On 24/04/10 01:18 AM, Dennis Clarke wrote:
On 04/23/10 10:04, Dennis Clarke wrote:
Should I then assume that the "tonic" labeled BFU ready-to-run tarballs
are gone? Those lovely non-debug tarballs are also gone? That the people
that were doing those lovely community type services have been taken out
into a field and buried?
... Really? Is this necessary?
Given recent events, yes.
You are being entirely too melodraomatic and fear driven.
We (the people you suspected of being buried in a field) are very
definitely still here. We are still doing the tasks we were doing
before Oracle started paying us instead of Sun.
We are, however, going through a period of immense organisational
change, and learning how our new corporation operates. For that
matter, Oracle is learning about us, too. It goes both ways.
As another comment to this thread has reminded you, BFU is going
away. Soon. This is part of the logical conclusion that is moving
ON to generating nightly IPS repos.
In the realm of things changed, this is a Good Thing (TM). The Mercurial
repository should be considered the canonical source for retrieving
source-based artifacts for ON. Period.
Its a trivial matter to re-create the source tarball contents via hg.
Tags have been in place for years now in the event you need to go back
to previous releases.
Explain on-closed-bins-nd.sparc.tar.bz2 to me please.
I really don't understand what you think needs explaining.
The on-closed-bins(-nd).$MACH.tar.bz2 are, as they always have
been, a tarball containing the closed binaries that you need
in order to actually build ON without have usr/closed present.
Yes, in the last few months the "nightly" version has been
generated in an ad-hoc fashion, due to the formerly tedious
way in which they were generated (build with closed, save
minimum necessary files, build only with open source, then
do a diff of the proto areas, tar up those binaries).
I pushed a changeset last Thursday which makes that easier
to do on a regular basis, and once I've rejigged some of the
other gatekeeping glue infrastructure (and dlc.sun.com comes
back online after its physical move), then we should be able
to have those closedbins being pushed out on a nightly basis.
The per-build closedbins (ie, for snv_136, snv_137, snv_138
etc etc), are all being built on the same schedule as before,
which is "after we've done all the other bits we need to get
the build to RE."
As others have pointed out, also, if you want to know what
goes into a respin of a build, then you should subscribe to
on-disc...@opensolaris.org, and look for the emails which
the gatekeepers (mostly me so far) send which provide that
information. For reference, here is the announcement I made
about it in November last year, and an example of the sort
of information you can expect to see in a respin email:
http://mail.opensolaris.org/pipermail/on-discuss/2009-November/001342.html
http://mail.opensolaris.org/pipermail/on-discuss/2009-November/001377.html
These respin informational emails do not contain information
about what goes into an OpenSolaris _Release_ (eg, 111b or
101a).
Now, if you want help with how to setup and run a Mercurial-
based ON workspace please ask specific questions.
James C. McPherson
--
Senior Software Engineer, Solaris
Oracle
http://www.jmcp.homeunix.com/blog
_______________________________________________
opensolaris-code mailing list
opensolaris-code@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/opensolaris-code
