So I think I may have been in mourning last week. A little, at least. I’m going to miss Sun Microsystems.
I wrote a big post full of doom and gloom for the Oracle acquisition. I speculated that this was the end of all that was good with Sun, all the extremely positive open source work it’s done, the end of its fantastic R&D efforts.
I predicted the death of MySQL. I wrote all that, but I didn’t file it. Maybe I thought that if I didn’t, it wouldn’t come true. I truly hope it doesn’t. Time will tell.
Instead, I’ve decided to post some memories of Sun over the years, since working with (and occasionally swearing at) Sun hardware and software was a large part of my formative years in IT.
I started waaaay back with the first iterations of FreeBSD and (later) BSDi, the Internet SuperServer. I became well versed in the BSD way of doing things, and when I chanced to work on some SunOS systems, there was little learning curve. Then came Solaris.
Logging into my first Solaris system was like taking a walk in bizarro world. ps auxww had no meaning here. Bash wasn’t an option. I quickly found sunfreeware.com and later blastwave.org. To this day, I still modify my Solaris systems to be more BSDish and less SysVish. It’s all in the foundation, I suppose.
I recall my first day working as a consultant at a major financial company, sitting down at my new workstation, which was a Sun Ultra 2 with 256MB of RAM (!!) and three 21-inch CRT monitors running Solaris 2.4 and CDE. I was blown away.
One of my first big Solaris projects was to build and deploy a large-scale NIS+ infrastructure to a 10-city WAN. At the time, NIS was viewed as legacy and NIS+ was the way of the future. That might have actually been the case if NIS+ wasn’t so amazingly convoluted and difficult to administer. The decision was made to convert from NIS to NIS+ across the entire network, and each site was to get their own pair of NIS+ servers, linked to a master running at headquarters.
I learned far more about NIS+ than any human should ever really know and built the whole thing using SparcStation IPX boxes, which were the odd Mac Mini-looking boxes Sun produced in the mid-nineties. They didn’t have much horsepower or RAM, but for these purposes, they fit the bill. My cube had 20 of these little boxes, stacked in columns of five, and I spent the better part of three weeks building them, configuring them, and labeling them for shipping to each site.
When they got there, they’d get plugged in, and theoretically Just Work. And they did. For nearly a month. At that time, a new admin on the site decided that he needed to rebuild the Kerberos keys on the master NIS+ server at HQ. Naturally, he didn’t tell anyone about this, and of course, he didn’t really need nor want to do this, since as soon as he did, the slaves at each office started failing when they tried to update their maps.
As soon as I realized what was happening, the race was on to salvage the maps before they were completely lost. I managed to pull raw copies of the maps from the last pair of slaves that hadn’t yet tried to update the maps and spent 24 straight hours rebuilding the whole authentication infrastructure. Good times.
Then there was the day that the first shipment of Sun E450s arrived. They were squat boxes that looked like they were made of Legos and packed a huge punch for the time. I recall conversing with a Sun engineer over dinner about the curious case of the multi-million-dollar Sun E10Ks that were inexplicably breaking at odd times, segfaulting all over the place for no rhyme or reason.
After many months of troubleshooting and general puzzlement, some bright engineer covered the RAM with aluminum foil, and the problems went away. Apparently, the shielding wasn’t sufficient and gamma rays were randomly flipping bits in active RAM. Talk about a non-obvious solution.
Burned into my memory is the 16-straight-hour effort I put in to attempt to salvage a Solstice Disksuite software RAID array that had fallen prey to an overzealous and underequipped admin. Naturally, there were no backups. Naturally, this was a mission-critical Oracle database installation. I was finally able to reconstruct the databases, but it wasn’t pretty.
I remember buying my first Sun workstation, an Ultra 10 with a 440MHz UltraSparc CPU with 512MB of RAM with a Creator 3D graphics card. I ran Solaris on it for a while, then put Red Hat Linux 6.2 on it. It was my main workstation at the turn of the century.
I also had my share of Sun migrations — away from Sun. Back in the dark days of the early 2000s, more than a few companies were looking at the cost of their Sun contracts, and this up-and-coming Linux operating system, and exploring alternatives. I can recall replacing SunFire V440s and 280Rs with Dell and HP servers running Linux (unsupported at the time) and showing performance increases in the triple digits.
I clearly remember walking into a datacenter that I had just finished moving to Linux and taking a long look at all the Ultra 60s, V440s, and 280Rs lining bakers racks near the door — powered off, cold, and silent. A dozen feet away, half a rack of first-generation DL360s were humming along, running Red Hat Enterprise Linux 2.1.
I remember unboxing several Sun Netra X1s and wondering what happened to make Sun produce such a terrible server. I also remember talking with Bruce Perens about Sun and Jonathan Schwartz at LinuxWorld several years ago. We decided that the pony tail is usually found at the back of the horse, not the front.
I remember buying six Sun GDM-5410 21-inch CRT monitors at an auction for $100 each. I still think they were the best CRT monitors ever produced. I still have them. All but one still work flawlessly.
I remember my surprise at Sun’s announcement of the first multicore Sparc chips and the introduction of the Sparc T1. I was also surprised at the announcement of OpenSolaris, and /usr/sfw, which contained a whole pile of open source tools, services, and applications that suddenly came with Solaris and didn’t require half a day spent pulling packages from sunfreeware.com and blastwave.org.
I remember Sun buying MySQL and thinking, “Well, I hope they don’t mess it up.” They didn’t – so far. Apparently they were waiting for Oracle to give it a shot.
I remember seeing the first Operton-based x86 servers from Sun and wondering why they were bothering – and I remember unboxing the first eval server and being completely impressed with the hardware design. It shouldn’t have been too much of a shock — Andy Bechtolsheim was back in the design seat.
I ran a Sun Ultra 40 M2 as my main workstation for several years. It is and was a fantastic workstation. When coupled with a few Sun 24-inch LCD monitors, it would make any geek salivate like one of Pavlov’s dogs.
My first look at ZFS blew my mind. To this day, I consider it to be the best file system ever designed. It can and should take over the world — as long as Oracle doesn’t screw it up. Dtrace is another example of “holy wow” engineering from Sun.
I took the Epcot Center-like tour of Sun’s Palo Alto showroom, highlighting all things Sun. From the StorageTek SL8500 running at full speed and looking like it could possibly pull a Terminator and take over the world to the SunRay laptop clients, it was terribly impressive. Speaking of SunRay clients, I can recall building a proof-of-concept SunRay implementation back when they required a dedicated VLAN and all the transport was broadcast UDP. Noisy, for sure, but oh so cool.
There are more memories, some cloudier than others. Suffice it to say that I’ll miss Sun. I just hope that Oracle understands that it’s purchased more than just a company — it now owns a huge slice of computing history and a significant part of computing’s future — and that it knows enough to not get in the way.