Let's make it official: Solaris 11 is closed source
January 19, 2012
You may remember back in August 2010 when there was a leaked Oracle memo that said, among other things:
At the time I noted that 'full releases' might be construed to be 'Solaris 11' instead of the next 'Solaris 10 update X' release and was unhappy about it. That was then. Now it's been a couple of months since Solaris 11 was officially released to the world and, well:
; cd onnv-gate ; hg incoming comparing with [...] searching for changes no changes found ; hg log | fgrep date | sed 1q date: Wed Aug 18 15:52:48 2010 -0600
I think it's safe to conclude that there will be no further updates to public (Open)Solaris source code from Oracle, ever. Solaris is now a closed source, 'source-not-available' operating system once again (and probably stronger than it ever was; it used to be sort of possible for universities to get Solaris source code, but I doubt that's on the table from Oracle).
(I'm sure that almost everyone concluded this some time ago. Sometimes I remain hopeful even in the face of all but certain disappointment.)
This matters a lot for us; our ZFS spares system and parts of our ZFS status monitoring system are built around information obtained from undocumented internal library interfaces because there is no other alternative. It seems extremely unlikely that we will ever upgrade to any future version of Oracle Solaris. Lack of (Open)Solaris kernel code also significantly reduces the usefulness of DTrace, one of the theoretical signature Solaris features.
(It is vaguely possible that some version of Solaris will sometime expose public interfaces for the information we need, but frankly I really doubt it. All evidence to date suggests that it is strongly against the engineering culture of ZFS; they had five years to do it and steadfastly didn't. I will skip any number of angry remarks.)
* * *
Atom feeds are available; see the bottom of most pages.