Re: [head tinderbox] failure on arm/arm

On Sat, Nov 10, 2012 at 4:54 AM, Brett <brett.mahar_at_gmx.com> wrote:

> On Sat, 10 Nov 2012 23:34:24 +1100
> Peter Jeremy <peter_at_rulingia.com> wrote:
>
> > On 2012-Nov-10 09:16:32 +1100, Brett <brett.mahar_at_gmx.com> wrote:
> > >Just an observation: a few years ago when I got sick of Linux's
> > >"headlong rush" development model, I subscribed to various BSD
> > >mailing lists to see what else was out there. I considered FreeBSD at
> > >the time - there was a neverending avalanche of "[head tinderbox]
> > >failure" messages.
> >
> > The Project tries to avoid it but occasional build failures on the
> > development branch are very likely to occur.  As a new user, you
> > would be much better off starting with a release branch.
> >
>
> I used 9.0 and release candidates for a couple of months beforehand so i
> would know what "usually" works and doesn't work before, trying current
> out. Compared to many of the old timers out there I guess this makes me
> very new still, though!
>
> > >This told me that I would be more likely to be running code written
> > >by people who knew what they were doing if I went with Open, Net, or
> > >DragonflyBSD.
> >
> > I think that's being unfair.  Do Open, Net or DFly have an equivalent
> > to the tinderboxes that do automated test builds and report failures?
> > And, since you have replied to an ARM failure, DragonflyBSD would not
> > be an option since it doesn't support ARM.
> >
>
> The point I was trying to make (context lost in the partial quote above)
> was not that it is better or worse than the other BSDs, but that at the
> time (maybe 3 years ago) when I was looking around to alternatives to Linux
> and reading the various mailing lists, this was the impression I got. I am
> sure other people must see these daily failures and get the same
> impression. Whether this is fair or not has nothing to do with what
> impressions people form, and what OS they subsequently decide to install.
>
> As I recall reading, the tinderbox was established due to the high
> incidence of build failures. In my original post on this thread, I was
> commenting not on the failure of ARM build in particular, but chiming in
> after Doug Brewer's request for the code to be tested before being
> committed. If anyone else had backed him up I would not have felt the need
> to write.
>

We heard your concerns loud and clear; there are several developers that
understand your concerns well and we're shooting for build stability now
(it's easy to achieve) and runtime stability as well (considerably harder
because the automated infrastructure isn't in place yet to run smoke tests
to verify that all commits are sane -- but hopefully that's coming sometime
in the near future ;)...).

So, as I said before, 1) let's come up with a proper solution that we can
pilot (even if it's a prototype) to extend tinderbox to allow developers to
toss code over the fence (branches/patches?) in order to get build results
before it hits the tree (Chuck offered the hardware.. someone needs to
offer the time to come up with a submit/receive mechanism) and 2) come up
with a proper test environment so we can avoid regressions like this as
well as the slew of clang related TB failures that are still being fixed.

Thanks!
-Garrett

PS Adrian is doing good work out-of-band (not work related), so let's work
with him in order to make his life easier -- not harder by stating the
obvious and bikeshedding him to death.