Forgot your password?
Close
typodupeerror
HP Unix Operating Systems

Workstation Owner Sadly Marks the End-of-Life for HP-UX (osnews.com) 152

Posted by EditorDavid from the GNU's-not-Unix dept.
Wednesday marked the end of support for the last and final version of HP-UX, writes OSNews.

They call it "the end of another vestige of the heyday of the commercial UNIX variants, a reign ended by cheap x86 hardware and the increasing popularisation of Linux." I have two HP-UX 11i v1 PA-RISC workstations, one of them being my pride and joy: an HP c8000, the last and fastest PA-RISC workstation HP ever made, back in 2005. It's a behemoth of a machine with two dual-core PA-8900 processors running at 1Ghz, 8 GB of RAM, a FireGL X3 graphics card, and a few other fun upgrades like an internal LTO3 tape drive that I use for keeping a bootable recovery backup of the entire system. It runs HP-UX 11i v1, fully updated and patched as best one can do considering how many patches have either vanished from the web or have never "leaked" from HPE (most patches from 2009 onwards are not available anywhere without an expensive enterprise support contract)...

Over the past few years, I've been trying to get into contact with HPE about the state of HP-UX' patches, software, and drivers, which are slowly but surely disappearing from the web. A decent chunk is archived on various websites, but a lot of it isn't, which is a real shame. Most patches from 2009 onwards are unavailable, various software packages and programs for HP-UX are lost to time, HP-UX installation discs and ISOs later than 2006-2009 are not available anywhere, and everything that is available is only available via non-sanctioned means, if you know what I mean.

Sadly, I never managed to get into contact with anyone at HPE, and my concerns about HP-UX preservation seem to have fallen on deaf ears. With the end-of-life date now here, I'm deeply concerned even more will go missing, and the odds of making the already missing stuff available are only decreasing. I've come to accept that very few people seem to hold any love for or special attachment to HP-UX, and that very few people care as much about its preservation as I do. HP-UX doesn't carry the movie star status of IRIX, nor the benefits of being available as both open source and on commodity hardware as Solaris, so far fewer people have any experience with it or have developed a fondness for it.
As the clocks chimed midnight on New Year's Eve, he advised everyone to "spare a thought for the UNIX everyone forgot still exists."
This discussion has been archived. No new comments can be posted.

Workstation Owner Sadly Marks the End-of-Life for HP-UX More

Workstation Owner Sadly Marks the End-of-Life for HP-UX

Comments Filter:

  • 20 year is forever in tech years. (Score:3)

    by CommunityMember ( 6662188 ) on Monday January 05, 2026 @01:52AM (#65902375)
    HP has a long history of supporting their hardware/software for decades after last sale. 20 years is certainly impressive. But, as the saying goes, all good things come to an end.

    • Re: (Score:2)

      by quenda ( 644621 )

      But, as the saying goes, all good things come to an end.

      Only MS DOS^WWindows is immortal.

    • HP has a long history of supporting their hardware/software for decades after last sale. 20 years is certainly impressive. But, as the saying goes, all good things come to an end.

      If 20 years is considered 'forever' in tech years, IPv6 is pulling a whole-ass Benjamin Button.

      • IPv6 was turned on in 2013. That is when the count should start - not in the year the first RFC was defined. So it will hit its 12th year on June 6th

        Since then, it has gained 49% of the market. This year, it should cross the 50% mark

        • Re: (Score:2)

          by weirdow ( 9298 )
          IPv6 was already in use in 2004. not much to do with it yet but it was still available and usable.

          • It couldn't have been. IPv6 backbones didn't exist, so any IPv6 traffic had to be tunneled in IPv4 packets. On June 6th 2012 - IPv6 day, they turned on all IPv6 services for a single day to see what happened. The next year, the same day, they turned it on permanently, and it's since then that IPv6 started growing

    • Re: 20 year is forever in tech years. (Score:5, Insightful)

      by Z00L00K ( 682162 ) on Monday January 05, 2026 @06:54AM (#65902615) Homepage Journal

      Even if something isn't supported the files should be left for download as 'legacy unsupported, use at your own risk'

    • HP-UX was never any good though, it was always the worst of the UNIXs

      • Comment removed based on user account deletion

      • Yeah, for sure. Sun were stubborn, HP were just weird. Nothing in HP-UX seemed to be as you'd expect (or want).

