Slashdot Mirror
HP Plots New Courses with HP-UX/Tru64
Uberhacker.Com writes "HP has given up on trying to bring key parts of Compaq/DEC's Tru64 operating system into HP-UX. They had once planned for the Tru64 goodies to arrive this year and made a big deal of this quick turnaround when it first acquired Compaq. Ironically, HP also announced today that it is expanding its Alpha RetainTrust program for Tru64 UNIX customers." The linked article also notes that HP has decided that it will proceed forward with purchasing some of the technology from Veritas.
it was a doomed project in the first place, they had no way of making it
this sig no verb
WHy would they give up already? I never knew that theyw ere weak once in this area, competitive yes but not weak. I always have heard about HP-64 servers, heck even RIT, has some of them running database or whatever they're running. I just don;t want to see a good company go down just simply because they gave up. Now who will rise to the occassion? The GEEKS!
May
My question to the world is this: where were you on the day when HP gave up on trying to bring key parts of Compaq/DEC's Tru64 operating system into HP-UX?
Some Guy: "Ummm.....uhhh......what?"
It would be cool if it didn't suck.
...some of us could have some fun again!
My mom always said, "Jim, you're 1 in a million." Given the current population, there are 7000 of me. God help us all!
Follow Sun's business model and benefit us all!
...yup...
HP-UX has always been clunky, trying to splice in some DNA from a totally unrelated (and more technically advanced) version of Unix was a pretty tall order. They probably would have had better luck porting Tru64 over to PA-RISC and trying to merge in the bits they really wanted from HP-UX.
Just junk food for thought...
Trying to port the Tru64 clustering features into HP/UX was a bit like trying to fit a jet engine into a Yugo.
I would expect two systems based on the same code to be pretty easy to merge.
I know that all UNIX vendors had fun adding incompatible extensions -- but I always figured that a company with the code to the original, to their own incompatabilities, and to somebody else's imcompatabilities wouldn't need to work that hard to get everything to work together.
Guess I was wrong.
If HP Unix is dominant in your business model, why not open source the other, more advanced offerings so that others not constrained by your business model can make it work?
Most sites that are migrating are going away from both as fast as they can. There are a small fraction that truly depend on clustering or other proprietary feature, unfortunately everybody is holding on tenaciously to said features despite the fact that they really do 99% of the applications no good. And most commercial applications have been somehow hoodwinked into the proprietary hooks.
Does HP do anything leading edge anymore? I love HP printers and have much better reliability and luck with HP than epson, lexmark and cannon. it's just too bad Tru64 and all that great Alpha technology is going down the tubes. Hopefully all that knowledge the Alpha guys acquired is being used to advance Opteron and Itanium.
Opening HPUX would benefit us all, or make us go insane? If they did do that though, at least they could get one accolade out of it.
Ultimately, manufacturers like HP and Sun are increasingly pushed into niche and legacy markets as PCs get faster and Linux and BSD become more capable. I would expect more withdrawals like this in the future rather than less.
More than that, HP has seen considerable pressure from younger webmasters who see their business practices as inequitous and dangerous. Admittedly such efforts are probably scattered and short lived, but they seem to have some sympathy amongst conservative culture warriors as well. Ultimately, only time will tell whether these efforts have any effect on HP's bottom line.
A Proud Member of the Reality Oriented Community.
Some of the TruCluster stuff is REALLY COOL!
For those not familiar, picture a filesystem that can be mounted on 2 or more hosts at once instead of mounted from one then NFS-exported (or Samba, either way) from one host to all the others.
TruCluster was way ahead of its' time, the Digital guys were WAY ahead of their time.
This just really ticks me off because the Veritas version is NOT AS GOOD and has FAR MORE BUGS.
Aaargh!
Some days, I hate HP.
HP-SUX.
I mean it in a nice way though.
They are letting Alpha CPUs die, even though they rock, because they sank so much money into Itanium.
