Follow Slashdot blog updates by subscribing to our blog RSS feed

 


Forgot your password?
Close
typodupeerror
×
Linux Business Sun Microsystems

The End of Sun's Cobalt Servers 88

Posted by CmdrTaco from the they-were-so-cute-too dept.
knighten writes "Sun Microsystems has taken the last of its Cobalt line of server appliances off the shelves in favor of the AMD based Sun Fire line." The article makes note of several relevant bits of history regarding Cobalt, the Appliance Server market, and Sun's Linux strategies.
This discussion has been archived. No new comments can be posted.

The End of Sun's Cobalt Servers More

The End of Sun's Cobalt Servers

Comments Filter:
  • but the upcoming new Sun Opteron systems look pretty sweet! :-)

  • Continued Support (Score:4, Interesting)

    by ohchaos ( 564646 ) * on Sunday December 28, 2003 @12:32PM (#7821923)
    This will be yet another good test for the opensource concept. As sun ends support for these devices, will someone else pick up the ball (be it in a commercial sense, or a free sense) and continue providing updates (at least security updates....) for these now orphaned linux-based products?

  • Ease of use (Score:5, Interesting)

    by mocker ( 197762 ) on Sunday December 28, 2003 @12:32PM (#7821924)
    As someone who maintains Cobalt servers on a daily basis I can say that this has been coming for some time. Sun has been very poor about releasing patches for exploits on the Cobalt server. These are fun servers to play with when you get the hang of it, but newer control panels (Plesk, CPanel) pretty much make them obsolete.
    • Yeah, these things were revolutionary for their time. I was disappointed when I learned that Sun had purchased Cobalt, apparently with good reason. It's a shame Sun let them stagnate.
    • I see that you didn't mention the popular Ensim control panel software.

      Good thing. Ensim is the worst, buggiest, least stable server software I have ever used. Ever. Period.

      I realize that this is a bit off-topic, but if it saves one person the trouble that I went through working with Ensim (and finally, the trouble of having to wipe several servers to get rid of it), then it was worth it.
      • Oh my God yes, Ensim is terrible. I was forced to work with it at a (somehow still in business) company I used to work for. Its feature set is tiny, and what is there doesn't work right.

        It's really surprising actually, since when you look under the hood, the methods it uses to do what it does (the whole "multiple servers on one physical CPU" thing) aren't all that revolutionary.

  • From the article:
    Sun is strategically focused on delivering choice and performance to our customers, offering general purpose x86 servers that can run Solaris SPARC [...] operating systems.

    An x86 machine that that can run Solaris SPARC operating systems? Clever... :-)

  • Sun is going down (Score:3, Interesting)

    by superpulpsicle ( 533373 ) on Sunday December 28, 2003 @12:33PM (#7821932)
    I am finding it difficult to see Sun's position in the market more and more each year. The best thing to have happened to this company was free downloads of solaris 9 x86. It is virtually impossible to convince my manager to buy Sun anything nowadays.

    For high end stuff we have AIX. It comes with LVM and other critical stuff. It has ridiculously stable support for fibre channels and just the most outstanding support.

    For middle to low end we have PCs with windows and linux.

    I can't seem to see where Sun (with or without their cobalt server) fits in today's market anymore.

    • Re:Sun is going down (Score:5, Interesting)

      by johnlcallaway ( 165670 ) on Sunday December 28, 2003 @01:00PM (#7822012)
      I don't have much AIX experience, but there is one reason why I suggested Sun for our data center servers, binary scalability. Sun servers scale from 1U/1CPU lower cost servers (5K) and developer stations to clusters of over 300 CPU servers, all with full binary compatiblity. I have yet to not be able to take software off a 1CPU low-end sun box and not be able to run it on the top-of-the-line servers without any recompiling.

      This provides the capability to develop on low-end boxes without the headaches associated with recompiling on production servers and shortens our development cycle.

      I will admit though, with most of the development moving into the Java world, maybe this doesn't make as much sense. However, we have still found it useful to do some of the development work on smaller Sun boxes for performance benchmarking and forcasting performance when something goes into production.

