Sorting out Fixpack confusion

I get different variations of this question all the time. "If I install a given fixpack on my ITM infrastructure (TEPS TEMS etc), how do I know what z/OS PTFs I need to apply"? The problem, of course, is that ITM fixpacks are not SMP/E maintenance, so how do I correlate the two maintenance items?

There are a couple web pages that may help you sort out some of this fixpack question. The following is a platform maintenance table web page that list the information by ITM version and fixpack level, and then shows the corresponding z/OS PTFs.

https://www-304.ibm.com/support/docview.wss?uid=swg27008514

The following page shows information on fixpacks for each of the various OMEGAMONs, and the corresponding z/OS PTFs for each of these fixpacks.

https://www-304.ibm.com/support/docview.wss?uid=swg21290883#kc5420

Getting a handle on your System z environment

Something has just crashed. Everything seemingly was working fine until just recently. Now things are not so fine. What's the first question people usually ask? How about "what's changed"?

What's changed is often the million dollar question because, let's face it, in most shops System z is well managed and quite stable. When all around me is suddenly crashing down, the first thing I will often wonder is what changed and who made that change (and better yet, how to put things back).

IBM Tivoli is never at a loss for acronyms. One you may not have heard of before is TADDM. What's TADDM? It stands for Tivoli Application Dependency Discovery Manager. What TADDM does is it goes out and dynamically discovers resources and components in your environment, and map these resources. TADDM output can be used for a variety of functions, including as input to a CMDB (Configuration Management DB) and as input to build application views in TBSM. But, one of the more interesting features of TADDM is it's ability to track and highlight system changes. Did something change? TADDM can show you the differences.

TADDM has components for both distributed and z environments. On System z TADDM can go out and map at different levels of discovery, including looking at such things as z/OS Parmlib information.

While I would not suggest that TADDM is a tool for change control, TADDM can be a tool to help you get a handle on what's on your system, what's talking to what, and what if anything has changed.

If you want more information on TADDM, here is a link:

http://www-01.ibm.com/software/tivoli/products/taddm/

ITCAM For Transactions V7.3 provides agentless transaction tracking

IT Composite Application Monitoring (aka ITCAM) for Transactions V7.3 has been available for a few weeks. ITCAM for Transactions provides many interesting features for tracking and analyzing complex transactions, and offers insight into where those complex transactions may be having issues.

One of the more interesting new features of ITCAM for Transactions V7.3 is the so-called "agentless" transaction tracking. This feature captures network traffic and reports the interactions between servers and applications. It integrates into the existing user interface, combining the breadth of agentless tracking with the depth of agent based tracking.

Agentless tracking is built upon facilities developed for web response time analysis. Now, in addition to HTTP traffic, the tool is now capable of monitoring and analyzing all TCP traffic.

The example I show gives you an idea of how agentless tracking can gather information and dynamically map the activity.

New - IBM zEnterprise 114 announced

Last week IBM announced a new entry in the zEnterprise line, the z114.

IBM introduced the IBM zEnterprise 114 (z114) as the newest member of the zEnterprise family, bringing the unique value of hybrid computing to a much broader set of businesses. The z114 offers a smaller mainframe footprint and lower entry cost than the zEnterprise 196 (z196).

Here's a link to the announcement letter:

http://www-01.ibm.com/common/ssi/cgi-bin/ssialias?infotype=AN&subtype=CA&htmlfid=897/ENUS111-136&appname=USN

Upcoming webcast on OMEGAMON XE on z/OS

On July 28th there will be a free webcast covering the capabilities of OMEGAMON XE on z/OS.

In this session, you’ll come away with details on how OMEGAMON XE can help:
• Reduce the cost of administrating availability and performance on the mainframe by reducing complexity and using fewer staff resources
• Automate availability monitoring to implement best practices and compliance requirements
• Analyze data to detect and respond rapidly to availability and performance events
• Proactively handle events with automated closed-loop remediation that closes exposures
• Consolidate cloud availability management for zEnterprise

The speaker is Joe Winterton of IBM. Joe is the OMEGAMON XE on z/OS Release Manager. The webcast will happen on July 28th at 11 AM ET.

