When MPE/iX systems, both virtual and physical, see their clocks tick over tonight at midnight, it will be a significant date. The end of Dec. 31 puts MPE/iX, as crafted by its creators — into single digits for years remaining. Nine is tomorrow's number.

Whether that's nine years until end of life depends on your IT plans. If like more than a few managers you're retiring clean -- with configurations in place to survive into 2028 — the nine years will show you're prepared. You've made your changes to work around the loss of accurate MPE/iX date keeping. At least one vendor is taking orders for this service.

Others, meanwhile, are doing the work and leaving the credit to others. Stromasys has a lot at stake in the 3000 market to make 2028 a year of smooth pavement. We've gotten word they're ready with a software solution to carry MPE/iX beyond HP's wildest visions.

For the IT manager who's retiring without a 2028 plan — and leaving Dec. 31, 2027 as a shutdown date — tomorrow is the start of the final nine years for that HP 3000. It goes without saying these managers have no current interest in the Charon virtualizer for HP's MPE/iX iron.

Everything ends sometime. 2018 wraps up this evening. Lau Tao wrote in another century, "New beginnings are often disguised as painful endings." May your year to come be a new beginning without such pain. We'll see you in a future where options are still emerging for a suprising decade-plus to come. Some 3000 managers will be joining the march toward a Double-Digit Future for MPE/iX.

By Gilles Schipper

While working with a customer on the design and implementation of disaster recovery (DR) plan for a large HP 3000 system, it became apparent the implementation had room for improvement.

In this specific example, the customer had a production N-Class HP 3000 and a backup HP 3000 Series 969 system in a location several hundred miles from the primary.

The process of implementing the DR was completed entirely from a remote location — thanks to VPNs and an HP Secure Web Console on the 969. One of the most labor-intensive aspects of the DR exercise was to rebuild the IO configuration of the DR machine (the 969) from the full backup tape of the production N-Class machine, which included an integrated system load tape (SLT) as part of the backup.

The ability to integrate the SLT on the same tape as the full backup is very convenient. It results in a simplified recovery procedure as well as the assurance that the SLT to be used will be as current as possible.

When rebuilding a system from scratch from a SLT/Backup tape, if the target system differs in architecture from the source system, it is usually necessary to modify all the device paths and device configuration specifications with SYSGEN and then rebooting the system in order to even be able to utilize the tape drive of the target system to restore any files at all.

(This would be apart from the files restored during the INSTALL process — which does not require proper configuration of any IO component at all).

Some would argue that this system re-configuration needs to be completed only once, since any future system rebuilds would require only a “data refresh” rather than a complete system re-INSTALL.

I say that this would be true only in very stable system environments where IO configurations — including network printer configurations — are static and where TurboIMAGE transaction logging is not utilized. Otherwise there could be unpleasant results and complications from using stale configurations in a real disaster recovery situation. In any case, there really is no reason to take any chances,

Earlier this year Jeff Kubler of Kubler Consulting was trying to label the status of MPE/iX security. The distinction between hardware and software is noteworthy. Whatever security the 3000s had confers onto the virtualized 3000s running under the Charon emulator from Stromasys.

Kubler built a list of the known conditions and advantages

• Unknown operating system
• Password protected
• Must know how to address it with HELLO
• Must know or guess the user
• Could have additional security like VEsoft strenghtening the additional login string
• Security on the account, user and group level could keep those who even know a login from getting anything important 
• No visiting websites while using an HP 3000 application

When Alan Yeo of ScreenJet said the 3000 security is weak ("if you have locked the doors, then it will stop someone who just tries the door handle"), Pro 3K's Mark Ranft wanted to disagree.

The correct description is Security through Obscurity. If your HP 3000 has VESOFT's Security 3000 installed, and it is properly configured with two factor authentication, I don't know if anyone, without physical access to the machine, or access to unencrypted backups media, that could break in.

Where the HP 3000 falls short is in encryption of data that is in transit between the user and the system.  For this, I recommend you turn to MiniSoft Secure 92 for terminal access.

