Are You Ready for World IPv6 Day

June 8, 2011, is World IPv6 Day. As I wrote a few weeks ago, the Internet is full; although IPv6 has been around for some time now, it is not widely used.

World IPv6 Day is an event to help with the transition to IPv6. Tim Mullenbach, who works on the IBM i networking development team and has been the lead developer on IPv6 for several years, has written an excellent article on World IPv6 Day. In this article, Tim also reviews the IBM i IPv6 strategy and discusses the IBM i certifications for IPv6 enablement.

Within Tim's article is one important question – does your IBM i connect as a client to any of the participants host names in the IPv6 Day event? You should read Tim's article to determine if IPv6 Day could affect you.

 

 

COMMON, the Minneapolis Advantage

With the COMMON conference being held in Minneapolis this year, there were a few unique opportunities that made this conference a little special that I thought would be nice to share with the larger i community.

The temperatures were very chilly early this week, unusually cold for early May, with even some snow flurries in the air – definitely not the special part of the week. I'm sure that made some of the attendees from warmer climates wonder why anyone would want live in Minnesota anyway!? Everyone was appreciative of the skyway system in Minneapolis to avoid braving the cold May temperatures. (OK, maybe not everyone, but this native Minnesotan sure was!)

COMMON and IBM worked together to provide the ability for people from IBM Rochester to attend COMMON for one day. Because most of the IBM i development team in Rochester is unable to attend COMMON, this gave them the opportunity to see what the conference is all about, attend a few sessions, the Expo, and the evening events. It also gave them the chance to “get out” and speak to some of our clients. I talked to some of the Rochester developers who rarely get this kind of an opportunity, and the feedback I got was overwhelmingly positive. They heard about interesting businesses, small businesses, facets of what customers really like about IBM i and Rochester (our support!), and many other things.

Likewise, I know many COMMON attendees appreciated the ability to talk to members of the development team. It also gave IBM the ability to provide a wide range of expertise in the IBM Power Bar in the expo, as well as giving additional people the opportunity to present technical sessions.

COMMON also offered tours of the IBM Rochester facility. Since my day job at IBM is working in the Performance and Scalability Services Center, which used to be known as the Rochester Benchmark Center, I rode the bus down to Rochester with those signed up to take to tour so I could lead the tour of our test facility. The excitement of our clients to see the Rochester facility is wonderful. The IBM Rochester Wiki page has some interesting information about the architecture of the big blue buildings.

And finally, the call for presentations for the fall COMMON conference is now out. The following information was in the COMMON Daily on Tuesday and Wednesday:

The call for presentations for 2011 Fall Conference and Expo is now open. The conference will take place October 3-5 at the Hilton St. Petersburg Bayfront, St. Petersburg, Florida. The education team is looking for sessions and all day workshops in every course of study. We are particularly interested in sessions on New Technology and RPG.

Current speakers should visit their speaker portal to submit sessions.  But any member can submit a new session by following this link.

The COMMON Education team is always looking for new content to add to the conference agenda and for new speakers. I'd also like to reiterate what I wrote last week about volunteers – there are several open positions on the Education team, and if you'd like to get involved and help to shape the agenda for future conferences, consider volunteering.


Twitter: #ibmi @DawnMayiCan #COMMONUG, #ibmiuser

Automate Extra IPL for PTF Install

This week's blog was written by Ellie Streifel.  Ellie works in the IBM i development lab on the Software Service team and has extensive knowledge on the IBM i PTF process.  Ellie worked on the Technology Refresh support as well as the function she describes in this blog article. Ellie will be at COMMON in Minneapolis (along with co-worker Dennis Martin) presenting the session, “What is an IBM i Technology Refresh – and Why do I Care?” 

Do you get annoyed when it takes two IPLs to install a PTF package? If so, we’ve got a great, new function for you in IBM i 7.1. When you’re installing PTFs, there are two conditions where we prompt you to perform an IPL to apply some of the PTFs, require you to restart the PTF install after the first IPL, and then make you perform yet another IPL to apply the delayed PTFs:

1.    When installing a cumulative PTF package that contains special handling pre-apply PTFs.

2.    When installing a technology refresh PTF at the same time as a technology refresh requisite PTF. This can occur when installing the HIPER PTF group, the technology refresh PTF group, or a cumulative PTF package.

We can’t eliminate the extra IPL, but this new function can automate it for you. If an extra IPL is required, we’ll save off of your PTF install parameters and use them during the next IPL. Instead of seeing “Confirm IPL for Technology Refresh or Special Handling PTFs” panel, you’ll see a new message CPF362E: “IPL required to complete PTF install processing”.  Actually, if you select Automatic IPL=Y on the “Install Options for PTFs” panel, you won’t see any messages or panels, we’ll just power down. On the next normal IPL, we’ll do your second “GO PTF” during the “PTF Processing” IPL step in the SCPF job, and then automatically do a second IPL of the partition. So when the system IPLs the second time up to sign-on, your PTFs are all activated and ready to go.

