From Fedora Project Wiki

< QA‎ | Meetings

(Updated notes)
m (internal link cleaning)
 
Line 27: Line 27:
: jlaska sent out an email to design-team this morning to collect some thoughts (see http://lists.fedoraproject.org/pipermail/design-team/2010-August/003190.html)
: jlaska sent out an email to design-team this morning to collect some thoughts (see http://lists.fedoraproject.org/pipermail/design-team/2010-August/003190.html)
# jlaska to publish F-14-Alpha QA retrospective page
# jlaska to publish F-14-Alpha QA retrospective page
: the wiki page is available for anyone to record thoughts about the Alpha ... (see https://fedoraproject.org/wiki/Fedora_14_QA_Retrospective)
: the wiki page is available for anyone to record thoughts about the Alpha ... (see [[Fedora_14_QA_Retrospective)]]
: it's not wiki formatted to my liking yet, but I'll get to that soon
: it's not wiki formatted to my liking yet, but I'll get to that soon


Line 75: Line 75:
; Owner - [[User:wwoods]]
; Owner - [[User:wwoods]]
; Summary
; Summary
: jskladan finished the "how to write autoqa tests" wiki draft (see https://fedoraproject.org/wiki/User:Jskladan/Sandbox:Writing_autoqa_tests)
: jskladan finished the "how to write autoqa tests" wiki draft (see [[User:Jskladan/Sandbox:Writing_autoqa_tests)]]
: wwoods updated the autoqa depcheck tickets to give a bit more information - including the "make it use mash" ticket (see https://fedorahosted.org/autoqa/ticket/201)
: wwoods updated the autoqa depcheck tickets to give a bit more information - including the "make it use mash" ticket (see https://fedorahosted.org/autoqa/ticket/201)
: vaschenb updated upgradepath, the new test... now it has improved output and is automaticaly started with post-bodhi-update hook
: vaschenb updated upgradepath, the new test... now it has improved output and is automaticaly started with post-bodhi-update hook
; Next steps...
; Next steps...
: jskladan will replace current wiki page [[Writing_autoqa_tests]] with https://fedoraproject.org/wiki/User:Jskladan/Sandbox:Writing_autoqa_tests
: jskladan will replace current wiki page [[Writing_autoqa_tests]] with [[User:Jskladan/Sandbox:Writing_autoqa_tests]]
: wwoods + jskladan investigating how best to handle mash with depcheck test
: wwoods + jskladan investigating how best to handle mash with depcheck test
: wwoods also investigating setting up depcheck to handle karma
: wwoods also investigating setting up depcheck to handle karma
Line 264: Line 264:
|- id="t15:08:11"
|- id="t15:08:11"
! style="background-color: #407a40" | jlaska
! style="background-color: #407a40" | jlaska
| style="color: #407a40" | https://fedoraproject.org/wiki/Fedora_14_QA_Retrospective
| style="color: #407a40" | [[Fedora_14_QA_Retrospective]]
|| [[#t15:08:11|15:08]]  
|| [[#t15:08:11|15:08]]  
|- id="t15:08:16"
|- id="t15:08:16"
! style="background-color: #407a40" | jlaska
! style="background-color: #407a40" | jlaska
| style="color: #407a40" | #link https://fedoraproject.org/wiki/Fedora_14_QA_Retrospective
| style="color: #407a40" | #link [[Fedora_14_QA_Retrospective]]
|| [[#t15:08:16|15:08]]  
|| [[#t15:08:16|15:08]]  
|- id="t15:08:19"
|- id="t15:08:19"
Line 326: Line 326:
|- id="t15:11:11"
|- id="t15:11:11"
! style="background-color: #407a40" | jlaska
! style="background-color: #407a40" | jlaska
| style="color: #407a40" | for those new to test days ... there is a great introduction to the process at https://fedoraproject.org/wiki/QA/SOP_Test_Day_management