      I've stayed away from IBM because of past bad experiences, providing quotes that are not complete solutions resulting in server cost overruns or software that is not yet written. (They once replied to a quote for a automated-failover system, and provided an neat OS2 solution. When pressed on how the failover worked, they finally admitted that once we ordered the system, they would write it.)

      If IBM costs have come down, and their ability to fully respond to quotes has improved, maybe they are worth another look. But if not, Sun is still my server of choice for critical production systems.

      • Re:Sun is going down (Score:3, Insightful)

        by argoff ( 142580 )

        Sun servers scale from 1U/1CPU lower cost servers (5K) and developer stations to clusters of over 300 CPU servers, all with full binary compatiblity. I have yet to not be able to take software off a 1CPU low-end sun box and not be able to run it on the top-of-the-line servers without any recompiling.

        This provides the capability to develop on low-end boxes without the headaches associated with recompiling on production servers and shortens our development cycle.

        Most datacenters I've been in could

        • Re:Sun is going down (Score:5, Insightful)

          by johnlcallaway ( 165670 ) on Sunday December 28, 2003 @02:25PM (#7822379)
          Your statement about a 'truly scaleable datacenter' does not make sense. You say you've never seen one, but then you mention that Linux has delivered scalabilty beneifts Sun could never imagine. Please make up your mind which it is.

          Myself, I have seen scalability. I've have seen applications start out on 2cpu $20K database servers and migrate to E15-type servers without any code change. I'm not saying Linux (or AIX) cannot do this, but this is scalability that Sun does provide at a competitive price.

          Having priced the cost of multi-CPU server-quality x86 platforms, there is very little cost benefit to going there. Multi-CPU server quality x86 boxes cost almost as much as the same Sun boxes.

          As for Java, since you did not mention any alternative, I cannot provide any response. However, since the application servers and web server provide enough bandwidth, there is no reason to switch to anything else. I am not ready to jump on the 'let's switch everything to Linux' bandwagon yet, but I am on the 'let's review it as we deploy new products or grow existing ones.'

          I am sure Linux will migrate into our datacenter, and eventually support production applications. It just doesn't make any sense to replace what works and is proven with something else unless there is a clear cost advantage.
          • OK, I am curious. What kind of applications does require THAT MUCH cpu power?

            And standart sql databases don't count, because you don't deploy them on your highend application server.

            • Databases are what require that kind of power. How about core banking systems, payment processing...databeses. Online, can't fail ever, fast as possible databases.

            • Re:Sun is going down (Score:4, Insightful)

              by johnlcallaway ( 165670 ) on Sunday December 28, 2003 @04:18PM (#7823081)
              Risk analysis applications that process 10 million complex transactions in a few hours and produce hundreds of megs of output.

              A large server may not be dedicated to just one database. There is a significant cost savings in purchasing a single high end system for several databases rather than splitting them over several smaller systems. With the partitioning technology available from Sun, you can start with a smallish (15 CPU) server and grow up as is needed. Using smaller servers to start with is cheaper, but the cost of swapping them out is expensive if the database requires more horsepower than the system can deliver. (Sun isn't the only one to support hardware/software partitioning, I'm just using it as an example.)

              It is always a juggling act to find the appropriate cost/performance mix that provides both long term and short term advantages. Purchasing systems that are not expandable is often cheaper in the short term. Purchasing expandable systems is often cheaper in the long run, but the risk is the application may not grow enough to realize the savings before the technology becomes obsolete.

              I hate estimating hardware requirements these days...

            • Re:Sun is going down (Score:3, Interesting)

              by GreggBert ( 89663 )
              Big, bloated CMS systems like Vignette Story Server need that kind of Power. Yes, there are better, more efficient choices for CMS software, but when you have trained a dozen developers in your I.T. Dept in Story Server development, bought license worth more than a house and management has you locked in, your choices are limited.

              Therefore, you need this kind of power. Sun servers have, so far, provided us with that power. That and heavy use of technologies like Akamai Edgesuite have allowed to handle som

        • IMHO, java still has too much overhead. We havent got the efficiency and equality that has been promised yet, but I assume we will get there someday.

