ReleaseEngineering/BuildFaster: Difference between revisions

 
(15 intermediate revisions by 3 users not shown)
Line 36: Line 36:
| style="background:#DDD" | '''Status'''
| style="background:#DDD" | '''Status'''
|-
|-
| ??
| {{bug|772446}}
| Use sdwilsh's notification bot to help devs more easily watch the tree
| Move all Linux builds to AWS
| harth
| jhopkins
| {{StatusHealthy|status=Mostly ready, needs polish, harth taking over}}
| {{StatusHealthy|status=All Firefox builds are there now, just waiting on Thunderbird to catch-up.}}
|-
|-
| ??
| {{bug|758624}}
| Profile mochitest plain
| Purchase new machines for win/linux testers that are not tied to aggressive obsolescence schedule
| jgriffin
| coop
| {{StatusHealthy|status=COMPLETE}}
| {{StatusBlocked|status=Underway. Various test OSes are being setup and validated.}}
|-
|-
| n/a
| {{bug|784913}}
| Get data on test run time for each of the moth individual tests
| migrate linux xpcshell testsuite to AWS
| ctalbert
| jmaher
| {{StatusHealthy|status=COMPLETE}}
| {{StatusBlocked|status=Blocked on panda work}}
|-
|-
| ??
| {{bug|784913}}
| Purchase new machines for win/linux testers that are not tied to aggressive obsolescence schedule
| migrate linux xpcshell testsuite to AWS
| zandr
| '''unowned'''
| {{StatusBlocked|status=Underway, getting gear for lab testing}}
| {{StatusBlocked|status=Needs an owner}}
|}
|}


== Long Term Ideas ==
== Long Term Ideas ==
* {{bug|657738}} - Automatically determine when oranges happen and auto-star them (no owner/decision on steps yet)
* {{bug|657738}} - Automatically determine when oranges happen and auto-star them (no owner/decision on steps yet)
* {{bug|630534}} - (related) Host tree status outside of tinderbox
* Windows builds on AWS are still too slow, but we can revisit this regularly
** This would help us move away from tinderbox and would free up many of the constraints we currently operate under
* Cross-compile Mac builds on AWS
* Analyze using VMs for "dial-up" capacity
** after legal review, *yes* we think we can do this. Need to find a developer owner to give this a try.
** Would need engineering resources to debug oranges that occur only on vms
* {{bug|772579}} - Can we run unittests on AWS for Linux (and eventually Windows)?
** Would need to run some low numbers of vms on an ongoing basis so that we continue to ensure they are providing results we can trust
** We may need to solve {{bug|617763}} so that these machines can be kept up to date (whether they live in the build VPN or not, i.e. if we use EC2 or something like that...).  Either way we must address the problem that they could be out of date between one dial-in and another dial-in.
* Analyze methods of understanding and optimizing what we run on a per-checkin basis
* Analyze methods of understanding and optimizing what we run on a per-checkin basis
** Test code related to the current patch?
** Test code related to the current patch?
** Run full tests only every x pushes or y time (whichever occurs first).  Between those only run a set of tests related to the patch at hand. (Would need developer override and on-demand testing ability)
** Run full tests only every x pushes or y time (whichever occurs first).  Between those only run a set of tests related to the patch at hand. (Would need developer override and on-demand testing ability)
* less frequent debug builds
* less frequent debug builds
* use an optimized xpcshell + httpd.js to run debug tests (depends on {{bug|669953}})
* use an optimized xpcshell + httpd.js to run debug tests (depends on <s>{{bug|669953}}</s>)


= Data we need =
= Data we need =
* Profiling on all major harnesses
* Profiling on all major harnesses
** Currently profiling is being done on mochitest harness and tracked [http://brasstacks.mozilla.com/testperf_dashboard/#/mochitest here]. Be sure to select either "opt" or "debug" builds, since choosing "any" is probably not what you want (profiling times differ greatly between the two).
* How many 10.5-only test failures have there been?  Do we need to run 10.5 tests on-checkin?
* How many 10.5-only test failures have there been?  Do we need to run 10.5 tests on-checkin?
* Where are we spending our time during build slave setup/teardown (catlee owns)
* Where are we spending our time during build slave setup/teardown (catlee owns)
Line 89: Line 88:
* Clint Talbert (:ctalbert)
* Clint Talbert (:ctalbert)
* Will Lachance (:wlach)
* Will Lachance (:wlach)
* Some as-yet undefined set of the A-team (probably Ted, Jmaher and Harth, but I'll put out the call today)
* Ted Mielczarek (:ted)
* Chris Cooper (:coop)


= Definitions =
= Definitions =
Line 101: Line 101:


= Meetings =
= Meetings =
* [[ReleaseEngineering/BuildFaster/Meetings/2012-10-11|2012-10-11]]
* [[ReleaseEngineering/BuildFaster/Meetings/2012-06-28|2012-06-28]]
* [[ReleaseEngineering/BuildFaster/Meetings/2011-10-26|2011-10-26]]
* [[ReleaseEngineering/BuildFaster/Meetings/2011-09-28|2011-09-28]]
* [[ReleaseEngineering/BuildFaster/Meetings/2011-08-31|2011-08-31]]
* [[ReleaseEngineering/BuildFaster/Meetings/2011-08-31|2011-08-31]]
* [[ReleaseEngineering/BuildFaster/Meetings/2011-08-17|2011-08-17]]
* [[ReleaseEngineering/BuildFaster/Meetings/2011-08-17|2011-08-17]]
canmove, Confirmed users
2,850

edits