And unfortunately, if you host a website on the HP 3000, I have to admit the HP WebWise MPE/iX Secure Web Server is not TLS 1.2 capable. This would be a showstopper for PCI certification. But this is only a big deal if you accept credit card or other protected information via the website.

Finally, depending on your location or customer base, you may also need to worry about GDPR.

That two-factor feature might not be fully available under MPE/iX, depending on your definition of 2FA.

Eleven years ago we wished for nine things that would help 3000 users in the years to come. At the close of 2007 there was no virtual HP 3000 product like Charon. We didn't even allow ourselves to wish for such a thing.

But here on the last office day before Christmas, it's fun to review our holiday wish list. Let's see what we got and what HP withheld until it was too late for the vendor to supply what the community requested.

We've heard these desires from HP 3000 customers, consultants and vendors. Some of the wishes might be like the Red Ryder BB-Gun that's at the center of the holiday epic A Christmas Story. As in, "You don't want that, you'll put your eye out." If you're unfamiliar with the movie, the line means "I don't want you to have that, because I worry what you will hurt once you get it."

1. Unleashing the full horsepower of A-Class and N-Class 3000 hardware
2. Just unleashing the power of the A-Class 3000s (since every one of the models operates at a quarter of its possible speed)
3. Well, then at least unleash the N-Class systems' full clock speeds
4. HP's requirements to license a company for MPE/iX source code use
5. A way to use more than 16GB of memory on a 3000
6. A 3000 network link just one-tenth as fast as the new 10Gbit Ethernet
7. A water-cooled HP 3000 cluster, just like IBM used to make
8. A guaranteed ending date of HP's 3000 support for MPE/iX
9. Freedom to re-license your own copy of MPE/iX during a sale of an 3000

HP finally supplied Numbers 4 and 8. The first created the Source Code Seven, vendors who hold licenses that let them create workarounds and custom patches for MPE/iX issues. Number 8 arrived during the following year. It can be argued HP didn't end all of its MPE/iX support for several years beyond that official Dec. 31, 2010 date.

Some of the more inventive indie support companies have devised ways to use 32 GB of memory for 3000s, too. Ask yours about Number 5.

The last two items seem like real BB-Guns. But they have a chance of helping the community see the 3000 future more clearly, instead of putting its eye out.

A guaranteed ending date for HP's 3000 support is something both homesteaders and migration experts desire. By moving the finish line twice already, HP has kept customers from finishing migrations, or even starting them, according to migration partners.

What's more, the "we're not sure when support is really done" message keeps the 3000's service and support aftermarket in limbo. Customers tell us that they will be using their HP 3000 systems until their business demands they migrate away. HP plans to change its business practices someday for the HP 3000. But nobody knows for certain what day that will be.

That brings us to No. 9, the freedom to re-license your own MPE/iX. HP development on this software ends in one year. That's the end of changes to the operating environment, a genuine Freeze Line for MPE/iX. HP should be able to compete on a level field with the rest of the community. HP Services seems to need those special 3000 licenses.

Number 10? A wish for a long life and continued interest in MPE/iX from the HP 3000 gurus of the community. Someone can bring some these gifts after there's no one inside HP to cares about the 3000 community.

I want to provide an option for rebooting an HP 3000 remotely using LDEV 21. How do I do it?  Can I using a modem and landline or an IP address to get to LDEV 21?

Gary Stephens replies

Yes, you can use a modem on the remote support port set to auto answer. This will definitely work. It was more about controlling access to the console remotely. I recall one site that had a modem with a remote call-back to a known inbound number that was effective. Upon answer you were prompted for a U and P that had a dial-back associated with it. Ultimately it's all down to risk and your appetite for change.

Billy Brewer adds

You could hook up a cheap laptop connected to the console port (with a USB to serial converter). Then use TeamViewer or any shared desktop utility you prefer.

Tracy Johnson reports

An older PC with one NIC (you can remote into) and a serial port will also do. These days we just have a PC as the console via a terminal emulator (Minisoft, Reflection, or QCTerm). We use two NICs and just remote into the PC and open a window from there to do our remote reboots.

Mark Ranft notes

All the newer A- and N-Class systems have IP configurable remote console. I assume yours is an older system.

