Friday, June 26, 2009

RealDolmen System z e-zine N° 13

The 13th issue of our RealDolmen System z Newsletter for our Belgian customers just got out yesterday. As we target the Belgian market, those who read Dutch or French can find this and previous newsletters on the RealDolmen Website : Dutch versions and French Versions.

We have two main articles : the first is focusing on Dynamic Infrastrusture but more specific on the role of System z. IBM has made a change of strategy in that it already gives us a view on its vision for the future System z while it is still under development. For the moment this concept is still called zFuture. Next we give an overview of Solid State Devices (SSDs) and how storage and our perception of it might change with the introduction of SSDs. Next to that there are also some hints, tips, links, an overview of recent announcements, EOS dates of software and operating systems and an agenda for Belgian events.

Enjoy the reading !

Dynamic Infrastructure Virtual Forum

This week there was the Worldwide Virtual IBM Forum on Dynamic Infrastructure. You can still access the forum via this link or if you've registered before you should've received a direct link to it.

Here's how the forum looks like :

Click on picture to enlarge in new window

You should go to 'Locations' (bottom left) and then choose for 'General Sessions' to have a replay of the presentations. You can also choose 'Resource Center' for more . . . resources on several related subjects.

Wednesday, June 24, 2009

Red Alert - Potential sysplex-wide outage due to XES signalling limit reached for CF Lock structures

Two red alerts in one week ! Read all about it ! Take a look over here.

Applications utilizing a Coupling Facility Lock structure may hang when the sequence number in a XES signal received for a connection to a lock structure reaches the sequence number limit x’FFFFFFFF.’ As a result the application may abend, transactions may begin to time out, or work queues may build to unacceptable levels. While any lock structure or serialized list structure may potentially encounter this issue, the problem has occurred with the DB2 IRLM lock structure and the VSAM RLS (IGWLOCK00) lock structure. Please see APAR OA27487 for further details.

From an operations view there are no XCF/XES messages or displays that surface the sequence number, therefore making it difficult to determine the problem or identify the system requiring recovery.

Recommended Actions:
Please install the PTFs for OA27487 and OA28470 to prevent the problem.

If you want to have en overview of all past Red Alerts, then take a look over here. You can also subscribe on the same page so you'll be notified by mail of any future Red Alert.

Tuesday, June 23, 2009

The Mainframe surviving and thriving in a turbulent world

I mentioned a CA mainframe survey a couple of posts before. Now it's out with an accompanying video by Marcel den Hartog.
I also already told you that the survey revealed that while on average 50% of all corporate critical data are on the mainframe, the cost on average is only 19% of the total IT-budget.
But there's more : there's also information on how in companies with a 'connected' mainframe (connected "within a distributed, web-enabled enterprise"), the mainframe even plays a more important role than in those that are not connected. And here I have to make the link to some of IBMs strategies which were presented during the last months. The concept of a new mainframe (zFuture) is also including more connectivity to other platforms. I'm sure I will come back to that in future posts.
Next to that there's also information on security and on how to find a solution for skills shortages.

Here's the video :

I'll let you discover the rest of this EMEA survey yourselves with the press release, the introduction page and the survey itself.

By the way, there recently was also a CA Survey predicting a growing use of Linux on the mainframe. Here are the press release and the survey : 'Usage and Plans for Mainframe Linux'.

Monday, June 22, 2009

Upcoming events

Holidays coming up ? Maybe, but those still wanting to grab some teleconferences on DB2, IMS, Tivoli Key Lifecycle manager and many more should take a look at the agenda. There are quite a lot of them, so I've also put them over here :

25/06/2009 :
Teleconference - Tivoli Key Lifecycle Manager for z/OS
at 17.00h Brussels Time


07/07/2009 :
Teleconference - Misconceptions and Old Wives Tales on DB2 Database Maintenance and Recovery
at 17.00h Brussels Time


14/07/2009 :
Teleconference - Get Smart IMS Applications with COBOL and Java Interoperability
at 17.00h Brussels Time


16/07/2009 :
Teleconference - Tivoli Provision Manager V7.1.1 for z/OS
at 17.00h Brussels Time


21/07/2009 :
Teleconference - DB2 9 for z/OS Utilities – Best Practices Update
at 17.00h Brussels Time


22/07/2009 :
Teleconference - Increase Business Agility through Innovative Tooling
at 17.00h Brussels Time


30/07/2009 :
Webcast - Enhancing Mainframe Security with Tivoli Security Management for z/OS
at 17.00h Brussels Time


11/09/2009 :
GSE Security Group Meeting on z/OS Security


23/09/2009 :
Joint GSE z/OS and GSE Storage Working Group Meeting
at IBM Forum Brussels


24/09/2009 :
GSE DB2 Working Group Meeting
at IBM Forum Brussels


5-8/10/2009 :
IDUG Europe in Rome


11/12/2009 :
GSE Security Group Meeting on General Security


Thursday, June 18, 2009

Red Alert - Potentially incorrect data for z/OS applications sending data using TCPIP ASYNCIO on z/OS 1.10

Here's a new Red Alert :

A logic error in z/OS 1.10 TCPIP can cause applications using ASYNCIO Writes to be driven twice. This can affect various applications and subsystems using z/OS TCPIP ASYNCIO on the sending side, such as Websphere MQ, JES and non-IBM products.