| style="color: #407a40" | for those new to test days ... there is a great introduction to the process at [[QA/SOP_Test_Day_management]]
|| [[#t15:11:11|15:11]]  
|| [[#t15:11:11|15:11]]  
|- id="t15:11:31"
|- id="t15:11:31"
Line 338: Line 338:
|- id="t15:12:25"
|- id="t15:12:25"
! style="background-color: #4b904b" | kparal
! style="background-color: #4b904b" | kparal
| style="color: #4b904b" | #link https://fedoraproject.org/wiki/Test_Day:2010-08-26_OpenSCAP
| style="color: #4b904b" | #link [[Test_Day:2010-08-26_OpenSCAP]]
|| [[#t15:12:25|15:12]]  
|| [[#t15:12:25|15:12]]  
|- id="t15:12:48"
|- id="t15:12:48"
Line 416: Line 416:
|- id="t15:18:08"
|- id="t15:18:08"
! style="background-color: #407a40" | jlaska
! style="background-color: #407a40" | jlaska
| style="color: #407a40" | #link https://fedoraproject.org/wiki/Test_Day:2010-09-02_Preupgrade
| style="color: #407a40" | #link [[Test_Day:2010-09-02_Preupgrade]]
|| [[#t15:18:08|15:18]]  
|| [[#t15:18:08|15:18]]  
|- id="t15:18:44"
|- id="t15:18:44"
Line 646: Line 646:
|- id="t15:29:04"
|- id="t15:29:04"
! style="background-color: #818144" | adamw
! style="background-color: #818144" | adamw
| style="color: #818144" | kparal: https://fedoraproject.org/wiki/QA:Testcase_initialization_basic , https://fedoraproject.org/wiki/QA:Testcase_initialization_tools
| style="color: #818144" | kparal: [[QA:Testcase_initialization_basic]] , [[QA:Testcase_initialization_tools]]
|| [[#t15:29:04|15:29]]  
|| [[#t15:29:04|15:29]]  
|- id="t15:29:14"
|- id="t15:29:14"
Line 662: Line 662:
|- id="t15:29:41"
|- id="t15:29:41"
! style="background-color: #407a40" | jlaska
! style="background-color: #407a40" | jlaska
| style="color: #407a40" | #link https://fedoraproject.org/wiki/Test_Day:2010-09-09_Systemd
| style="color: #407a40" | #link [[Test_Day:2010-09-09_Systemd]]
|| [[#t15:29:41|15:29]]  
|| [[#t15:29:41|15:29]]  
|- id="t15:29:41"
|- id="t15:29:41"
Line 686: Line 686:
|- id="t15:30:10"
|- id="t15:30:10"
! style="background-color: #4b904b" | kparal
! style="background-color: #4b904b" | kparal
| style="color: #4b904b" | adamw: it's not available from https://fedoraproject.org/wiki/QA/Fedora_14_test_days
| style="color: #4b904b" | adamw: it's not available from [[QA/Fedora_14_test_days]]
|| [[#t15:30:10|15:30]]  
|| [[#t15:30:10|15:30]]  
|- id="t15:30:15"
|- id="t15:30:15"
Line 961: Line 961:
|- id="t15:46:44"
|- id="t15:46:44"
! style="background-color: #539e9e" | jskladan
! style="background-color: #539e9e" | jskladan
| style="color: #539e9e" | https://fedoraproject.org/wiki/User:Jskladan/Sandbox:Writing_autoqa_tests
| style="color: #539e9e" | [[User:Jskladan/Sandbox:Writing_autoqa_tests]]
|| [[#t15:46:44|15:46]]  
|| [[#t15:46:44|15:46]]  
|- id="t15:46:51"
|- id="t15:46:51"
! style="background-color: #4b904b" | kparal
! style="background-color: #4b904b" | kparal
| style="color: #4b904b" | #link https://fedoraproject.org/wiki/User:Jskladan/Sandbox:Writing_autoqa_tests
| style="color: #4b904b" | #link [[User:Jskladan/Sandbox:Writing_autoqa_tests]]
|| [[#t15:46:51|15:46]]  
|| [[#t15:46:51|15:46]]  
|- id="t15:47:13"
|- id="t15:47:13"

Latest revision as of 09:02, 18 September 2016

Attendees

People present (lines said):

  • jlaska (98)
  • adamw (66)
  • kparal (41)
  • wwoods (41)
  • jskladan (22)
  • skvidal (16)
  • Viking-Ice-Home (15)
  • fenris02 (10)
  • zodbot (4)
  • robatino (3)
  • jsmith (3)
  • vaschenb (2)

Unable to attend:

  1. Rhe
  2. Newgle1

Agenda

Previous meeting follow-up

  1. adamw and jlaska to propose artwork final release criteria
jlaska sent out an email to design-team this morning to collect some thoughts (see http://lists.fedoraproject.org/pipermail/design-team/2010-August/003190.html)
  1. jlaska to publish F-14-Alpha QA retrospective page
the wiki page is available for anyone to record thoughts about the Alpha ... (see Fedora_14_QA_Retrospective)
it's not wiki formatted to my liking yet, but I'll get to that soon

F14 Test Days

Aug 26 - OpenSCAP

Owner - kparal
Summary
Aug 26 - OpenSCAP
the test day was quite successful, quite a lot of problems were identified and they were fixed
the new openscap release, which is just waiting for pushing into updates, should fix all those issues
Next steps...
kparal will send a recap to the list later today

Sep 02 - preupgrade

Owner - rhe
Summary
Sep 02 - preupgrade
Wiki page and test cases in place. As rhe mentioned on the list, testing preupgrade-cli will be included in this event
Next steps...
<adamw> will get some publicity out for this one as we'll want to have as many people as possible test
<jlaska> to reach out to rhe for any final wiki modifications needed

Sep 09 - systemd

Owner - adamwill
Summary
Sep 09 - Systemd
Adamw has the test day mostly set up, he would appreciate review of the monster test cases (which are heavily based on notting's systemd 'acceptance criteria')
Current systemd package in F-14 stable should resolve most of issues identified on mailing list (https://admin.fedoraproject.org/updates/systemd-8-3.fc14,initscripts-9.17-2.fc14)
Next steps...
Adamw add a few other test cases, he'll check in with lennart on what he thinks so far
Viking-Ice intends to complete the systemd debugging page

Proventester (metrics and sponsor upgrade)

Owner - jlaska
Summary
Received a request from a proventester to be upgraded to a sponsor, so that they could mentor others. This got me thinking whether we have (or need) expectations around becoming a sponsor.
In addition, are we able to gather metrics about proventesters?
Next steps...
Adamw directed jlaska to the policy regarding becoming a mentor.
Jlaska will elevate mcloaked to sponsor status.
Adamw recalled previously filed metrics ticket (https://fedorahosted.org/bodhi/ticket/456) and took action to follow-up at the next meeting

AutoQA update

Owner - User:wwoods
Summary
jskladan finished the "how to write autoqa tests" wiki draft (see User:Jskladan/Sandbox:Writing_autoqa_tests)
wwoods updated the autoqa depcheck tickets to give a bit more information - including the "make it use mash" ticket (see https://fedorahosted.org/autoqa/ticket/201)
vaschenb updated upgradepath, the new test... now it has improved output and is automaticaly started with post-bodhi-update hook
Next steps...
jskladan will replace current wiki page Writing_autoqa_tests with User:Jskladan/Sandbox:Writing_autoqa_tests
wwoods + jskladan investigating how best to handle mash with depcheck test
wwoods also investigating setting up depcheck to handle karma
autoqa team agrees now is a good time to make a new autoqa release and deploy it (coming soon ... autoqa-0.4.0-1)

Open discussion - <Your topic here>

Upcoming QA events

Action items

  1. adamw contact lmacken about proventester metrics https://fedorahosted.org/bodhi/ticket/456
  2. jlaska build and deploy new autoqa release