To sign up here is a URL:

www.ibm.com/software/systemz/telecon/jul28

NetView V6.1 and Canzlog

The new NetView release V6.1, went GA in May of this year. There were several interesting new features in NetView V6.1. I had mentioned an important one in an earlier blog, the enhanced TCP/IP trace function. Another feature that is generating a lot of interest is Canzlog. What is Canzlog? Canzlog is something of an acronym; Consolidated Audit, NetView, and z/OS Log (hence Canzlog).

Canzlog is based on a data-space that works with the NetView SSI to capture messages. Canzlog can hold millions of messages, after which it wraps. What Canzlog provides is the ability to capture messages from a variety of sources, and then provide a means to view and filter those messages quickly and efficiently.

I found a nice write up on Canzlog that goes through a description of how to set up and use the tool. Here's a link to the document:

http://www.ibm.com/developerworks/wikis/download/attachments/109642291/NetView+Canzlog.pdf?version=1

Overhead considerations when using the SOAP interface

I did a series of posts a few months ago on using the SOAP interface to issue commands, stop/start situations, and query monitoring data from the ITM infrastructure. The nice thing about the SOAP interface is that you can interface with ITM monitoring from a variety of sources, be it distributed sources or even IBM SA z/OS automation.

When doing things like querying information using SQL via the SOAP interface, be aware that the usual rules of thumb for effective/efficient SQL coding apply here, as well. For example, consider things like the number of SQL calls, the frequency of the calls, or the amount of data potentially retrieved by the calls. It's a typical monitoring rule of thumb, the more data you request, the more it will likely cost.

There is a nice write up that goes into these considerations. Here is a link to the document, it's interesting reading:

https://www-304.ibm.com/support/docview.wss?uid=swg21498390

All those storage mesages in RKLVLOG

So you are sifting through the RKLVLOG for an OMEGAMON agent (TEMA) address space or the TEMS address space on z/OS and you see all these storage messages in the log, such as the following:

KLVSD001 PRIMARY MAIN STORAGE INFORMATION:
KLVSD003 ALLOCATION DETAIL:
KLVSD004 SIZE(1-16) USE(27) TOTAL(521) ACCESSED(18864)
KLVSD004 SIZE(17-32) USE(93) TOTAL(102) ACCESSED(386)

It's normal for these tasks to generate these messages on an ongoing basis. Most of the time you won't need to worry about these messages.

So what do these messages mean, and do you need to do anything? In general, I don't recommend going into the RKANPARU members and playing with the storage settings unless directed to by IBM support. But every once in a while, it may help to know what these numbers are about. I have seen scenarios where storage creeps can occur, and adjustments may need to be made.

Here's a link to a nice description on what these messages mean:

https://www-304.ibm.com/support/docview.wss?mynp=OCSSZ8F3&mync=R&uid=swg21501180&myns=swgtiv

Upcoming webcast on managing security in the Cloud

Cloud computing is one of the hot IT buzzwords/phrases floating around currently. Cloud computing offers interesting capabilities, but like anything nowadays, security is an essential component that needs to be considered carefully.

"Consolidated Security Management for Mainframe Clouds" will take a look at the challenges of security in a mainframe Cloud environment. The premise is that better management of mainframe security simplifies complex IT environments, and allows the cost-effective consolidation of security management and compliance reporting for the enterprise.

If you are interested in learning more about Cloud computing, the security challenges in this paradigm, and what Tivoli is doing to manage security in the Cloud, take a look at this webcast.

The event is July 14th at 11AM ET. Here's a link to sign up:

http://www.ibm.com/software/os/systemz/webcast/jul14/

Upcoming webcast on solving network problems

The Share user group is sponsoring a webcast on July 7th on "Solving Application and Network Issues with OMEGAMON and NetView". The presenter for the session is Kirk Bean, the product manager for NetView and OMEGAMON Mainframe Networks.

Kirk is very knowledgeable and I think you will find this to be an interesting session.

The event is July 7th and 11 AM ET. Here's a link to register:

http://www.share.org/SHAREeLearning/Webcasts/tabid/700/Default.aspx

A unique alert

When you generate alerts you always want the alert to be unique and meaningful, and also have a solid means of notification. You may want the alert to be visible or have some other means of notification.

Here's a unique example of an alert. This was taken from a road sign on I-75 just outside of Cincinnati. Clearly someone had a sense of humor, but once you see the alert, you'll remember it.

