EngineeringProductivity/Meetings/2010-08-23: Difference between revisions

m
 
(4 intermediate revisions by one other user not shown)
Line 1: Line 1:
= The Boulders =
= The Boulders =
''' Android '''- ''ctalbert, bmoss, jmaher''
''' Android '''- ''ctalbert, bmoss, jmaher''
* Bear made progress on getting stuff integrated with buildbot
* Chrome package failed to land - but errors have been fixed
* Ready to land it again
* New Agent will land this week


''' Mozmill Buildbot Integration ''' - ''jhammel''
''' Mozmill Buildbot Integration ''' - ''jhammel''
* Talked to Jhford he put up another version of the UnittestPackageBuildFactory patch, currently reviewing
* Jhford is slated to land the UPBF patch next week


''' Better Performance Metrics ''' - anode, ?
''' Better Performance Metrics ''' - anode, ?
* Mostly setting up auto-tools staging environment
* Keeping bug up-to-date.  It's mostly poking holes in firewalls to access various utilities we need.
* Hoping it will be up soon - mostly IT issues left.
* Worked with jmaher to land E10S talos stuff.
* Alterations to Talos to make it work with new features around XUL permissions
* Cycle Collector test has stalled out waiting on dev's for the hooks we need; need to follow up with them.
** We can now force a cycle collection
** But we still need to disable auto-memory collection so that we know what state we're in after forcing a cycle collection
* Addons performance testing - Wil has a meeting between Alice, Harth, Nick & Wil to figure out how they want addons perf testing to take place in the future and how they want reporting to happen.


''' CrossWeave ''' - ''jgriffin''
''' CrossWeave ''' - ''jgriffin''
* No change from last week; still need to get hosted on permanent machine and add a couple of features requested by devs.
* No change from last week; still need to get hosted on permanent machine and add a couple of features requested by devs.
* Ctalbert fix up esxi host today for this


''' Firebug Test framework''' - ''ahal''
''' Firebug Test framework''' - ''ahal''
Line 16: Line 32:


''' Profile Manager ''' - ''jgriffin, jhammel''
''' Profile Manager ''' - ''jgriffin, jhammel''
* Got UX feedback last week from the UX team
* Finish this week - implementing the last set of features
* Once that is done, send to bsmedberg and beltzner to get their comments on it


''' Bug Hunter ''' - ''bc''
''' Bug Hunter ''' - ''bc''
* ESX 4 host is up and running in Phoenix
* Getting the winxp reference machine ready to go
* Vista and centos vms - we should get phong to copy those directly out to phoenix.  Ctalbert will help with this
* Over the weekend, finished processing July 12's data, finding a crash and valgrind regression
* VM setup and moving to Phoenix.


= The Gravel =
= The Gravel =


'''Mozmill 1.4.2''' - ''harth, ctalbert et all''
'''Mozmill 1.4.2''' - ''harth, ctalbert et all''
* Should release this week!
* Need to update the timeout bug error message.


'''GrafxBot''' - ''jgriffin''
'''GrafxBot''' - ''jgriffin''
Line 34: Line 60:


'''Manifest for the manifest-less''' - ''jmaher, jgriffin, anode''
'''Manifest for the manifest-less''' - ''jmaher, jgriffin, anode''
* Tried to land a prerequisite patch last week and it got backed out
* Need to re-land this week.


'''Jetpack Test Buildbot Integration''' - ''ctalbert''
'''Jetpack Test Buildbot Integration''' - ''ctalbert''
* Need to land the runcfx script this week into jetpack repo
* Need to work with Lukas to make sure it gets run in staging or run it myself in Alice's staging system.


= The Round Table =
= The Round Table =
* The content team is ready for us to jump in and help remove universal xpconnect from mochitests.  Can we assemble a tiger team for either later this week or next to do this? (see {{bug|582472}})
* The content team is ready for us to jump in and help remove universal xpconnect from mochitests.  Can we assemble a tiger team for either later this week or next to do this? (see {{bug|582472}})
* Trying out packaging Ted's idea on packaging debug runtime on windows machines
* Filed a bug last week on doing stuff with debugCRT messages: {{bug|587982}}.  It might really help on the randomorange issue.  Having the ability to not have these modals show up should help us not have to stop the tests when this happens.
** We can probably just fix that straight away.


= The Pulverized =
= The Pulverized =
Confirmed users
1,905

edits