        Years ago, my first job was in an engineering outfit. I thankfully didn't have to go to this meeting, but my boss's boss's boss was chairing an 'evaluation' of a variety of file server technologies. We'd had (a very young) Netapp, Sun and HP in for evals - and it was pretty clear Netapp were far and away the leaders. However, we had HP systems because the company had moved to SAP, and we had Sun sy

      • I thought that the worst of the unixes was DECs - both OSF/1 and Ultrix

        • Re: (Score:2)

          by kriston ( 7886 )

          That's cutting it close. There is the bottom of the UNIX barrel where OSF/1, ULTRIX, Irix, original SunOS, Xenix, and HP/UX once fought for the dregs.

          • I thought that Irix was always good. Also, for SunOS, while I didn't care for OpenLook, the NeWS architecture looked good

      • Re: (Score:2)

        by kriston ( 7886 )

        It notched below SGI Irix as the worst UNIX. Even Xenix/SCO UNIX (prior to UNIXware) was better than HP/UX. Imagine a generic 4.3BSD with dozens of proprietary features and closed-source drivers and go from there.

    • True. Although I don't think they did much good in moving OpenVMS to Itanic. They could have just ended sales, while continuing support on Alphas

      I am also surprised that HP/UX wasn't EOLed earlier - w/ Itanic. There was no point moving it to the x86-64

  • Also, Itanium (Score:5, Interesting)

    by Tailhook ( 98486 ) on Monday January 05, 2026 @02:02AM (#65902379)

    With the end of 2025, the last commercial support obligations for Itanium hardware have ended as well. Essentially, Itanium finally, officially died 4 days ago.

    • Re: (Score:3)

      by gweihir ( 88907 )

      It is really surprising how long bad ideas keep staying around. Itanium is a monument to the sheer incompetence of Intel when it comes to CPU design.

      • Re:Also, Itanium (Score:5, Interesting)

        by Tailhook ( 98486 ) on Monday January 05, 2026 @03:13AM (#65902435)

        Itanium was the longest lived (25 years) of Intel's various failed attempts to kill x86. It obviously failed in this regard, but it was successful at killing other RISC CPU architectures, including PA-RISC and Alpha.

        Previous attempts include iAPX 432, i960 and i860, all now consigned to the dustbin of history, along with IA-64 (Itanium,) although the i960 had some success in embedded IO controllers.

        • Re:Also, Itanium (Score:4, Informative)

          by Sique ( 173459 ) on Monday January 05, 2026 @05:27AM (#65902557) Homepage
          Parts of Alpha AXP live on in AMD's processor line-up though. The front side bus is a direct successor of Alpha's front side bus, and x86-64 is heavily inspired by Alpha's architecture too.

          • Parts of Alpha AXP live on in AMD's processor line-up though. The front side bus is a direct successor of Alpha's front side bus, and x86-64 is heavily inspired by Alpha's architecture too.

            From what I understand, PCI Express is a direct descendant of the "Ropes" I/O subsystem invented by HP right at the time we were working with Intel on Itanium systems. I was working with some of the developers who came up with the concept.

        • Re: (Score:2)

          by gweihir ( 88907 )

          I would expect that competent people make it on the second or at least the 3rd attempt. Intel failed 6 times by this count. AMD, on the other hand, needed one attempt to succeed and that is why it is the AMD64 architecture. What a crass difference.

          • Re: (Score:2)

            by Tailhook ( 98486 )

            Intel has suffered from having fuck-you money since the 1980s. Such companies can suffer from accumulating faddish management and misallocating resources. Intel is a poster child example of this.