Security and alerting on z/OS

Reading the papers lately, one thing that seems interesting is a spate of very sophisticated hacking attacks on various entities and companies. A new group called LutzSec has recently claimed to have hacked the CIA, PBS, SONY and who knows what other sites. And there are other groups out there doing what they do. One thing that seems apparent is these are very sophisticated attacks, done by very tech savvy individuals.

System z is blessed with very robust security capabilities driven by tools, such as RACF. You have a lot of power and flexibility to control access to System z. But, if someone is trying to do something nefarious, how quickly will you be notified? If a hacker attack is underway, wouldn't you want to know as soon as possible?

One interesting tool is Tivoli zSecure Alert. zSecure Alert comes with a pre-defined group of alerts, plus the ability to add additional user defined alerts, as needed. So you can alert on things like logon failures with powerful IDs, people attempting to access sensitive files, or people using system authorities that you may not want them to be using. Using zSecure Alert you can generate an alert in the form of an email, an SNMP trap, or a WTO to the z/OS console.

Or you could take that alert even further. Once the alert is on the console, how about having SA V3.3 send the alert message to the Tivoli Portal, fire a situation alert in the Portal, and have that situation forward the alert to OMNIBus, if desired.

The bottom line is zSecure Alert is an intriguing tool in that it can let you know right away if a potential attack is underway. You can then drive your notification in a number of ways. Stop them in their tracks before your company or government agency is the next one in the news (for all the wrong reasons).

Webcast tomorrow on Integrated Service Management on System z

Here's an interesting webcast happening tomorrow on Integrated Service Management for System z.

"Learn from System z customer experience: Cookbooks that can speed up deployment of integrated service management" In the session they will cover the new ISM for System z cookbooks with detailed information on how to implement ISM, and reference architectures. The details include prescriptive best practices that have been created, based on customer deployments, and are now available to speed your time to value of ISM solutions running on the mainframe.

The event will happen on June 23rd, at 11 AM ET. It's a free event, so the price is right. Here's a link to sign up:

http://www-01.ibm.com/software/os/systemz/telecon/jun23/index.html?S_TACT=101HX11W&S_CMP=twr

Upcoming System z technology virtual summit

This is a similar event to the System z summit I mentioned a few weeks ago, only this one is a virtual event, so you can attend from the comfort of your own office. This event will have four tracks: DB2 for z/OS technical update, Modern application development and deployment, CICS technical update, and Data Center operations and systems management update.

The virtual event will happen on June 29th. It's a free event, so the price is right. The agenda looks good, with good technical content on several fronts. Here's a link to sign up:

http://www.ibm.com/software/systemz/virtual/summit

System z jobs

System z jobs is a web site that connects IBM System z clients, partners and businesses with students and professionals seeking System z job opportunities.

The fun thing about System z is it keeps going strong. I began working in the MVS arena in the early 1980s. In 1992 I took a job with Candle Corporation. At the time some of my colleagues asked me "why take a job with a mainframe vendor, everyone knows the mainframe is going away"? Well, it's been almost 20 years later, and as Mark Twain may have said, reports of the mainframe's death have been "greatly exaggerated". System z continues to grow and evolve. I have the pleasure of working with many shops, large and small, that continue to invest heavily in the technology.

I've also had the pleasure at IBM to mentor younger colleagues about System z technology. What's interesting is that we are now seeing a new generation of technical talent coming into the System z realm. It will be enlightening to see where they take the platform in the future.

If you are interested in the System z jobs web site, here's the link:

http://systemzjobs.com/home/index.cfm?site_id=11677

New Version of ITCAM for Transactions

IT Composite Application Monitoring (aka ITCAM) Version 7.3 went GA in May. ITCAM is one of the most interesting and powerful tools in the Tivoli suite of solutions. ITCAM provides end-to-end transaction tracking and end user experience monitoring capabilities. ITCAM also provides capabilities for robotic monitoring and Internet service monitoring. By using ITCAM you can visualize and more easily isolate problems, leading to faster problem resolution. ITCAM is a very big solution, with quite a bit of technology and capability bundled into it. There are many ways you can apply this technology to monitor and track transaction activity.

