migrate more buildbot macosx test machines to taskcluster

RESOLVED FIXED

Status

Infrastructure & Operations
CIDuty
P1
normal
RESOLVED FIXED
a year ago
2 months ago

People

(Reporter: kmoir, Assigned: aobreja)

Tracking

Details

Attachments

(2 attachments)

(Reporter)

Description

a year ago
The pending counts for tc macosx tests are high
https://queue.taskcluster.net/v1/pending/releng-hardware/gecko-t-osx-1010

and the bb machines are not busy
https://secure.pub.build.mozilla.org/builddata/reports/slave_health/slavetype.html?class=test&type=t-yosemite-r7

With the next uplift, mac builds and tests will run on tc on beta.  We should consider migrating some of the remaining pool of machines to tc to reduce the wait times, since macosx tests will only run on release and esr with the next one.  The current pool of machines that are not converted are
0001-0030
380-392 (some loaners in here)
Adding this to the P1 pool to tackle after mergeduty 2nd of August.
Priority: -- → P1
We currently have 42 t-yosemite-r7 machines running in BB: t-yosemite-r7-[0001-0029], t-yosemite-r7-[0380-0392].
I think it would be a good start to move 13 of them (t-yosemite-r7-[0380-0392]) to TC, then monitor the load on BB and see if more such machines can be migrated.

@kmoir: how does that sound to you?
Flags: needinfo?(kmoir)
(Reporter)

Comment 3

a year ago
Sounds great!  As a side note, I think there are some currently some loaners in that range.  At least there were when I checked yesterday.
Flags: needinfo?(kmoir)
Assignee: nobody → aobreja
Attachment #8890892 - Flags: review?(aselagea) → review+
These machines t-yosemite-r7-[0380-0392] are now in Taskcluster and generic worker is running.
Will keep monitoring for a few days before closing the bug
Created attachment 8891961 [details]
yosemite.PNG

Seems jobs are working fine,only one last thing need to be changed, the view in Slavehealth so that we can see all machines if are taking jobs.
Wander do you know how can we make the latest migrated machines to be seen as the one that were migrated before (check yosemite.png to see the difference),or should we ask Greg.
Flags: needinfo?(wcosta)
(In reply to Andrei Obreja [:aobreja][:buildduty] from comment #6)
> Created attachment 8891961 [details]
> yosemite.PNG
> 
> Seems jobs are working fine,only one last thing need to be changed, the view
> in Slavehealth so that we can see all machines if are taking jobs.
> Wander do you know how can we make the latest migrated machines to be seen
> as the one that were migrated before (check yosemite.png to see the
> difference),or should we ask Greg.

303 :kmoir
Flags: needinfo?(wcosta) → needinfo?(kmoir)
(Reporter)

Comment 8

a year ago
I just pushed a fix to change this.  The next time the page is updated it should reflect the new range.
Flags: needinfo?(kmoir)
Status: NEW → RESOLVED
Last Resolved: a year ago
Resolution: --- → FIXED

Updated

2 months ago
Product: Release Engineering → Infrastructure & Operations
You need to log in before you can comment on or make changes to this bug.