Infrastructure People,
In the interest of maintaining good will with our mirrors and appreciation for the great service they provide to Fedora... I'm looking for some guidance and preferences from the Infrastructure team on how to go forward.
During the schedule planning for Fedora 13, Jesse suggested looking at the other releases to see where our final date was landing. Jesse dug up the dates and I add my own editorial comments (see below)
Thanks, John
Here are the questions as it relates to the mirrors that carry Fedora and other distros:
1) What is the minimum distance in days that we have to put between our release and other distros?
2) Is it the release date that matters most or the date the staging starts?
3) On a scale of 1 to 5 (1 = doesn't matter, 3 = moderate impact, 5 = critical, we'd be out of our minds not to) how important is it that we bend our schedule to not land as close as we might be potentially landing with the dates below (assuming we do not slip)?
I'm really looking for a scale to get a sense of the severity and not subjective responses like "it will be bad" or "people will hate us." I want to know "how much."
4) Our current options as I understand them are: a) Overlap closely (see below) b) Add two weeks to Fedora 13
Is there an "option C" that we could implement or go with so as not to have to add two weeks to our scheduled GA date if the consensus score from #3 is high?
------------------------------------------------------
Debian 2010-March (Freeze, not an actual release date)
Fedora 13 2010-04-27 (since Fedora 8 every release has been late >= 2 weeks)
Ubuntu 10.4 2010-04-29 (I'm told they have never slipped)
OpenSUSE 2010-05-05 (no idea on their "on time arrival history")
On Wed, Nov 25, 2009 at 09:48:56AM -0800, John Poelstra wrote:
- What is the minimum distance in days that we have to put between our
release and other distros?
The mirrors spike for about 2 days at the release of Ubuntu and Fedora.
Ubuntu 9.10 released on Thursday, Oct 29. Looking at the bandwidth logs from mirrors.kernel.org, they spiked through Thursday, but by Saturday were back to normal loads.
openSuSE 11.2 released on Thursday, Nov. 12. They peaked on the 13th, and were back to normal by the 15th.
Fedora 12 released on Tuesday, Nov. 17. Due to DNS problems at kernel.org, they didn't actually bitflip and start serving users until late the 17th, but peaked and were back to normal by Saturday.
From this, I'd argue that the minimum distance is 3 days, though both
the Ubuntu and Fedora have higher demand on these mirrors than openSUSE (we could probably move closer to openSUSE if we had to). 5 days would be plenty.
- Is it the release date that matters most or the date the staging starts?
It seems we can stage to our mirrors and have openSUSE release at the same time without problem. I doubt that's true of Ubuntu though - I wouldn't want their release spike to coincide with our staging.
- On a scale of 1 to 5 (1 = doesn't matter, 3 = moderate impact, 5 =
critical, we'd be out of our minds not to) how important is it that we bend our schedule to not land as close as we might be potentially landing with the dates below (assuming we do not slip)?
Depends who we collide with. 5 for ubuntu, 1 for anyone else.
Also, it's only a problem for mirrors carrying both. I don't know how many that is (MM doesn't track ubuntu URLs). our Tier 1s certainly, and maybe half of our Tier 2s.
I'm really looking for a scale to get a sense of the severity and not subjective responses like "it will be bad" or "people will hate us." I want to know "how much."
- Our current options as I understand them are: a) Overlap closely (see below) b) Add two weeks to Fedora 13
Is there an "option C" that we could implement or go with so as not to have to add two weeks to our scheduled GA date if the consensus score from #3 is high?
Debian 2010-March (Freeze, not an actual release date)
Fedora 13 2010-04-27 (since Fedora 8 every release has been late >= 2 weeks)
Ubuntu 10.4 2010-04-29 (I'm told they have never slipped)
OpenSUSE 2010-05-05 (no idea on their "on time arrival history")
pull F13 in a week, or push 2 weeks. Both falling the same week would cause slower delivery to both our sets of end users.
infrastructure@lists.fedoraproject.org
-
John Poelstra -
Matt Domsch