          I think this really depends on what one is doing. On server-side web app (and similar) processing, Java (with 1.4 JDK) seems to be pretty low overhead, considering what it's doing (as in fully standards compliant with unicode encodings etc). Most overhead comes (IMO) from various (often unnecessary complicated, for what they do) frameworks, and from people

          • Most overhead comes (IMO) from various (often unnecessary complicated, for what they do) frameworks, and from people who have no experience in considering performance impacts on designs or implementations.

            I have to agree. I hate using generic packages to do tasks in Jazilla, which makes the process more complicated and wastes code on inefficient conversion routines. Whenever possible, I try to code stuff in the simplest form I can, and without anything Jazilla doesn't need.
            • Ok, this is off-topic, but how is Jazilla doing? I probably need to have a look at home page... I'd be very interested in having a look, perhaps even helping if there's something interesting I might have expertise on. Back in the day I did write a simple HTML rendering component (JDK 1.1, before swing), and nowadays am interested in implementing efficient data structures. Plus it'd be fun to get to do some actual app development (not just server-side) for a change. :-)
              • I haven't had a lot of time compared to the start of the year to work on it, but Milestone 3 has just come out of the door.

                Funny how I'm getting people who work on Server side stuff wanting to work on Jazilla... I'm only starting to work on Servlets :)
        • Scalability may be no longer SUN's 'domain' or key strength, but when it comes to 'keep going', solaris on sun hardware is almost a class on it's own. Where I've seen Linux on intel come to a grinding halt due to excessive load, I have seen solaris going and going, at loads which are really bizarre.
          I am a devoted Linux user, but I can say that SUN's products are top-class, and thus deserve a place in today's IT market.
          My work requires absolute reliability from the systems we purchase, and so far SUN require

  • What credentials do you need to report on tech? (Score:3, Insightful)

    by DShard ( 159067 ) on Sunday December 28, 2003 @12:36PM (#7821944)
    The article reads like the cobalt was the only true "server appliance" left. As I see it this forgets all about blade servers, network attached storage and a nifty box I saw from IBM that allowed you to stack multiple servers to form a larger one footprint box. They have even moved into the consumer space with media servers and firewalls. The author obviously knows nothing about what is or isn't a buzzword.

  • Those Cobalt cubes were cute... (Score:4, Interesting)

    by MsGeek ( 162936 ) on Sunday December 28, 2003 @12:40PM (#7821954) Homepage Journal
    I suppose that all those looking for a similar device should look into VIA Mini-ITX. This one in particular [viavpsd.com] looks like it could be very useful with its twin Ethernet interfaces and four (count 'em) serial ports. Router/RAS anyone?

    Still...they just don't have the Kawaii factor of the Cobalt cube. I want one but I can't spare the money, dammit.

    • and four (count 'em) serial ports.

      Uh, yeah. I suppose if you count USB too, then you would have five, but there is only one RS-232 port on that board. It isn't often that I see USB referred to as a serial port, while technically true, most people just call it USB.
    • I have a Qube2 that has held up pretty well. I bought it on eBay for $170, and it functions as a gateway, Web, and Mail server for my home network. In a way, the end of the Cobalt line does have some benefits; there may be more of these that hit the market as used items. I wouldn' mind getting a bunch more and stacking them in my basement to look like an even bigger cube...

  • Should have never bought it (Score:5, Insightful)

    by argoff ( 142580 ) on Sunday December 28, 2003 @12:49PM (#7821980)
    I know I'm gonna get nailed for this, but the fact is that Sun should have never bought them. In a way, Sun was unworthy - Sun's CEO was too jealous of Microsoft to ever make a service based approach work, or at least be competitive price wise.

    From a data center perspective, yeah its true that Sun boxes can do some things better than x86 boxes running Linux, but I can't tell you how many times I've seen companies buy 100K worth of Sun servers to do services that I know darn well could just as well by an x86 box or two. It always amazed me to see the salesman talk "scalable" for systems that were really farmable. Yeah, experience with high end Sun boxes was great for my resume, thanks, but I wanted my career to have meaning too - and having a bunch of overpriced toys just for the sake of ego seems a little shallow, don't you think. (Sorta like Sun's CEO, :)

    IMHO, the Sun just needs to set. Now that 64 bit Opetrons are out, they will have almost nothing to offer in the midrange. The lost the lowrange a long time ago, but are still in denial. And in the high range, the IBM and HP can beat them out in all categorises.

