For Back-Level IBM i (iSeries/AS400) Users-Risks Are Greater Than They Can Imagine
IBM 9406-170 User Encounters Unexpected Hurdles To Move To New Tech

For Back-Level IBM i (iSeries/AS400) Users-Risks Are Greater Than They Can Imagine IBM 9406-170 User Encounters Unexpected Hurdles To Move To New Tech

The purpose of this blog is to alert back-level IBM i users to the grave risks to recover should their server fail or need to migrate to new IBM POWER technology, especially those servers that are 17 years and older.?

Disclaimer: IBM i is an operating system. iSeries and AS400 are servers. I use these terms interchangeably to make it easier for folks to find this kind of information on the web.

In the last few months, a 9406-170 user reached out to my company to get new technology for accessibility to current IBM support.

While I can understand how many businesses can defer technology and support to improve the bottom line, it can be risky to delay investments to stay current and safe.

Let me provide an example.

9406-170 User At V5R1

I have a soft spot in my heart for this client of many years. I really like the people that work at this company.

Finally, after a decade of nudging, they finally agreed to upgrade to a new IBM POWER server.

While the migration from V5R1 to V7.3 should have been straight-forward, we encountered more challenges than we ever expected.

First Migration From V5R1 to V7.3 Fails

Many times we can take the custom RPG code from V5R1 and load it in a higher IBM i level, like V5R4 or V7.3, and we can get it to run. Generally, the custom software author can tweak the newly recompiled version to work with V7.3.

Well, when we attempted that approach, it failed.

So we determined to move the software from V5R1 to V5R3 on an interim 9406-520 server so we could upgrade to V6.1, and then to V7.3.

Interim Server Rebuilt From Two 9406-520 Servers With Missing And Failed Parts

We harvested parts from two 9406-520s to create one interim server at V5R3.

We discovered each server had failed or missing parts. These servers had not been used in so long it was not until we staged and tested them we discovered they needed repairs. The effort to make this 9406-520 functional took extra weeks to get the reliable missing parts (we had received some replacement parts that failed. Grr!).

3 QIC Tapes Failed To Restore

While restoring the V5R1 Save21 from QIC tape to the interim 9406-520, 3 tapes failed.

The senior IBM i engineer opened up the QIC tape to discover that a tape belt (a little rubber band) coordinates the two tape spindle rotation. Overtime, the rubber band disintegrated. Yikes!

Even though this client stocked up on new QIC tapes in 2015 for future backups, the tape belts that synchronize the reels (a rubber band!) had decayed. ?Because of the tape belt’s fragility, three QIC tapes failed during the save/restore process.

V5R3 No Longer Supported

After 2 more weeks delay because of the failed QIC tapes, we finally migrated from V5R1 to V5R3.

Next, we discovered missing IBM PTFs in our “version” of V5R3 to upgrade to V6.1.

We had to seek help from friends with more complete versions of IBM i OS software to access the missing PTFs to continue because IBM no longer supports V5R3 …. These PTFs were not available from IBM.

These friends also helped us to get to V6.1 on our way to V7.3.

What Should Have Been Completed In A Weekend Took 2 Months

The culmination of all of these unexpected issues extended our “weekend migration” to two months.

We are grateful our client was patient and understanding.

We DID get them to their new POWER server at IBM i V7.3!

The customer is thrilled. The POWER server is lighting fast and so much more reliable … WITH IBM SUPPORT! YAY!

Conclusion

The IBM i has a well-earned reputation for rock-solid reliability. My sense is that is why so many cost-sensitive businesses delay upgrades to currently supported IBM POWER servers.

With the unexpected issues we encountered, I believe if this 9406-170 failed we could have had serious issues to recover the system.

First, reliable parts for these older server are harder to come by.

Second, the skills of senior IBM field engineers is shrinking (retirement, death, etc.).

Third, their tape backups were marginal. Backup tapes do not last forever, even if you bought them brand new 8 years ago.

Back-Level IBM i Users Are At Great Risk

This story is not just about a single 9406-170 user.

My belief is most all back-level IBM i users are at great risk. If their servers fail, they may not be able to recover … ever.

And, in the last few months I hear more frequently from these cost-sensitive, back-level IBM users.

To Stay Safe:

1) Keep moving to current technology where you can get IBM support for hardware and software (especially software, so you have access to PTFs for critical exposure to cyber-attacks, like from Log4j).

2) Move off outdated tape media (QIC, 4MM, 8MM, etc.) and replace it with LTO, which is a far more reliable tape media with a longer shelf life (estimates up to 10 years).

3) Test your tape backups! You want to confidently know you can recover when needed and your tapes are not so fragile they decay in storage.

Need Help?

Call me at 714-593-0387 or email me at?[email protected]. Let us know how we can help!

To learn more about us, and view our customer testimonials, please visit our website:?www.Source-Data.com





S.E. Yarbrough Jr

AS400 / IBM i & Z0S / SW Engineer

3 年

Bob great post, upgrades are important no matter how great things are at present. Thanks for sharing.

要查看或添加评论,请登录

Bob Losey的更多文章

社区洞察

其他会员也浏览了