            That's ended very recently. Intel's net income graphs since 2022 are astonishing: the wheels fell off entirely. We'll see if they get their poop back in a group with Panther Lake next week. It's looking pretty solid.

            • Intel spent a whopping $30.2 billion over the study period. With that sum, the U.S. chipmaker couldâ(TM)ve given each of its 124,800 employees a $48,000 bonus every year from 2019 to 2023, or increased their R&D budget over this period by at least 30.2 billion. Intel is in line to receive as much as $8.5 billion in CHIPS subsidies â" the most of any firm. https://ips-dc.org/report-maxi... [ips-dc.org]

              • Big numbers sound scary but Intel's yearly R&D budget is $15bn. This isn't a big expense for them and quite inline (or at least was 4 years ago before their sales tanked) with a 10-20% revenue reinvestment on R&D. And they spent this money while returning profits to shareholders. And that R&D wasn't a complete loss, they did claw back nearly $9bn in sales.

            • Re: (Score:3)

              by gweihir ( 88907 )

              Intel has suffered from having fuck-you money since the 1980s. Such companies can suffer from accumulating faddish management and misallocating resources. Intel is a poster child example of this.

              Indeed. Incapability from having too much money, arrogance, power.

            • Re: (Score:2)

              by Junta ( 36770 )

              And nVidia's income graphs since 2022 took off. The big winner of the 'fuck-you' money in tech went to nVidia thanks to LLMs...

              AMD deserved to usurp, but they had long executed better technology and the market didn't reward them to the extent you would have expected..

          • I would expect that competent people make it on the second or at least the 3rd attempt.

            Competence often has little to do with market failure. Some 90% of ideas never succeed for various reasons. For Intel the biggest problem wasn't competence it was that the world was seeking compatibility, not another vendor lock-in. IA64 may have been a roaring success had it come a decade earlier when the wild west of different incompatible arch types were still demanded in a market place.

            Intel was it's own worst enemy, they were beaten by their own competent engineers and Xeon dominated.

          • It's easy to criticize, harder to execute.

          • AMD tried something no other CPU vendor considered - extending the x86 instruction set to 64-bits. In the process, they ensured that the 64-bit extensions alone would be RISC, so that theoretically, if they ever dropped 32-bit and 16-bit support, the result would be a pure RISC CPU

          • Re: (Score:2)

            by Junta ( 36770 )

            Itanium could have ultimately succeeded after some fixups if not for AMD.

            Like everyone *knew* it was obnoxious but for a time, everyone assumed that it was going to be the only path for a multi-vendor ecosystem to get 64-bit addressing, which was obviously going to be needed with PAE being a crappy limited workaround. All the other viable 64-bit architectures of the time were locked into vendors, and if you were going to have to recompile everything for a new architecture *anyway*, why not Itanium? It would

            • No it couldn't have. The design was discovered (to be fair, thanks to Itanium's failure) to be fundamentally broken. The compiler can't know what will be hitting the CPU when the program is actually run.

              • The original premise of VLIW - Multiflow and Cydrome - both of which were acquired by HP - did have an interesting research project going. RISC proved that a CPU didn't need a plethora of instructions, given that most higher level languages use only a subset of all the available instructions in the instruction set. VLIW tried to go a step further - trying to move hidden functionality such as register renaming, out of order execution, instruction scheduling - from the hardware to the compiler. It is argua

              • Re: (Score:2)

                by tlhIngan ( 30335 )

                No it couldn't have. The design was discovered (to be fair, thanks to Itanium's failure) to be fundamentally broken. The compiler can't know what will be hitting the CPU when the program is actually run.

                Yes it does know, because the CPU only executes one massively wide instruction at a time.

                The problem with VLIW is that the compiler technology back then wasn't very sophisticated. These days the modern optimizing compiler is shockingly good at optimization - even figuring out what you're trying to do and re-

          • Innovation happens from failure. I like that Intel was trying new things. Even failures bear fruit (Sandybridge benefited from a lot of things introduced by Netburst). Just like how AMD learned from the disaster that was Sledgehammer.