    • Sunset (Score:3, Interesting)

      by fm6 ( 162816 )
      Remember the "network computer" thing 5 years ago? Sun bought a whole string of software companies so they could have a head start on the necessary Java applications, only to shut them down when the NC market didn't materialize. In some cases, these companies went away only a few months after being acquired.

      Scott M. keeps making expensive blunders like this, but nobody seems to hold him accountable. Very disturbing.

    • The Opterons are priced very competitively, have you seen the prices? These things are fast and cheap.

      Sun's Opteron servers will fill the midrange, and even low-midrange, quite nicely.

      • Sun's Opteron servers will fill the midrange, and even low-midrange, quite nicely.

        Hmm. I have trouble believing that sun can sell opterons cheaper than traditional x86 vendors--they have terrible distribution channels and a lot of overhead. So if there are two opterons, one expensive one from sun and one cheap one from someone else (and the cheap one probably has a better warranty and cheaper support costs) why would I buy the sun version? This is why sun isn't going to be able to compete. They're currentl

        • What flaws have the distribution channel got exactly then?

          Sun are doing fine with the pricing of the Xeon based kit they sell. Why do you assume that Sun are randomly going to sell extra expensivie Opteron kit and expect people to buy it?
        • I don't know about distribution channels and prices, but as far as the previous comment about Sun not having any mid-range product, the Opteron fits that bill.

          Opteron is no Itanium, it's priced to compete in the x86 world.

          I sure hope it takes off. A lot of the types of applications I run would benefit greatly from 64-bit computing.
    • I disagree: the Cobalt folks knew how to do two things Sun (and Dec, SGI, IBM, etc) didn't know how to do:
      • sell equipment with a low margin in quantity one, and
      • build miniturized appliances.

      With a tight economy, the first is a basic competance any company needs. Sun and DEC used to be quite bad at that, back when I bought both.

      The second was a gateway to the current 1U and blade servers, which are cheaper than Dell, the usual low-cost-leader.

      --dave (biasd, you understand) c-b

    • Why should the Sun set? Just because new technology comes out doesn't mean that a company can't adapt and use it, hence the collaboration with AMD to product Opteron based servers, the existing Xeon based machines, the UIIIi CPUs for one to four CPUs and the continuing development of UltraSparc for the really high end. Don't forget too the thousands of commercial apps for Sparc that aren't available for any Opteron based platforms. In the high range IBM and HP can certainly not 'beat them out' in all cate
    • >
      the Sun just needs to set. Now that 64 bit Opetrons are out, they will have almost nothing to offer in the midrange.

      RISC is still a superior architecture. Maybe Sun won't ever adapt to higher volumes, but it is not clear that x86-64 is the answer to RISC.

  • not a surprise (Score:5, Interesting)

    by USAPatriot ( 730422 ) on Sunday December 28, 2003 @12:52PM (#7821990) Homepage
    Anyone recognize the niche as that of Cobalt, before Sun took them over? Did those do well enough that this can be popular?

    Unfortunately, as an operator of a Cobalt RaQ for many years, I found it to be very limiting once we did figure out how to really use it and how little the custom interfaces allowed, but it was great for people who just wouldn't learn that stuff.

    I hope no one thinks these are patch-proof though,. Our Cobalt needed patches and even with them had trouble avoiding a few compromises since patches were so delayed. Now it runs Debian and I couldn't be happier with the little box.
  • The cobalt raq3 and 3i used AMD K6-2 350mhz i586 chips and the raq 4 used a K6-2 450. It would seem that sun is just re-kindling old business partnerships held between cobalt corp and AMD (before sun bought cobalt).

  • Sun has released all code under BSD license (Score:4, Informative)

    by Anonymous Coward on Sunday December 28, 2003 @12:58PM (#7822002)
    Sun has been very generous and released ALL the code from the Qube 3 and now the RaQ 550 under BSD license. See open.cobaltqube.org [cobaltqube.org] for more info.

