A timeline? [was Re: Scheduling the Bugzilla Upgrade]
- From: Owen Taylor <otaylor redhat com>
- To: Max Kanat-Alexander <mkanat everythingsolved com>
- Cc: gnome-infrastructure gnome org
- Subject: A timeline? [was Re: Scheduling the Bugzilla Upgrade]
- Date: Fri, 14 Aug 2009 10:59:04 -0400
On Wed, 2009-08-05 at 05:25 -0700, Max Kanat-Alexander wrote:
> I was talking with bkor just now, and I think that we should schedule
> the Bugzilla upgrade for August 15-16. I will have a feature-complete
> test installation containing all of GNOME's data by this weekend or
> earlier, and I think 7 days is enough testing to shake out any critical
> issues before we go live.
Do you have a detailed plan of how you see this happening?
- What needs to be done before we start the move
- When exactly do we start the move?
- What are the major steps of the move?
- Do we want to do validation before we open up the new instance
to the world?
- What cleanup has to be done afterwards?
So far, as prep I've:
- Increased the configuration of bugzilla-web from 2 VCPUs to
6 VCPUs [*]
- Set up backup for bugzilla-web and drawable
- Dropped the TTL for the gnome.org zone file to 30 minutes
Are there are any other prep steps needed from the sysadmin side?
- Owen
[*] My understanding is that Xen VCPUs are both a limit (a VM
with 6 VCPUs will never use more than 6 real CPUs) and a sharing
factor if there are more VCPUs than real CPUs. I have
git.gnome.org set up with 4 CPUs and bugzilla-web set up with 6
at the moment.
[
Date Prev][
Date Next] [
Thread Prev][
Thread Next]
[
Thread Index]
[
Date Index]
[
Author Index]