        • Re: (Score:3)

          by kertaamo ( 16100 )

          25 Years! Wow! Has anyone actually seen an Itanium machine? I thought it was still birth.

          Back when the i860 was released I attended a 1 day workshop on its architecture and how to program it, in assembler, at Intel. That was enough to convince me that the thing was impossible to program if you wanted to achieve its performance promises. So it turned out to be, the compiler writers could not do it.

        • I always wanted an Itanium box. You'd think these things would be dirt cheap at some point because no one wanted them. They were never ever cheap.

        • Yeah, that's the worst part of the Itanic legacy. It was no good itself, but the hype around it terminated far better CPUs already out there, like PA-RISC and Alpha. Not to mention, even SGI abandoned MIPS for that boondoggle, although they reduced their expense somewhat by adapting Linux on Itanic instead of porting IRIX there. Sun, despite initial commitments, switched to Opteron

          In fact, even from a VLIW POV, there were some advanced compiler techniques that were already adapted by both Alpha and Pow

        • Even without Itanium they would have died. Intel and AMD caught up to the performance and were far cheaper with a larger user base. With the exception of ARM the RISC cpus were on the retreat from workstation to server to specialty case.
      • Comment removed based on user account deletion

        • Once that was dropped, the original raison d'être of the Itanium - doing a pure VLIW CPU for HP, and maybe other Unixes - should have been resurrected. At the time, there were moves to merge the various Unixes - USL, OSF/1, BSDs - into one. That could have been done on this

          One thing I do wonder - did they ever manage to do a compiler that would do all the scheduling and runtime activities, moving it to compile-time? Maybe they could try it again today, using compilers that do machine learning (not

    • With the end of 2025, the last commercial support obligations for Itanium hardware have ended as well. Essentially, Itanium finally, officially died 4 days ago.

      Negative, Ghostrider. While standard support is up, HPE has an extended support system called "mature support" covering HP 9000 and Integrity servers until 2028, including HP-UX support [hpe.com]. So it's not truly dead just yet.

    • Re: (Score:2)

      by xanthos ( 73578 )
      Time to shutdown that last Superdome.

  • is it me? (Score:4, Informative)

    by diffract ( 7165501 ) on Monday January 05, 2026 @02:48AM (#65902407)
    Am I the only one who finds it hilarious that there's a news article about someone who turned off their old computer?

    • Right! And I was wondering who this "workstation owner" was, and thinking maybe he's the *only* one still using HP-UX! I wonder if he's got a couple of TRS-80s sitting around too!

  • I mean (Score:5, Interesting)

    by Anonymous Coward on Monday January 05, 2026 @03:09AM (#65902431)

    HP-UX was always the "busted" Unix. In the 1990s when you'd read instructions for compiling software, there were always workarounds for deficiencies with HP-UX.

    Irix was more than a "movie star" (appearing in Jurassic Park) -- all the graphics innovations we have today started with SGI, and Irix was solid. The first 64 bit Unix was IRIX. The Nintendo 64 was the direct child of SGI workstations. But the SGI keyboards were mushy as all hell so I guess nobody is perfect.

    Solaris had an enormous amount of effort put into making it as good as possible. Like the version of Troff that came with Solaris is light years beyond anything else available even today. And do you know why? Because there was a brief time in history that Unix typesetting was considered state of the art and people literally bought Unix workstations to run troff. So Sun dumped a ton of money into making Troff as good as humanly possible. Sun gave us NEWS and Java and all kinds of innovations.

    I don't doubt that HP-UX was capable but it's exactly the situation that the guy in the article is describing -- it was 100% an enterprise product sold to banks and similar customers with zero effort made to make it sexy or accessible to even broader commercial customers. When freaking IBM outshines you by a mile (like, you can get a Visual Studio integration for z/OS today) you've completely lost the plot.