They are dropping a Unix better than their own, because they can't suck it up and admit Tru64 is better. (I am taking your word for it #6336)
HP-SUX
From the process scheduling code:
_O_
.|< The named which can be named is not the true named
What happens when their customers already committed to Tru64? Are they left out to dry? (My work was going to go to Tru64)
got sig?
There are a number of open source projects that are trying to add things like SSI to Linux. Why not join them and make it happen for everyone.
So *that's* why they laid us all off...
Just some days? I hate them EVERY DAY!
-- I am. Therefore, I think!
"It's a good thing that HP never acquired the rights to penicillin. If they had, mankind would have perished from widespread disease while HP tried to figure out how integrate it with anthrax."
"It's the height of ridiculousness to say for those 9 lines you get hundreds of millions."
In further news, HP is offering HonestToGod 128 support with optional WeSwearWeWon'tScrewYou coverage.
-Derek
Something that Intel can't do.
And apparently neither can HP.
Sad to see how superior technology gets caught up in corporate mergers and gets killed. First, the DEC faithful had to swallow up the indignity of seeing DEC swallowed up by a Compaq, and then this...
I spent many a nights hacking Fortran on DEC boxes running everything from Ultrix to Digital Unix 4.. those were the good times..
The perfect sig is a lot like silence, only louder
HP is nothing more than an ink-cartridge-delivery system.
I understand why they bought Compaq, but why not spin off Digital. I wish Compaq had never bought Digital. Digital did a lot of cool things, Compaq was able to help them some, but HP has no idea what to do with their stuff.
"brxref
Having first used VAXClusters in 1987 eveything else I've come across seems toy in comparison. A VAXCluster gave us disk that worked just like a local disk but was shared across the cluster. No one VAX 'owned' it (no LAVC here!). It's peer to peer disk sharing with all the lockin problems sorted. We ran a navigational database (VAX RMS) over the VAXCluster with 3 VAXes and hundreds of users.
If VAXCluster technology is lost then it's a tragic waste of a good technology.
I guess that's the problem. It is soo cool that it requires a really cool OS to work with, it has too much karma to even recognize HP-UX as an operating system.
My boss and I have been just talking about this. HP is junking all of their best technologies. Ttu64 had a best of breed clustering. So, what does HP do? Junk it and buy the technology from Veritas.
No surprise they junked the Alpha. No surprise they even junked the PA-RISC. No wonder they are becoming another Dell. Yep, HP used to mean quality at a higher cost - but people were willing to put up with that because HP anything was going to work with precision, reliably for the next century. Now, the HP servers and spares we are getting are less and less reliable.
Sigged!
Some days? Why not everyday! You're not being negative enough.
both the reg and the inqwell have more information. Somewhere buried in the stories of the past few weeks is the strange fact that Compaq/DEC had a license for Veritas storage technology file system which they folded into Tru64 and now HP is going to pay for the licence a second time for HP-UX. Truly a sign of a management team that does not know what it has.
4 _l etters/
http://www.theregister.co.uk/2004/12/02/hp_ends_tr u64/
THe letters from customers are interesting as well.
http://www.theregister.co.uk/2004/12/03/hp_tru6
http://www.chipzilla.com/?article=20021
putting the 'B' in LGBTQ+
well, I'd love them to open source it but I doubt that'l happen. but if they won't give it away, perhaps they'd sell it? and you know, there is a unix (not (tm)) vendor who I think could really use alot of this technology. I'm talking about apple. I'm not suggesting that they try and take over the old customer base and support, but I would think that they could make some serious use of the technology.
:)
they obviously want a piece of the enterprise market but right now, I don't think they've got any real good reasons someone would choose them over any other unix vendor, but if they could bring in some of the (apparently) classy digital clustering technology, and indeed, any other useful things, then they could really make a go for some enterprise stuff.
I just want to see what they do with an 8U server, xserve style
dave
Didn't DEC name it Ultrix at one stage, or did I just dream that part?