  • Just a clarification (Score:1, Informative)

    by weebler ( 661013 )
    Poster says "in favor of the AMD based Sun Fire line"
    This means some products in the Sun Fire range, with Opterons. The poster's line sounds like all SF products will be sold with Opterons and the UltraSparc will be EOLd -- Not the case! You wont see a SF15k with Opterons any time soon ;)

  • Cobalt was dying anyway (Score:3, Informative)

    by VJoseph ( 79722 ) on Sunday December 28, 2003 @01:12PM (#7822037) Homepage
    Netcraft has some information about a decline in the number of sites running on Cobalt servers, and about Sun discontinuing them.

    http://news.netcraft.com/archives/2003/12/19/sun _d iscontinuing_cobalt_linux_servers.html

    It's kind of sad that they puchased Cobalt for $2 billion, not too long ago, and now they're discontinuing the Cobalt line. That's $2 billion down the drain. When Sun is making business decisions like this, it's hard to image them being a major force in the computer industry for much longer.

    • Re:Cobalt was dying anyway (Score:1, Informative)

      by Anonymous Coward
      2 billion in stock, when the stock was $60. Now it is $4. so they "paid" how much, again? Less than the capitol they acquired from Cobalt. Itls Cobalt that got the short end of that one.
    • It's kind of sad that they puchased Cobalt for $2 billion, not too long ago, and now they're discontinuing the Cobalt line. That's $2 billion down the drain.

      An anon coward pointed out that it was $2B in dot.com stock that is now worth $150M .bust. Furthermore, what Sun got was Cobalt's market -- the ability to walk SUN salespeople into the offices of every company that ever bought a Coboalt, and up-sell them sun-blades and *86 boxes that are roughly aimed at the same market as the cobalts were.

      They a

  • Oh yes, I'll buy another Sun.... (Score:5, Interesting)

    by carndearg ( 696084 ) on Sunday December 28, 2003 @01:18PM (#7822065) Homepage Journal
    Quite a few of my customers have used Cobalt Raq servers in hosting facilities. In my view they were a fantastic product, offering a very useful Linux box to me the developer and one of the best www based admin frontends around to my non-tech-savvy customers.

    I know that Sun paid well too much for the company and that perhaps in a post-dotcom culture the market for server appliances may have contracted somewhat, but it surprises me that there was aparently no money to be made from selling Cobalts. I have met more than one hosting provider desperate to source more Raqs over the past year.

    In my view Sun have damaged their reputation in my sector of the marketplace. Fair enough they're dropping the range, so I guess they expect customers to be happy to migrate to equivalent Sun kit. But how can I trust to buy a replacement Sun brnaded server from a company whose idea of support for a range of web server appliances was to stick with PHP 4.0.6, a rather aged piece of software that simply doesnt run everything these days. Leaving people like me to either compile our own or scour the web for install-and-pray packages would be fine for a geeks-only free distribution but is not what you expect from a product you pay good money for.

  • I wonder what my hosting provider [simpli.biz] plans to do about this. I should ask them [slashdot.org].

  • Another info point (Score:2, Informative)

    by Anonymous Coward
    There's more in this eWeek article [eweek.com], especially stuff implying that the Cobalt acquisition might not have been all peaches and cream for Sun.
  • the SunFire line is not only comprised of AMD based x86 machines. Mostly it is SPARC machines, but the first x86 SunFire was the v60x and the v65x. Both are based on Intel Xeon DP chips.
  • screaming, while i clutch the pizza box! they took commodity hosting to a great level.
  • Umm, Sun is very actively selling their new Xeon based systems currently as well. I actually just installed a V65x last Monday in our cluster.

    Of course we won't mention that these machines are just OEM'ed from Intel... I know this because I bought the equivilent of a V60x directly from Intel this fall... for much less then what Sun charges. :)
    • The first Opteron servers that Sun announced are also OEMed. They showed off two servers a couple months back, one was a Newisys 2100 (1U, 2P) and the other a Newisys 4300 (3U, 4P). You can already buy the 2100 from a number of other companies (Racksaver, Appro, Angstrom, and probably a handful of others).

