Closed Bug 827475 Opened 12 years ago Closed 10 years ago

Chief for AUS2_PRODUCTION deployments

Categories

(Infrastructure & Operations Graveyard :: WebOps: Product Delivery, task, P1)

x86
macOS

Tracking

(Not tracked)

RESOLVED WONTFIX

People

(Reporter: nmaul, Unassigned)

References

Details

(Whiteboard: [2013Q3] [triaged 20130111])

According to bug counts, this is one of the most-frequently-pushed things... and frequently a blocker bug, at that. Let's figure out how to get us (IT/WebOps) out of the way on this.

Chief is the normal method for non-IT-driven deployments, so that's where I'd like to start looking on this. It gives you a big red button to push. We've only ever used it with github-based projects, but I see no obvious reason why it shouldn't work for SVN or even CVS repos just as well, since we have to tell it how to update anyway.

One obvious problem is that the current aus2/aus3 site is managed by ip-admin02, which is not really ideal for running Chief. It would be better if we can move the current prod site over to the newer AUS cluster, running alongside the current dev/stage environments for aus4 (there should be zero concerns with running them side-by-side). That's obviously a bit bigger project, but worthwhile in its own right, IMO. If we have a bug for that already, it should block this one. If not, we should probably make one.
Priority: -- → P3
Whiteboard: [triaged 20130111]
Assignee: server-ops-webops → bburton
Priority: P3 → P1
Blocks: 854251
No longer blocks: 827462
This is being put on hold for a minute while we look at how things will go with getting aus4 into prod this Q
Assignee: bburton → server-ops-webops
Whiteboard: [triaged 20130111] → [triaged 20130111][waiting][aus4 prod timetable]
FWIW, we're not planning on having aus4/Balrog code complete and everyone switched over this quarter. Maybe part of the way there, but release channel people are very likely to still be on aus3.
Whiteboard: [triaged 20130111][waiting][aus4 prod timetable] → [2013Q2] [triaged 20130111][waiting][aus4 prod timetable]
Ok, given that we'll discuss how we can allocate time for this, during our meeting on Thursday, as setting up Chief requires we move the admin node to ausadm and get all the prod web servers using it.

ip-admin02 (where current pushes happen) is RHEL5 and we can't install Chief on RHEL5

Thanks fr the update
Whiteboard: [2013Q2] [triaged 20130111][waiting][aus4 prod timetable] → [2013Q2] [triaged 20130111]
Whiteboard: [2013Q2] [triaged 20130111] → [2013Q3] [triaged 20130111]
Component: Server Operations: Web Operations → WebOps: Product Delivery
Product: mozilla.org → Infrastructure & Operations
The plan is now that we'll abandon AUS2/3 and focus only on AUS4, so this will die naturally in the next few months... only a couple more releases to go. :)
Status: NEW → RESOLVED
Closed: 10 years ago
Resolution: --- → WONTFIX
Product: Infrastructure & Operations → Infrastructure & Operations Graveyard
You need to log in before you can comment on or make changes to this bug.