Slashdot Mirror
Solaris No Longer Free As In Beer
rubycodez writes "Oracle, having acquired Sun Microsystems, including its Unix, will no longer give away free Solaris licenses. Oracle also states that some features of its Oracle Solaris will not appear in OpenSolaris, which means OpenSolaris may start to die."
For trying to get people to want to use the OS, Sun and Oracle sure like to piss people off.
Oracle just seems to make it more pronounced.
Twitter supports and protects racists - by smearing their critics with the "Hate Speech" label.
An honest question from someone who has never been involved in OSS development: how 'different' does a Linux distribution have to be in order to count as a separate branch? Is someone allowed, for example, to take the current release of Solaris, remove anything Oracle may own the rights to (does that include code? just graphics?) and redistribute it?
Where is the line drawn, legally, in the OSS community?
Will. Oracle is not in the business of giving stuff away for free.
Have you heard? They license their database software not by the servers it runs on, nor by the processor, but by the core. How absurd is that? Does it cost them more to produce a database that works on more than 4 cores, or to support it? Believe it or not, they also charge extra for installed memory, as if that had anything to do with their production or support costs. Failover? Now you're into serious money. And don't you dare run it on stuff that's not on the secret list, or your support contract is invalid.
If Cisco's motto is "that feature is enabled through the purchase of an optional license", Oracle's is more so.
I guess Oracle doesn't get that we have options, and the pace of hardware technology will quickly erase any software advantage they think they have.
Help stamp out iliturcy.
that thing has been dead for years. Which is a huge pity because solaris and sun's hardware was some sweet gear.
If you mod me down, I will become more powerful than you can imagine....
What makes you think that Solaris's death by neglect is not part of Oracle's plan? Milk those who are locked-in to Solaris for as long as possible and for as much money as possible, while putting the least possible resources into it. Classic corporate-raider tactic for medium term (3-5 yr) returns. Strip the assets for as much as you can squeeze them for, then sell the trademark name of the carcass off to the highest bidder.
mdb, the solaris modular debugger, that's what I will miss the most, it's not a product (comes with solaris) but there just is no open source equivalent. People that tell you otherwise have never run into a problem that was too much for truss or dtrace but one where gdb simply did not work or got in the way.
Clearly, very few people here have any enterprise-level Solaris experience. In terms of stability and performance I compare Linux to Solaris like you compare Windows to Linux. Well, this may be too harsh but this is mostly addresses to the fat dorks on Slashdot screaming "death to Solaris". The biggest file server guys like that had to support is the one sitting under their desk with all the porn on it.
When I transitioned from Solaris and AIX to supporting RH and SuSE several years ago, I experienced somewhat of a shock: servers hanging on shutdown, lousy NFS performance, Samba slowing down to a crawl under moderately heavy load and a crapload of other issues I never thought a unixoid OS can suffer from. All these problems coupled with consumer-grade hardware and what you get is one big, never-ending downtime. Something is always down or barely limping along.
There were times when all our servers were running Solaris, AIX or HP-UX. I could come to work, drink my coffee, read the news, space out for a couple hours, then break for lunch, work a couple hours on some project and go home. As more and more real servers are being replaced by cheap HPs and Dells running the blasted RHEL or, worse yet, SLES, all this free time I used to have is a distant memory.
Why AC? I second you with your effort (trying fscking hard to fall in love with ZFS). I was hoping for a ground-breaking filesystem for the rest of my lifetime. After some bad crashes and loss of data (well documented in the Internets), we had to declare it a full-blown failure.
You're both right.
Generally, Sun hardware made before 2000 seemed almost "unbreakable". Since then, likely due to wildly fluctuating financial conditions, reliability has been all over the map. My (pre2k) E4000, E4500 and E450 systems never dropped. When they had any hardware problems, they'd announce it in syslog preemptively (ie. "DIMM J3201 correctable errors blah replace module").
I ordered about a dozen E420R's around 2001. E420 and E220 servers made 1999 and later had problems specifically with their stupid "memory riser boards". Each board had two torque screws that needed to be tightened precisely. If they were even slightly under-tightened (as many were from the factory), vibration would eventually shake the connectors slightly loose, causing intermittent system freezes.
A hard-freeze with nothing in the logs was almost unheard of, but that's why we have serial console logging, right? Getting nothing off the serial console post fail, now that was novel. Back then, Sun field engineers were a great bunch, but they were as in the dark on the issue as we were, performing the same checks, making sure everything was clean and properly seated. The same systems would fail a week later.
Eventually, I noticed that the memory board screws on a failed system that I had tightened had actually gotten looser. This led to a simple solution: over-tighten them. Granted, this isn't really an acceptable solution on systems sold for $30k+, and it was inexcusable that Sun didn't vigorously inform its customers (or employees for that matter).
rogerd, Are you talking about E420 rackmount systems that you have now? If so, then I'd say the exception proves the rule, as you're talking about 10 year old systems, over double the 5 year MTBF advertised (lie) for PC hardware. Try bunging the screws next time one goes haywire. I've never played with an M5k, personally. If that's true, (bad) SPOF fan, (worse) faulty temp sensors that (horrible) failed to save the hardware, that's, well, kinda sad.
/* MAGIC THEATRE
ENTRANCE NOT FOR EVERYBODY
MADMEN ONLY */
Yes, Linux distributions will release new packages.
*Because* Windows has no packages, it can't do the same.
So you are seriously arguing that those huge downloads from Microsoft are actually binary diffs? They are not.
I cannot say where I work but yes, we use a lot of Solaris. Quite a few people who attend the various USENIX conferences use Solaris. Especially 10 works like a champ, and the Zones/ZFS/FSS/DTrace bundle alone is pretty nice.
But yeah, we use Solaris on 75+ boxes doing a lot of work with my country's law enforcement agencies. A lot of it's critical stuff and in our heterogeneous environment (we use Linux, Windows, and Solaris, each in a different capacity) of several hundred systems the 75+ Solaris boxes are the *only* ones without issue. That either means I'm a god-like admin (especially considering some of the setups I've had to build and maintain...they are NOT easy) or Solaris is a durable, solid OS worthy of enterprise-level use.
And I personally think I'm far from a god-like admin.
"Just a fox, a whisper."
"Linux distributions release entire new packages - not patches."
Uhm. Have you ever looked inside Windows patches? They are in essence what Debian Stable updates are. I.e. new versions of affected software.
They are definitely NOT patches in original sense (i.e. a diff between two versions).