Your total PTF install time will be shorter since we won’t start any of the system jobs during the first IPL when we restart the partition. However, you’ll have a somewhat longer IPL time since we’re doing the work you previously did interactively – the second GO PTF to set all PTFs for delayed apply. If you’re wondering why the IPL is taking so long and what it’s doing, you can always bring up the console. The “Operating System IPL in Progress” panel is shown in the following PDF link with the “PTF Processing” IPL step active (Download ICan 3.15.11 Figure1).  Previously, we only displayed “Applying PTFs” for the Activity in this step.  But now you will also see “Loading PTFs” or “Setting IPL Action for PTFs” if the previous PTF install was incomplete. When all PTFs have been set for delayed apply, you’ll see “IPL requested by PTF processing” status message at the bottom of the screen and then the partition will be restarted again to apply the delayed LIC PTFs.  The next time we reach the “PTF Processing” IPL step, we’ll display the usual “Applying PTFs” and the IPL will continue.

There are a few things to keep in mind if you want to take advantage of this new function:

1.    You must have PTF SI42445 (PTF management code) temporarily applied before you perform your PTF install. Note: if you have HIPER PTF group SF99709 installed with level 26 or higher, you have PTF SI42445 applied.
2.    If an IPL is required for a technology refresh PTF, the new function only supports installing from a virtual optical device or *SERVICE (PTFs downloaded electronically to save files).  If you’re installing from a physical optical device, you’ll still need to perform the extra IPL and second GO PTF manually. If you received your PTFs on physical DVDs, just create yourself an image catalog from the DVDs and use the new support.
3.    If you’re doing your PTF install and you’re not yet ready to IPL the partition, remember, we’ll use the same PTF install parameters the next time you IPL. That includes the image catalog and virtual optical device you were using … so don’t delete the image catalog until PTFs are active.
4.    If you’re installing PTFs from a shared virtual optical device or a network optical device, we’ll access the remote device from the SCPF job on the next IPL without bringing up TCP/IP. Make sure the server partition and remote optical device are available when you IPL the client partition.  Also, PTF SI42302 must be temporarily applied to the server partition that contains the image catalog with the PTFs you are installing. PTF SI42302 contains the required image catalog and optical code to support the new “Automatic-IPL” function.
5.    Since this new support was delivered via a PTF, the new messages seen from GO PTF and the “Operating System IPL in Progress” panel are English only. They will not be translated until the next IBM i release.

When a new Technology Refresh is announced, a corresponding Licensed Internal Code Resave is also released. If you want to avoid, rather than automate, the extra required IPL, install the corresponding LIC Resave. The latest LIC Resave for 7.1 has the latest technology refresh PTF already permanently applied.  For more information on Technology Refresh, see the IBM i Technology Refresh website.

Name that Thread

Did you know in the 7.1 release, you can now name the secondary threads you create in a process?  Prior to 7.1, the system assigned a thread ID and there was no interface to specify a thread name.  However, thread IDs can be difficult to work with, particularly when doing problem determination, because the IDs provide no indication of what that thread is for. You’d have to go in through the work management interfaces to display each thread and examine its call stack to have some idea of what that thread may be doing.

Likewise, the synchronization mechanisms commonly used in multi-threaded applications--mutexes,  condition variables, and read/write locks--couldn't be managed by names prior to 7.1; name support was also added for these in the 7.1 release.

Of course, the documentation in the IBM i Information Center doesn't exactly tell you this; you must do some review of application programming interfaces to find out about this little enhancement.

The work management “Work with Threads” interface that displays thread information (option 20 when using WRKJOB on the green screen) was updated to display the thread name. The thread name isn't displayed through the IBM Systems Director Navigator Web console or in System i Navigator. Also, very few of the IBM-supplied multi-threaded processes have taken advantage of using the thread naming capability; probably because these interfaces are not portable.

You can set the name of the target thread with the pthread_setname_np() function.

You can retrieve the name of a thread with the pthread_getname_np() function.

Likewise, there are setname and getname interfaces for mutexes, condition variables, and read/write locks.
pthread_mutex_setname_np()
pthread_mutex_getname_np()
pthread_cond_setname_np()
pthread_cond_getname_np()
pthread_rwlock_setname_np()
pthread_rwlock,getname_np()

The Internet is Full - i is Ready

A few weeks ago, I discovered the Internet was full. A Google search will find several articles on the topic.  I found it somewhat entertaining to think of the Internet as full, because it's really far from full; content will forever be created and available at our fingertips. But, in the context of these articles, the Internet is full, not because it has no more capacity for content, but because it has no more capacity for additional users. Of course, all of us technical people know the real story …

We in the IBM i development community have known for years that the IPv4 address space was going to be exhausted. Prior estimates used to say 2014, then it was 2012, but here we are in 2011 and it has happened. Of course, i is prepared with support for IPv6.

IP Version 6 (IPv6) work started on i way back with the V5R2 release (V5R2 was available starting in 2002). There were many restrictions with the IPv6 support for the first few releases it was available on i--it was more of a development environment so that if you had your own sockets application, you could enable and test it for IPv6 support. None of the IBM i supplied TCP applications, such as FTP, Telnet, etc., had IPv6 support at that time.

In the 5.4 release, IPv6 support became consistent with the IPv4 support, with the initial restrictions removed.  However, TCP application support for IPv6 was still lagging. While the TCP stack supported IPv6, the applications provided by IBM didn’t.

Finally, in the 6.1 release, many of the TCP applications were enhanced with IPv6 support, and in 7.1 that work is essentially complete--nearly all TCP applications that IBM provides are fully IPv6 enabled.

The 7.1 release provides the richest IPv6 support and is the first i release to be certified as meeting the current U.S. government IPv6 requirements. When you need to move to IPv6, 7.1 is the release that IBM recommends you use.