${YEAR+1} is going to be the year of Linux on the desktop!
Its a shame that DEC / Compaq and now HP have not really supported one of the most advanced and stable operating systems the world has ever seen. Having been a long term alpha processor / server user with both OpenVMS and Tru64 it was such a shame what happened to DEC, Alpha, OpenVMS, Tru64. It just goes to show that at the end of the day the best products may not survive.
Aaah! You de-referenced his name 3 times!
I am TheRaven on Soylent News
When Amazon switched over to HP servers running HP/UX 11 in 2000 there were a lot of annoying things about the change in operating systems but as far as the filesystems went I thought that I had died and gone to heaven. LVM on HP was rock stable and simple compared to the insanely complex LSM on Digital Unix, and the HP's filesystem didn't shit itself the way that AdvFS, which we referred to as the "Adventure FileSystem" because using it was a real adventure in finding out whether or not your files would be available in a day's time, did. I for one won't miss AdvFS.
cheap labor conservatives - they want to keep you hungry enough to be thankful for minimum wage.
The Tru64 license came from AIX, but the code is based directly on 4.3-Reno (the last 4.3BSD release before 4.4, and functionally almost identical to 4.4BSD/FreeBSD 2.x/Darwin 1.x) and Mach. It's unrelated to the original code donated to OSF. The code is close enough that I used to use the FreeBSD source tree when I was tracking down and fixing problems in our Alphaservers. HPUX is based on 4.2BSD with many generations of hacks and System V code imports. It would have been more logical to do the port in the other direction.
Urban myth. AIX had nothing to do with Tru64, aka OSF/1. There were DEC, HP and IBM copyrights in there because of the joint development effort, i.e., each contributed code.
What rubbish. I've used both, and H-POX is about 3 years behind Tru64 in development terms. Not to mention very proprietary in its feel. Some simple examples:
-- you can't get a process listing from ps(1) that shows you the run state ( ps -ef doesn't support that) cos it doesn't understand any BSD flags. Tru64 has supported both bsd and sysv flags from inception.
-- Every Unix I've ever come across supports "ifconfig -a" to list all available network devices. Even OSX supports it. But H-POX has to be different, you need to use 'lanscan' to find the device name, and then use ifconfig on each device seperately to see the same info.
I can go on
For those not familiar, picture a filesystem that can be mounted on 2 or more hosts at once instead of mounted from one then NFS-exported (or Samba, either way) from one host to all the others.
Serious question: At work i do some testing with iSCSI (SCSI over TCP/IP). The target-device is being provided by a Linux-PC (see Linux Enterprise Target for details). We tried a little and were able to mount that target at three different Windows 2000/XP machines using Microsofts iSCSI Initiator.
We had the same filesystem on three PCs and could read / write / erase.
Is this quite similar or were there other features in tru64 clustering that one would miss in todays OS'es (Linux/BSD/Windows)?
AIX had nothing to do with Tru64, aka OSF/1
I didn't say it did: quite the opposite, I was talking about the license... this was before the USG/CSRG lawsuits: you needed a System V source license for 4.3-Reno.
True64 is a very insecure OS. Having run a true64 server. and after seeing that it was using CRYPT for passwd encryption and then not supporting shadow passwords as well as TONS OF other crap problems I switched the server (dual COMPAQ alpha blah blah blah) over to FREEBSD and it has hummed like a bird ever since. I would suggest to HP to leave true64 in the trash can as well as with HPUX but then.... these big guys keep dragging along there old unixes like there is no tomorow.
http://blog.evogts.com
Presumably it was not an HP person that would moderate a reasonable request to be a flaimbait. There are lots of them and their customers, I am sure, who still know how great it is. Digital stuff was great stuff, and they had many technologies before the market curve, some that have never appeared outside. How about their spiralog filesystem. I'd love to be able to try to make those concepts work with Linux.