      I haven't seen anyone selling the Newisys 4300 server yet (most companies selling 4P Opteron servers are using the Celestica model instead), but they may arrive by the time Sun starts shipping theirs.

  • Good Bye....A rant :-) (Score:3, Interesting)

    by MegaHamsterX ( 635632 ) on Sunday December 28, 2003 @05:48PM (#7823540)
    After admining many of these machines and becoming an expert of sorts, I can rebuild one of these after an intrustion in a few minutes and have it patched.

    I know all about them .pkg files, I have built and released quite a few internally for customers in need of patches now, not when Sun/Cobalt felt like it.

    I know how to get a borked interface working again, all the tell tale signs of an exploit, placating customers as they plead and ask why their was an intrusion as they patch it the minute Cobalt releases a patch.

    The hardware in the raq3 and 4 servers look like a modified laptop design minus video.

    Actually I'm probably wrong about this, but laptops have better performance for the same spec processor.

    On the units with SCSI why are the drives IDE?

    What exactly is the PCI slot for?

    I have seen so many fail right out of the box, sometimes 2 out of the carton of 5 with the rest failing over the next 6 months.

    The perl scripting was totally horrid, the web interface runs as root, why isn't dns in the postgres database, why does it have it's own unique flat file.

    All the commands and tecniques I used were unsupported, the backup through the web interface was broken for sometime before they fixed it, tho I fixed the mangled backup and made them work anyway. These machines were unsupported if you wanted them to actually work correctly, the interface fell short in so many areas as to be useless. Let's not forget the main webserver authenticating through PAM by default......why??

    I can go into many more reasons why I hate these machines, they certainly don't fail safe, fill the disk up with logs and watch as the machine borks all of it's conf files.

    Bad engineering all around.

    I am glad to see them go, while Sun may not be perfect, these little bastard appliances gave Sun a black eye in my view.

    I thought Sun might be able to put them back on track, they did by disco'ing them.

    A Cobalt rep (pre Sun) paid us a visit to show fail-over in a demo....it just failed...I asked her if they were designed in someone's garage, she said basicly yes.....2 Billion dollars later this realization hits Sun.
    • Your story about Cobalt shipping DOA boxes reminded me of my experience with a dead Cobalt server years ago...

      The first 2700WG (the original Qube model) I ordered arrived DOA, so I went through the RMA process and returned it for a replacement. When no replacement showed up I called them again and the Cobalt rep told me they'd credited my Amex card for the purchase price (I'd returned it for a replacement, not a refund). I told him I wanted a replacement server, so he sent one out, but they never charged

  • Qube 3 Sourcecode (Score:2, Informative)

    by mcbridematt ( 544099 )
    The Qube 3 sourcecode was released to the Cobalt Users Group of Japan at open.cobaltqube.org [cobaltqube.org] (down at the moment) :(

    What a sad ending. I am still drooling over this sexy Cobalt Qube 2 advertisment [dhs.org]
  • Having been in a position where I was required to work on cobalt servers, I can only say, its about damned time.

    In their day, I am sure that these little toys were great for a small business looking for a simple end to end solution, and I must admit that the user interface for hosting customers was great, not to mention the fact that a rack of the Cobalt RaQ servers just looks damned cool in a dark server room, but, they were and are an incredible PITA to work on.

    Upgrading the software was difficult, inst
  • When they came out, pretty cool. One shop I worked at had one. Made it simple for the non-UNIX people to administer (pretty much everybody but me) but if you poked around at all under the web GUI you broke stuff. And you waited forever for patches, a security hell mitigated only partially by the fact that our Qube was a MIPS (a lot fewer script-kiddie hacks for anything non-x86). Our boss said not to compile stuff for it, void warranty, yadda yadda.

    I still wanted a MIPS one for home, once NetBSD got po
  • With the exception of building out a linux box and manually configuring the myriad of services that are integrated into a Qube does anyone know of some good, simple to use, alternatives to the Qube 3? I have one of these in a small business now and would like to swap it out for something better but I am having a hard time finding something that would work. Any leads?

Slashdot Top Deals

"Protozoa are small, and bacteria are small, but viruses are smaller than the both put together."

Close