    • Re: (Score:3)

      by Tailhook ( 98486 )

      IRIX was the origin of XFS in Linux. The design and command line syntax of the LVM subsystem in Linux was modeled after the HP-UX LVM implementation: HP's LVM was a leading implementation at the time and it worked well: I used it on HP-UX servers for backing up Oracle volumes, among other things.

    • Re: (Score:2)

      by Viol8 ( 599362 )

      Some versions of HP-UX 9 had a nasty bug whereby a fork() loop could hard crash the system because IIRC there was no bounds checking on the process table. I know, I found it by accident when learning to code multiprocess in C at uni. I was definately not Mr Popularity in the sys admin office that day.

      • Some versions of HP-UX 9 had a nasty bug whereby a fork() loop could hard crash the system because IIRC there was no bounds checking on the process table. I know, I found it by accident when learning to code multiprocess in C at uni. I was definately not Mr Popularity in the sys admin office that day.

        Been there, done that. HP-UX 10 was much better than 9 in many, many ways. I left the HP-UX world before v11 really hit the streets so I have no experience with it.

    • Yeah... I haven't touched a pa-risc workstation since '96, but the systems were pretty sharp at the time. HP-UX was always sort of just okay. Still, I feel for the guy and have a fondness for the systems. For context, I was doing most of my work on a Sun with 2GB of RAM, which at the time was insane. I had a ridiculously over-speced dual pentium desktop for running linux so I could edit code in emacs, but ran my code on the Sun -- which of course sat in my pointy haired boss's office so he could check h
      • I think the PA-RISC CPUs were great, but HP/UX dragged that down. If only NEXTSTEP had been their OS throughout once it came out

    • HP-UX was always the "busted" Unix. In the 1990s when you'd read instructions for compiling software, there were always workarounds for deficiencies with HP-UX.

      I worked on HP-UX in the mid to late '90s and I concur. HP-UX was always a weird love child of BSD and System V. It was virtually impossible to get open source packages to build, even things we supported like X-based programs.

      Regarding the availability of patches, well, I'm not surprised. It would take time and money to make patches openly available and by the time this became an issue, HP-UX revenue must have been rapidly dropping to zero. It is a shame HP didn't just beam all the patches to achive.org and

  • We had a bunch of HP-UX machines and a handful of Sun boxes running Solaris. Personally, I've been much happier working in an all-Linux environment.

    • Re: (Score:2)

      by Viol8 ( 599362 )

      Back when Solaris and HP-UX was in the ascendent Linux frankly wasn't really up to much either at the application, graphics or even system level (proper pthreads didn't arrive until kernel version 3, before then it was faked with seperate processes) no I'm not sure what the appeal of early linux was other than being able to run it on a PC. And yes, I've used all 3 and still using Linux today.

      • One major issue I had w/ the earliest Linuxes - be it Caldera, Mandrake, Corel,... as well as the earliest FreeBSD versions - was networking. Not WiFi, but networking. Sometimes, even X11 didn't work, but when it did, it was impossible connecting to the internet. One disclaimer here - those were the days of dialup internet, so the OSs weren't familiar w/ PPP or phone protocols. It wasn't until DSL and Cable took over and had to be connected via Ethernet that that situation changed

    • When NEXTSTEP was ported to both PA-RISC and then Sparc, I was excited. I wasn't too enamored those days w/ the Unix shells, so when it looked like those workstations would ship w/ NEXTSTEP, I was pretty excited. But then Apple bought out NEXT, and that was that

      Recently, on YouTube, I saw people running NEXTSTEP on PA-RISC and Sparc workstations, and was pretty impressed. If only NEXT had those CPUs throughout, rather than the Motorola 680x0

    • But Linux did exist on both Sparc and PA-RISC. Maybe you should have taken a particular distro and installed it on all of them, then had the same operating environment on all your boxes

  • Used to sysadmin hpux (Score:5, Interesting)

    by caseih ( 160668 ) on Monday January 05, 2026 @03:31AM (#65902443)

    In fact HPUX was the first Unix I was introduced to as a CS student back in 1997. I remember the large monitors that were super high resolution and CDE looked great. You didn't need anti-aliasing because the fonts (well-designed bitmap fonts) looked crisp and readable on those enormous 19" 1600x1200 CRTs. When LCDs came out they were a real step backwards compared to those monitors. Has take a long time to finally exceed the quality of those 1600x1200 displays with modern 2k and 4k displays.