Yes AdvFS was a bit buggy at V4. It was also funneled and tied to running on the base CPU, which created a real bottle neck.
But it was completely re-written for V5 to be fully multi threaded and could run on any CPU. This was a real turning point for AdvFS, it was incredibly stable a very fast after that.
I've been building TruCluster since the beginning, hundreds of them; and since V5 I've never had a cluster go belly up to the point of being unrecoverable. AdvFS is rock solid now.
In fact, I'm the only person in TruCluster circles I know that's ever had to do a full TruCluster restore from backup. And the one and only time I did, it was a planned move to new systems on a different site with different hardware.
It's the most solid O/S I've ever had the privilege of working with.
For starters, it's Tru64, there's no "e" in the name. Seeing as you don't know how to read, it's no surprise to me that you didn't read up on it's security features. The default install comes configured with "base" security. But if you'd taken any time to learn, you've quickly have discovered the existence of "enhanced"security, which implements "bigcrypt" and moves the password out of
When the enhanced security mechanism is installed and configured, the system is referred to as a trusted system. Enabling all enhanced security features will result in a system that can be configured to meet the C2 class of trust, as defined by the Trusted Computer System Evaluation Criteria (TCSEC, also called the Orange Book). The system also meets the F-C2 functional class as defined in the Information Technology Security Evaluation Criteria (ITSEC).
Enhanced security extends BSD security by providing:
* Login control enhancements
* Password enhancements
The following lists enhanced security features for login control:
* Recording the last terminal used for a successful and unsuccessful login
* Recording the time of the last successful login
* Recording the time of the last unsuccessful login attempt
* Recording the number of consecutive unsuccessful login attempts
* Automatic account lockout after a specified number of consecutive bad access attempts
* A per-terminal setting for the delay between consecutive login attempts, and the maximum amount of time each attempt is allowed to complete the login before being declared a failed attempt
* A per-terminal setting for the maximum consecutive failed login attempts before locking any new accesses from that terminal
* Display information about last successful and last unsuccessful login attempts at login time.
* Recording login information whenever a login occurs over the terminal.
Enhanced security provides the following features for password control:
* Configurable maximum password length, up to 80 characters
* Configurable password lifetimes
* Variable minimum password length
* System-generated passwords that take the form of a pronounceable password made up of meaningless syllables, an unpronounceable password made up of random characters from the character set, or an unpronounceable password made up of random letters from the alphabet. (All letters are from ASCII.)
* Per-user password generation flags, which include the ability to require a user to have a system-generated password
* Record of who (besides the user) last changed the user's password
* Password usage history
I could go on and on about the auditing capabilities too, which enables you to capture the calling of virtually every system and library call on the system, trackable by an Audit-ID that stays the same even if you change your Real UID to some other user.
And then there SIA, the modular security architecture that enables you to write and plug-in your own security if you so desire.
even better. seriously, if there is a box you use to run a web server and you never license it, you can still run the thing off a serial console and it is a beautiful thing indeed. just watch the hackers break their noses trying to bust in...
I love tru64. i use tru64 every day and i will continue to use it as long as i can get spare parts because, frankly, it really is the best.
maybe one day i'll be smart enough to come up with a cool sig, too.
Ah, but what would you give just to have even LAVC-like clustering available for Linux? Is there anything better than the silly failover-to-a-hot-spare style clustering available for Linux?
Uh, I hope you meant "locking". :-)
CUR ALLOC 20195.....5804M
IBM does some pretty rotten and underhand things when configuring systems for benchmarks. For example, they'll take a 16-way box and disable half of the CPU cores to double the cache available to the remaining 8, and submit the results as for an 8-way box. The other thing they do is put small amounts of really expensive DDR RAM in. Finally, they tout results from benchmarks that are 10 years old and no longer representative of true workloads.
IBM sure knows how to play the benchmarketting game and impress the PHBs.
Simultaneously? You had three Windows PCs mount the same FAT filesystem and modify it at the same time? How well did locking work?