Seed Windows machines with unified Firefox repo under pooled storage

RESOLVED FIXED

Status

RESOLVED FIXED
2 years ago
6 months ago

People

(Reporter: gps, Assigned: gps)

Tracking

(Blocks: 1 bug)

Firefox Tracking Flags

(Not tracked)

Details

(Assignee)

Comment 1

2 years ago
PR submitted at https://github.com/mozilla/build-cloud-tools/pull/234.

Hopefully grenade will look at it soon.
(Assignee)

Comment 2

2 years ago
PR is merged.
Status: ASSIGNED → RESOLVED
Last Resolved: 2 years ago
Resolution: --- → FIXED

Comment 3

2 years ago
how does this go live in upcoming AMI? or is it already?

I'm seeing lots of hg-shared purge errors in release automation.


context:
14:05:33 <jlund> lot's of sad windows slaves: https://secure.pub.build.mozilla.org/builddata/reports/slave_health/slave.html?name=b-2008-spot-147
14:08:06 <jlund> wonder if https://bugzilla.mozilla.org/show_bug.cgi?id=1286430 is at play
14:09:27 <jlund> gps: I'm getting lot's windows purge errors in our hg-shared dir
14:09:35 <jlund> gps: example https://irccloud.mozilla.com/pastebin/EHs334oC
14:09:54 <jlund> this patch looks suspicious: https://github.com/mozilla-releng/build-cloud-tools/pull/234/files
14:10:21 <gps> that patch would have fixed it if AMI seeding was actually working
14:14:43 <•catlee> it needs to be done manually iirc
14:13:54 <jlund> it's not? is there a bug?
(In reply to Jordan Lund (:jlund) from comment #3)
> how does this go live in upcoming AMI? or is it already?
> 
> I'm seeing lots of hg-shared purge errors in release automation.
> 

I sent an e-mail around about this, and the AMI process is not finishing at present for some reason. but markco wrote a cloud-tools patch to help us here which does not depend on new AMIs

https://github.com/mozilla-releng/build-cloud-tools/pull/238
Component: General Automation → General
Product: Release Engineering → Release Engineering
You need to log in before you can comment on or make changes to this bug.