The PC options are excellent. If you have a DTC, you can set up back-to-back DTC switching. You configure a host (outbound) port on the DTC and you choose it by IP and TCP port ID.

The Distributed Terminal Controller was a networking device with intelligence that stood between an HP 3000 and a peripheral. We use the past tense to describe the DTC usage for many of the homesteading 3000 sites. In some places, DTCs continue to let 3000s shake hands with other devices.

At TE Connectivity in Hampton Roads, Va. the box works between an N-Class 3000 (the ultimate generation) and an impact printer (of considerably older peerage). Al Nizzardini makes the pair work for the company that employs 3000s across the globe, from North America to China.

"Our DTC 48 with 3-pin ports died on us," Nizzardini said. "We have an impact printer connected to the 48, the only thing that is hanging off that DTC." At first the solution to the blocked connection was to use an even older controller, the DTC16 with modem ports. That would've involved shorting out pins on the DTC 16.

Nizzardini asked and a few veterans answered. Francois Desrochers said Nizzardini would need pins 2, 3 and 7 (send, receive, ground). "You may have to short out 5 and 20," he added. Another combination from Gary Robillard suggested connecting 4 and 5 together and 6, 8, and 20 together. "We always had 2 and 3 crossed—2 to 3 and 3 to 2," he said.

It's been 20 years since HP last released a DTC, something that's still useful for older peripherals. The intel to keep one connected to the latest 3000s is still available in the 3000 community. Old doesn't mean dead when someone remembers the essentials. Nizzardini solved his problem without shorting out pins, just by locating another working DTC 48. MANMAN drives the workflow at TE Connectivity, but the real driver is pros like Nizzardini, helping one another remember.

Editor's Note: Initial HP 3000 hardware networking can be like a trip down a Yellow Brick Road. Here's a primer for the administrator who's wondering if that HP 3000 can link to a network

By Curtis Larsen

Auntie MAU! Auntie MAU! A Twisted Pair! A Twisted Pair!

Once upon a time networks were as flat as the Kansas prairie, and computers on them were a lot like early prairie farmsteads: few and far between, pretty much speaking to each other only when they had to. (“Business looks good again this year.” “Yep.”) Most systems still used dumb terminals, and when speaking to anything outside the LAN, system-to-system modem connections were the way to do it.

A tornado named the Internet suddenly appeared in this landscape. It uprooted established standards and practices, swept aside protocols and speed limitations, and took us into a Technicolor networking landscape very different than what was there before.

Toto, I get the feeling our packets aren’t in Kansas anymore

Smaller companies were tossed before the tornado to eventually land and quickly begin growing again in the new environment. Large companies like IBM, HP, Digital, and Microsoft, who were rooted and established in their own proprietary standards (it sounds like an oxymoron, but it’s true) survived by generally ignoring the howling winds. Eventually, munchkin-like, they all came out to see what the general fuss was about, and found that a house-sized chunk of change (pun intended) had landed.

Networking, and the TCP/IP protocol had truly arrived in style, bringing strange new applications and markets. Serial connections and proprietary networking (“What do you mean we don’t need SNA to connect to the Wichita office anymore?”) gave way to a new kid on the block. And her little dog, too.

Follow the Yellow-Colored-Cable-and-Labeled-at-Both-Ends Road!

So then the HP 3000 managers found themselves sitting in a strange new networking land of strange new networking things. And for some of us, trying to understand the whole of it all — especially in relation to “legacy” system like the HP e3000 — was a little daunting. What are all these networking black boxes we plug the system into, and what do they all do? How can they make life better? (How can they make life worse?) If you’re not sure (or just plain curious) read on.

Ten years ago this week the 3000 community was in a state of anticipation about MPE/iX. HP had an offer it was preparing that would give select vendors the right to use the operating system code. The vendors would have a reference-use-only license agreement for MPE/iX. No one knew whether the source would have any value, said Adager CEO Rene Woc.

Adager, the company whose 3000 products are so omnipresent they held a spot on the Hewlett-Packard corporate price list, believed there was potential for independent support and development vendors. What was far less certain was how far HP would let source go to solve problems for the 3000 community.

