager with the full details.

  • SiteScope targets also appear in the Operations Manger Service Map.

  • Operations Manager lets you control multiple SiteScope servers, including transferring configuration information from one SiteScope instance to another and synchronize settings between multiple SiteScope servers.

The ability to monitor your IT infrastructure using a combination of both agent-based and agent-less technology lets you simultaneously improve the quality of service and reduce IT management costs.

For HP Operations Center, Peter Spielvogel

Showing results for 
Search instead for 
Do you mean 
HP Blog

HP Software Solutions Blog


Follow Us
The opinions expressed above are the personal opinions of the authors, not of HP. By using this site, you accept the Terms of Use and Rules of Participation.