ITCAM V7.3 adds new capabilities including:

• Agentless tracking that provides improvements in time-to-value

• Integrated view of agent based and agentless data through a single user interface on a single topology view

• Scalable monitoring roll outs that leads to improvement in consumability

• Enhanced transaction tracking that provides support for multiple domains including Microsoft (.Net), Distributed (Websphere, J2EE, etc.), and Mainframe

For more information on ITCAM for Transactions, here is a link:

http://www-01.ibm.com/software/tivoli/subscriptionandsupport/whats-new-itcam-transactions/?cm_sp=MTE18609

About the PDS (the Persistent Data Store)

Anyone who has done an install of any of the OMEGAMONs on z/OS has gone through the exercise in the ICAT installer of configuring the Persistent Data Store (PDS). Odds are if you've done it, you have executed the configuration steps multiple times (once for each TEMS and once for each agent TEMA task on each LPAR). This adds up to doing what seems to be the same task many times over and over. The question may be, is this really necessary?

If you plan on using the history functions of the Tivoli Enterprise Portal, displaying the data in the TEP, and optionally sending the history data to the Tivoli Data Warehouse (TDW), you will need this infrastructure in place to some extent. If you are not planning on using these functions, then you can get by without doing this work. One exception to this would be OMEGAMON XE For Storage, which will require the PDS for all its various features.

My recommendation is to go ahead and set up the PDS, at minimum, at the level of the TEMS. This gives you the option of collecting at least some history data in the TEP. Keep in mind though, that this may cause some confusion with users if they are trying to collect history at the level of the TEMA, but the PDS in the TEMA is not set up.

If you can, configure the PDS for each component, TEMS and TEMAs. Often I will scale down the default size allocations of the PDSs for the various components, just to save space (this stuff can add up across many systems and agents). But, I will go ahead and create the files per the steps in ICAT. This means that a user will be able to get the history to function from the TEP without having to worry about the underlying infrastructure being in place. It may not be optimal, but it will at least function.

With that being said, you may need re-create the files with more space if usage goes up. So if users are taking advantage of the history facility, it is a good idea to monitor usage, and make adjustments if usage increases.

What is tacmd?

tacmd is a command line tool used for ITM 6.x monitoring infrastructure management. There is a wide array of functions that you can use tacmd for. For example, you can use tacmd at the command line to edit agent configuration options, add a situation, create a situation, delete a situation, edit managed system lists, stop/start agents, and more. tacmd is a handy tool to have around.

The ITM Command Reference documents the capabilities of tacmd. Here's a link to the doc:

http://publib.boulder.ibm.com/infocenter/tivihelp/v15r1/index.jsp?topic=/com.ibm.itm.doc/itm_cmdref34.htm

An interesting subtlety of using the KILL command with OMEGAMON CICS

OMEGAMON CICS offers levels of KILL and KILL FORCE commands to kick a problem task out of CICS. This is a handy command that users need on occasion to deal with problem transactions. Users will sometimes ask me, when is it appropriate to use KILL versus KILL FORCE? I usually suggest starting with the least powerful command and progressively working your way up to KILL FORCE.

APAR OA35253 discusses an interesting scenario when using KILL versus KILL FORCE. While OMEGAMON does not require that a user issue a KILL before doing a KILL FORCE, there is apparently an occasional timing issue where if you try doing a KILL without first doing the KILL FORCE, the command may not work, and generate an error message.

Now you may see the following if a KILL FORCE is attempted without a prior KILL. The tool will process the command as a KILL, and a message will be displayed:

OC0550: FORCE OPTION REMOVED - MUST TRY KILL FIRST

Upcoming webcast on OMEGAMON historical data collection

On June 16th I will be doing a webcast entitled "Become More Proactive through Effective Historical Analysis". In this webcast I will be looking at each of the OMEGAMON products, and their historical data collection and analysis capabilities. I'll look at the various types of historical data available, and how it can be viewed and analyzed. I will focus on the specific historical data collection capabilities unique to each of the OMEGAMON products, and then look at historical analysis functions common to all OMEGAMON and other IBM Tivoli® solutions.

I'll get into the cost/benefit aspects of historical data collection, and try to help advise on a strategy for effective efficient historical data collection. Then I will look at examples of how you can use history to be more proactive from a monitoring perspective.

The event is June 16th at 11 AM Eastern Time.

Here is a link to the event:

http://www.ibm.com/software/systemz/telecon/jun16