"Source code is important whenever these kinds of [vendors] have support from HP, which most of them do," he said in that month of 2008. But HP engineers can look at source, just as third parties will do, "and the answers won't come instantaneously. In the meantime, you have to get your business back on track, and I think that's what the customer is eventually interested in. It will be nice to have that additional [source code] resource — especially in the sense that it will not be lost to the community."

There was a chance that HP's source licensing terms would be too restrictive, "to the point where you say that you are better off not knowing, because then we're free to use all the methods we've worked with while we didn't have source." After getting a license to source, Woc added, "you might have to prove that you got your knowledge through a difference source than HP's source code. We will see."

That sort of proof has never been required. Not in a public display, at least. Source code, held by vendors such as Pivital Solutions and others, has been a useful component in workarounds and fixes. HP never gave the community the right to modify MPE/iX. This turned out to be a good thing, as it kept the 3000s stable and made support a manageable business for application vendors.

There was also the wisdom that the resource of HP's code would have to prove itself. At least it held a chance for rescue and repair.

The source code "is probably a security blanket," Woc said in 2008. "In that respect, it's good that it will be available, that they're starting to offer some things. We'll have to see what kind of conditions HP will offer in their license agreements." 

Having source access though a license did not automatically make license holders better providers of products and services, he added. "You cannot assume, even with good source code readers, that the solutions will pop up," he said. "A lot of the problems we see these days are due to interactions between products. So the benefit for the customer would be based more on the troubleshooting skills that an organization can provide."

"The basic resources [of source] won't make things better by themselves," Woc said. "It's a matter of troubleshooting." 

Almost two years ago, an expert in HP's 3000 systems was working to use solid state disks (SSD) with the computer. John Zoltak was trying to link the server to microSD cards late in 2016. He checked in with us this week to report success on the project.

SSD on 3000 hardware from HP has been a dream for several decades. Imperial Computer had a solid state unit early in the 1990s that held a promise of faster IO transfer on MPE/iX. The cost was astounding compared to moving media and the capacity was a fraction of spinning disk drives'. Much later, SSD has become something of a desktop standard and is an active choice in enterprise servers, too.

The MPE/iX hardware from HP -- to us, something called an HP 3000 -- wanted to play from SSD, too. In his prior report, Zoltak was trying to copy one 917LX disk to a new disk on the server's SCSI bus. A 4GB drive is standard on a 917, so just about any microSD card would match that storage. Now there's a V6 edition of SCSI2SD, a combination of hardware and software that delivers SD storage to HP's 3000 iron.

The combination now works beautifully, said Zoltak, who's working at Fives North American Combustion in Cleveland, Ohio. "You want the V6 boards," he said. "The V5's are much slower. The V6 takes a full size SSD card and up to 128GB has been tested." Michael McMaster, the inventor based in Australia, has engineered the latest version of his product "as a complete redesign for the V6 boards, which use a completely different microcontroller." The device is for sale online at Intertial Computing. Today's price is $105 including 16GB of microSD.

The product employs a SCSI-2 Narrow 8-bit 50-pin connector. It does SCSI FAST10 synchronous transfers at 10MB/second. Zoltak is reaching way back into the HP 3000 hardware closet to test. He's attached the SCSI2SD to a Series 917.

"I have the board sitting on top of the system with a cable around to the back on the same SCSI as the 917's DAT and DLT drives. I did a reconfigure and a restore to the SSD. Seems to be fairly quick. While restore was running I used HP Glance and saw that the disk was doing about 65-70 IO's per second. This is not as fast as the Nike array it came off of, but then it was on a differential wide SCSI."

The bigger benefit is that the HP MPE/iX iron can rely on SSD instead of moving media. Disks are among the leading culprits in HP's 3000 failures in 2018. Tape is a close second. Storing and moving bits gets complicated while using the hardware that HP certified for storage with 3000s than a decade ago.

Newer storage reduces the risk of homesteading. This is one of the benefits of using a virtualized 3000, too.

NewsWire Classic

By Jeff Kubler

You need to get management support for your efforts to keep your systems performing at their best. Memory and disk are two components of your performance picture under MPE/iX. Main Memory is the scratch pad for all the work that the CPU performs. Every item of data that the CPU needs to perform calculations on or updating to must be brought into Main Memory.