    When I got my first job as a sysadmin at a uni in the CS dept we had a whole bunch of HPUX workstations, and also a file server running HPUX. I've forgotten nearly everything I knew about HPUX but I vaguely remember the SAM utility. Around that time I was introduced to Linux. Eventually as Linux got better and better I managed to convince the department to let us roll out Linux in one computer lab as a test. It was so popular with the students compared to the Solaris and HPUX labs that by around 2000 we had replaced all of the open lab HP and Solaris workstations with Linux on x86 PCs (RedHat if I recall) and everyone from the faculty to the students were quite happy with them.

    I feel somewhat bad at helping an entire CS department migrate away from HPUX machines as they really were high quality machines, but it was good move, for the department, the students, and for me professionally.

    • Re: (Score:3)

      by Sique ( 173459 )
      I worked with HP-UX in 1990/1991, when it was still HP-UX 7.1. I was once credited in Emacs 19.10 for porting it to HP-UX 9.x in 1993. At the time, a friend of mine coded a variant of Atari's MIDImaze for UNIX, and we were playing netmaze on a pool of HP-UX workstations in the university's computer pool, because the new PA-RISC architecture was REALLY fast compared to anything Sun SPARC or x86 had to offer.

    • I remember all that.
      In fact, I bought a bunch of those HP workstations when they sold them off.
      I got a bunch of the C180s from the CS dept. and a few C240s (probably from the Engineering labs) to play with at home.
      I learned basic HP-UX admin, SAM on them. CDE wasn't terrible.
      I used one with Debian Linux on it for my webserver for years--since non-x86 adds an extra layer of protection from exploits--a buffer overflow full of x86 instructions won't do anything on a PA-Risc cpu except properly crash--not provi

  • Fuck Carly et all (Score:2, Funny)

    by Anonymous Coward
    Fuck Carly and the rest of the assclowns that destroyed HP, Compaq, and seriously wounded Agilent. Special mention CEO's of Sun Microsystems, Commodore, Atari, and Palm.

    all of you can eat giant bags of dick

    • Commodore especially whoever fucked that company, man...

    • She and her gaggle of over-promoted airheads also wreaked havoc in Lucent.

      I don't remember which one was which but one had a BA in Medieval History and the other had a BA in Political Science: Carly Fiorina and Pat Russo.

      • If Carly's was medieval history, she was humiliated by someone who pointed out that her praise of the "islamic golden age" was horsemanure. If it was political science, she showed what a political genius she was first running for senate against Barbara Boxer, and later for POTUS in that 16 man race that Trump ultimately won
    • The cascaded takeovers - first of DEC by Compaq, and then of Compaq by HP - saw to it that there was a reduced competition. Too bad the FTC was asleep at the wheel

  • Don't tell me it's MacOS. (Also don't be an ignorant ass and say Linux.)

    • Re:What is the last unix standing? (Score:5, Informative)

      by 93 Escort Wagon ( 326346 ) on Monday January 05, 2026 @04:12AM (#65902485)

      Is a fork still considered Unix? AIX still exists, as does Solaris and of course SCO Unix.

      The BSDs are also still around, although according to Wikipedia [wikipedia.org] they're considered "Unix-like"... as is Linux.

      Also according to Wikipedia, both macOS and Z/OS are current, certified Unix systems. SCO OpenServer is also listed there, but it doesn't appear to be maintained.

