On 2013-06-28 17:51, Chris Murphy wrote:
On Jun 28, 2013, at 4:20 PM, Adam Williamson <awill...@redhat.com>
wrote:
I think you may be labouring under a bit of a misapprehension about
what
should be tested, here. The distinction between a TC and an RC is not
large. An RC can only happen after freeze and must have all blockers
fixed: if a build after freeze doesn't have all blockers addressed, we
call it a TC.
The distinction between TC and RC is large in that an RC1 *could* be
final release.
Yes, I did address that in the rest of the mail. I was making a general
point, not a specific one.
And since this particular bug wasn't in TC6 but was in
the very next build, RC1, I think that's consistent with suggesting
more time between an RC and a go/no-go.
If we do this, we are going to have more slips. _Considerably_ more
slips. That is the trade-off. I don't think it's one people will be
happy with. It's really that simple.
Although even better than that would be more recruitment of Mac users
to do installation testing, so that it's possible to get installation
tests done for at least RC1s. I tested it in a VM, foolishly not
testing it on baremetal. RC2 hit, and I missed that since RC3 came
soon after and immediately tested that on baremetal. But between RC3
and go/no-go was a matter of hours.
Because RC3 was RC2 with a single very precise change; we were basically
expecting to be GO on RC2 when I went to sleep on Wednesday night, then
an issue was found just after I went to sleep, and we decided to go
ahead and just fix it and respin. The change was isolated in the
anaconda TUI interface's Software Selection spoke so really couldn't
possibly affect anything else.
We do actually have more than a few people with Macs: you, Fedora QA's
Brno team, mjg59, and I think someone on anaconda team has one. It would
be nice if at least one of the above could test each *C, though I
realize bare metal testing is a PITA and I hate it myself.
If we actually blocked on Mac dual boot by policy, we would have a test
case for it, and we would not have considered releasing without having
that test case run in at least one of the RCs, FWIW.
--
Adam Williamson
Fedora QA Community Monkey
IRC: adamw | Twitter: AdamW_Fedora | identi.ca: adamwfedora
http://www.happyassassin.net
--
devel mailing list
devel@lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/devel