CPU used to manage Main Memory: The CPU must manage memory. It must cycle through the memory pages, marking some as Overlay Candidates (this means that new data from disk may be placed here), noting that some are in continued use, and swapping others out to virtual or what is called transient storage. Swapping to disk occurs when data is in continued use but a higher priority process needs room for its data. To accommodate this higher priority process and its need for memory space, the Memory Manager will swap the memory for the lower priority process out to disk. The more activity the Memory Manager performs, the more CPU it takes to do this. Therefore it is the percentage of CPU used to manage memory that we use as a measurement.

Page Faults per Second: A Page Fault occurs each time a memory object is not found in memory. The threshold for the number of Page Faults per second that can be incurred before a memory problem is indicated varies with the size and the power of the CPU. Larger machines can handle more Page Faults per second while a smaller box will encounter problems with far fewer.

An exceptional number of Page Faults should never be used as the sole indicator of memory problems but when observed should be tested with the memory manager percentage. If both agree, you have a memory shortage. There are some strange things that I have observed with Page Faults, so it does not stand alone as an indicator of memory shortage.

The number of Page Faults per second and the amount of CPU needed to manage Memory are always evaluated in conjunction with each other. That is to say the high Page Fault Rate will not be considered a problem if the Memory Manager Percentage is not above 4 percent.

In a webinar this week, Stromasys made its case for how shutting down HP's 3000 hardware can reduce an IT budget. Using data from Gartner analysts and other sources, the company estimates that downtime costs companies $1 million per year on average. Any alternative to 15- to 25-year-old servers is a good shot at making the future more stable.

There still hasn't been a computer system built that will never fail. Hot-swap backups with automatic server failovers were never a big part of the 3000 datacenter experience. If you had to handicap which server was a likely failure candidate, HP's MPE/iX hardware would give you short odds of failure. In this case, short is not a good measure.

One million per year in losses is a big enough number to get the attention of a corporation's C-level. It's the same number, coincidentally, that Stromasys used this week to describe the costs of migrating MPE/iX apps. The text circled in the slide above "implies investments of $1 million+" for migrations.

These millions, lost through downtime or surrendered in datacenter budget, are averages. Smaller 3000 customers may not approach the $1 million in yearly lost revenues. Migration costs track closer to that number, but they're a one-time hit. The alternative is Charon, of course. During the webinar we learned that an additional HP market is coming online to use Charon. HP's Unix PA-RISC servers will be the latest Stromasys virtualization segment, according to Dave Campbell.

An HP 9000 HP-UX virtualization product is in development. In that kind of design, a single Intel server with enough computing power (concurrent threads) could host both HP 3000 and HP 9000 virtualizations. HP had the same objective almost 20 years ago for its largest enterprise platforms.

Early in 1999 HP's Harry Sterling spoke at an all-day user meeting in the UK hosted by Riva Systems. Sterling, who'd retire before the end of that year, said a multi-OS server was within HP's vision for the 3000 and 9000 customers.

Sterling’s mentioned the possibility of running MPE, NT and Unix concurrently on the HP 3000 "sometime in the future." There was even the possibility of a “hot-swap” version of MPE alongside the production system. John Dunlop reported for us at the time.

The passing mention indicated that separate processors in one box would be able to run different operating systems. Sterling did suggest that a hot-swap version of MPE might be a valid use, so that there would be some redundancy with the live operating system.

This seemed to lead to the subject of more uptime. From these comments, it’s possible that HP is looking at allowing online changes to a hot-swap system and then just switching it over to achieve the so-called “magic weekend.” This is a system upgrade that occurs seamlessly and transparently to both the users and management.

That would be a dream not realized. Hot-swap didn't make it any further into the customer base than architect discussions. Sterling noted that in 1997 customers expressed concern about the future of the 3000. To counter that feeling and give the customers more confidence, he outlined in 1999 a five-year roadmap for the 3000.

Marketing was on board as well in that year that led to Y2K. It would take another 13 years before a multiple OS host for MPE/iX would emerge.