      • SCO Unix, which used to be SVR4.2, totally vanished. A different company Xinuos [xinuos.com], acquired them, and now market SCO OpenServer 5 & 6, as well as Unixware 7. I believe all they do is maintenance, rather than sell anything

    • A couple of Linux distros have been certified as UNIX(tm) before: Inspur's K-UX and Huawei's EulerOS.

      • Re: (Score:2)

        by flink ( 18449 )

        A couple of Linux distros have been certified as UNIX(tm) before: Inspur's K-UX and Huawei's EulerOS.

        That would be POSIX certification. That doesn't make it UNIX. POSIX certification just guarantees compatibility with certain de facto standards that grew out of the UNIX ecosystem, not that the OS is UNIX(tm).

    • AIX is still a current IBM product. It runs on their POWER line of processors.

    • If you want the USL blessed editions, it would be OpenIndiana - dunno whether Oracle still sells Solaris. If you want the BSD blessed editions, it would be NetBSD, FreeBSD and all their derived distros

      Incidentally, when you say "Unix", do you mean distributions who pay X-Open an arm and a leg for the Unix-certification logo? Or just any OS that supports the entire POSIX suite of commands?

  • We're waaaay past the steam-age of microcomputers and even further past proprietary software of this kind even being a thing. There is absolutely _nothing_ left on this sort of OS (and hardware) that can't be replaced by some rasberry pi and way superior FOSS in less than an hour two orders of magnitude cheaper and 3+ orders of magnitude more performant. My cheap-ass 200 Euro smartphone has 12 GB of RAM and can run circles around this dinosaur of a microcomputer "workstation" in power-saving mode. That UI l

  • It was a nightmare to maintain. As far as I can remember, patches sometimes clashed. So if patch X was required for application A, it might break application B.

    The build team configuration managed the patches.

    Was thrilled to move to Solaris on Sparc

  • When I first started in IT at the Enterprise level many many years ago I use HPUX for developing telephony software. At the time we used HPUX as a Command and control of our systems. It was also used to develop software for the Motorola 68020 switching systems.

    HPUX was not actually a multithreaded OS then. It had fake threads implemented with interrupts. This caused us no end of grief at the time. All of these low level hacks to ensure that things didn't lock up. We needed something that could handle

  • Won't miss it. (Score:3)

    by kschendel ( 644489 ) on Monday January 05, 2026 @09:14AM (#65902735) Homepage

    Back in the late 80's and 90's, I did a lot of porting among the various Unixes, mostly Solaris, AIX, HP-UX, DG, and a couple others. The HP-SUX port was always the most troublesome. The PA-RISC hardware was nice enough, but not HP-UX. It seemed like every time we turned around there was some new quirk or incompatibility. I'm glad it's gone.

  • When software can't be freely shared and hardware not freely purchasable, the whole ecosystem fails. Now everything Unixy is Linux and BSD, with Illumos if you want the remains of Sun from the great Oracle lockdown.

  • Which is what the City of Chicago 911 system ran on in 2001.

    Why, yes, I was one of the two admins for it.

  • starbase graphics (Score:4, Interesting)

    by belmolis ( 702863 ) on Monday January 05, 2026 @12:22PM (#65903243) Homepage
    Does anyone remember Starbase graphics, the graphics/windowing system that came with HP-UX? Back in the 1980s we got a bunch of Bobcat's running HP-UX, with, for the time, lovely big color graphics terminals and thought they were great. Coming from 4.2BSD on Vaxen, HP-UX was a little weird - I still have code with #ifdefs in it for HP-UX - but it wasn't that weird, Starbase was easy to use. It was also closely tied to HPGL, so if you had something that ran a terminal, it was easy to send it to a pen plotter. Even so, it was an improvement in the end to move from HP-UX to BSD and from Starbase to X Windows.

  • Hpsux was the worst Unix. Their documentation for ipsec was backwards (confused the source and destination) which was fun for me the only time I worked on it. Fuck the HP of today in every way.

Slashdot Top Deals

Lead me not into temptation... I can find it myself.

Close