A logic error in TCPIP on z/OS 1.10 can cause applications using ASYNCIO Writes to be driven twice. The application data presented on the socket write operation is sent over the TCP connection twice, thus corrupting the data stream. This can affect various applications and subsystems using z/OS TCPIP ASYNCIO on the sending side, such as Websphere MQ, JES and non-IBM products. Please see APAR PK86842 for the details and symptoms that may result.

Recommended Actions:

z/OS 1.10 users should install the APAR fix for PK86842 that is currently available from IBM Support. z/OS releases prior to 1.10 are not affected.

If you want to have en overview of all past Red Alerts, then take a look over here. You can also subscribe on the same page so you'll be notified by mail of any future Red Alert.

Wednesday, June 17, 2009

CICS TS v2.3 End of Support next September

I'm just grabbing this from the CICS e-Newsletter I mentioned a couple of posts ago :
"IBM has announced the withdrawal of support for CICS Transaction Server V2.3, effective September 2009 - see details in the announcement letter here. (...)

When planning version-to-version migrations, don’t forget that CICS Interdependency Analyzer V2.2 now contains new queries to help to identify those applications most sensitive to change, helping target testing where it is most needed".
I guess the people involved already know they are going EOS with CICS TS V2.3, but I'm not sure every one knows this CICS Interdependency Analyzer. I, for one, did not. So I gathered some material about it :
But there's more : the CICS Interdependency Analyzer is one of the IBM CICS Tools for which you can download a free 60-day trial version. If you're not interested in this one, you might want to check out the other IBM CICS Tools at the same page.

Friday, June 12, 2009

IBM Lifecycle Extension for z/OS V1.8

Last year in August IBM announced the Lifecycle Extension Service Offering for z/OS 1.7. z/OS 1.7 seemed to be a very popular release (like z/OS 1.4 was too) and many people still had to make the upgrade to z/OS 1.8 or z/OS 1.9. It's a monthly fee-based offering which offers you an additonal 2 year maintenance after the EOS date. This 2-year period is a maximum : you can take a minimum of 6 months with the option to add additional months. I'd advise you to upgrade as soon as possible, because I wouldn't call it cheap.

Now we have the same announcement for z/OS 1.8 : "IBM Lifecycle Extension for z/OS V1.8 offers fee-based corrective service to users who have not completed their migration from z/OS V1.8 to either z/OS V1.9 or V1.10 (ZP09-0213)". The conditions are pretty much the same except that you can start with 3-month purchase. The offering does not surprise me, not only because it can generate some extra money for IBM, but it definitely has its use. Many customers cannot keep up the pace of an upgrade every two year. On the one hand the question is often raised why people can't stay longer on a certain release without having to pay extra money for it. Not every one needs the 'new' enhancements now. On the other hand staffs have been cut down in order to reduce cost, but the same work still has to be done. So, the offering can help you out, but it can leave a bad taste in some mouths.

Just a hint : this doesn't alter the z/OS coexistence policy. So if your not planning to upgrade in the next couple of months, be sure to order z/OS 1.10 before the end of September because it's the last release you can upgrade to from z/OS 1.8 within that policy and you'll have to pay for it if you order it later.

Addition : there's a FAQ too which you can find over here. It also mentions that as "a statement of direction, IBM intends to provide the IBM Lifecycle Extension for z/OS offering for future releases of z/OS" as well.

Tuesday, June 9, 2009

Short Takes

Here are some more short takes
  • CA, Datacom and zIIP
    Last week, CA announced that the newest release of CA Datacom now also makes use of the zIIP engine. I must admit I have the feeling I lost a bit track of how many products now use specialty engines. But I guess that's a good thing : more and more ISVs use them because every one realizes it's a great way to help reduce cost for the clients ànd at the same time keep the mainframe alive and kicking. The counter for CA seems to be at 14 products now. Still recently, they gave an overview over here.
  • zPrime
    Talking about saving money : a couple of weeks ago I received this mail from Neon Enterprise Software about zPrime. What is zPrime ? I don't know and they're not going to tell you straight away, so if you want to take a guess yourself, take a look at the zPrime Site. Here's the initial explanation in the mail : "zPrime is a radical new technology in mainframe computing. It cuts the cost of mainframe computing significantly and immediately. No new budgets, no complicated projects, no expensive consultants, no new equipment purchases. Want to know more? Go to and learn about zPrime -- and have some fun and maybe win some prizes at the same time". We'll just have to wait and see. Oh, you can also follow zPrime on Twitter !
  • When will I use Solid State Drives ?
    I know I'm a bit late with this. One very interesting presentation on the System z Conference some weeks ago was the one of Pat Artis of Performance Associates on SSD. According to Pat Artis the question is not whether we will switch to SSD but when. He expects that very soon every one will only use SSD drives for production data. Interested ? You can now find the slides and the video if this presentation on their site.

Thursday, June 4, 2009

IBM Dynamic Infrastructure Worldwide Virtual Forum

I've talked about Dynamic Infrastructure before. We had the announcement on April 28, 2009 and there was a promise of a Worldwide Virtual Forum where Dynamic Infrastracture would be discussed into much more detail. There's a specific page about it now where you can get all the details on where you can register for the 4-hour event. Be sure to click on your country because every country has another timing and language. The agenda :
Jim Stallings : A smarter planet needs a dynamic infrastructure
Erich Clementi : The emerging role of cloud computing in enterprise infrastructure
Steve Forbes : Global outlooks and the challenge of change
Richard Lechner : Building an energy-efficient dynamic infrastructure