IRC Transcript

jlaska #startmeeting Fedora QA Meeting 15:00
zodbot Meeting started Mon Aug 30 15:00:17 2010 UTC. The chair is jlaska. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00
zodbot Useful Commands: #action #agreed #halp #info #idea #link #topic. 15:00
jlaska #meetingname fedora-qa 15:00
zodbot The meeting name has been set to 'fedora-qa' 15:00
jlaska #topic Waiting for critical mass 15:00
* adamw gets critical 15:00
* kparal enjoys brand new QA meeting :) 15:00
* jskladan tips his hat 15:01
jlaska adrianr: not sure why that makes me think of an Olivia Newton-John song :) 15:01
* vaschenb is here 15:01
jlaska adamw: ^^^ 15:01
jlaska kparal: jskladan vaschenb welcome gang 15:01
adamw red dwarf: still, you know what they say - 'better to have loved and to have lost than to have listened to a record by Olivia Newton-John'. But then, anything's better than listening to a record by Olivia Newton-John... 15:02
jlaska kparal: adamw: you both may be spared my grammar and spelling mistakes for half of the meeting today 15:02
jlaska #chair kparal adamw 15:02
zodbot Current chairs: adamw jlaska kparal 15:02
* wwoods is here 15:02
jlaska I may need to drop out halfway through the meeting 15:02
jlaska hey wwoods 15:03
jlaska I saw robatino and Viking-Ice-Home join earlier ... you guys lurking? 15:03
Viking-Ice-Home yup 15:03
jlaska welcome 15:03
robatino yes 15:03
jlaska double welcome :) 15:03
jlaska okay, I think we all feel critical (or massy) ... let's get started 15:04
jlaska #topic Previous meeting follow-up 15:04
jlaska #info adamw and jlaska to propose artwork final release criteria 15:04
jlaska nothing big to report back yet on this ... I sent out an email to design-team this morning to collect some thoughts 15:05
jlaska #link http://lists.fedoraproject.org/pipermail/design-team/2010-August/003190.html 15:05
jlaska Feel free to weigh in, otherwise I'll collect some feedback and we can make a more precise criteria proposal 15:05
adamw thanks for handling that 15:06
jlaska certainly, sorry took so long :) 15:06
Viking-Ice-Home you might want to add copyright clause 15:06
jlaska definitely relevant as it came up again in the blocker meeting 15:06
Viking-Ice-Home displaying correct year :) 15:06
jlaska Viking-Ice-Home: heh, good point 15:07
jlaska #info jlaska to publish F-14-Alpha QA retrospective page 15:07
jlaska the wiki page is available for anyone to record thoughts about the Alpha ... 15:07
jlaska it's not formatted to my liking yet, but I'll get to that soon 15:08
jlaska Fedora_14_QA_Retrospective 15:08
jlaska #link Fedora_14_QA_Retrospective 15:08
* jsmith stumbles in, a bit late 15:08
jlaska jsmith: welcome 15:08
jsmith jlaska: Thanks :-) 15:08
jlaska the 2 other action items were autoqa specific, so I'll save those for that topic 15:08
jlaska anything else to discuss from last week? 15:08
jlaska alrighty ... moving on 15:09
jlaska #topic F14 Test Days 15:09
jlaska if you haven't noticed, test days have started in full swing again 15:10
adamw yaaay test days 15:10
jlaska hoooray! 15:10
jlaska :) 15:10
jsmith w00t! 15:10
jlaska I was hoping to spend just a few minutes rounding the bases </poor sports analogy> for previous and upcoming events 15:10
* fenris02 smirks 15:11
jlaska for those new to test days ... there is a great introduction to the process at QA/SOP_Test_Day_management 15:11
jlaska #topic F14 Test Days - Aug 26 OpenSCAP 15:11
jlaska kparal, you and wrabco collaborated in setting up this event. How'd it go last week? 15:12
kparal #link Test_Day:2010-08-26_OpenSCAP 15:12
jlaska #link http://kparal.wordpress.com/2010/08/25/openscap-test-day-thursday-2010-08-26/ 15:12
kparal the test day wast quite successful, as opposed to what test matrix might look like :) 15:12
adamw sorry i didn't get on the publicity wagon for this one, got it a late mention in fwn but that's all 15:13
kparal well, the red fields are just best results from QA perspective, aren't they? :) 15:13
kparal quite a lot of problems were identified 15:13
jlaska it seems so ... a fair number of attendees as well 15:13
kparal and they were fixed. the new openscap release, which is just waiting for pushing into updates, should fix all those issues 15:14
Viking-Ice-Home thats great work 15:14
* kparal will send test day recap into list shortly 15:14
fenris02 kparal, does this replace sectool ? 15:14
kparal fenris02: I am not sure I can answer that, openscap developers might be a better choice :) I believe it should 15:15
kparal to tell the truth, I've never heard of openscap nor sectool before this test day 15:15
jlaska yay, learning all around :) 15:16
kparal but let's hope it got at least some attention and more people learned what it is good for 15:16
jlaska kparal: nicely done on the setup with wrabco ... look forward to the recap later this week 15:16
jlaska kparal: feel free to note any good/bad/ugly on the retrospective page 15:17
* kparal will try 15:17
jlaska #topic F14 Test Days - Sep 02 PreUpgrade 15:17
jlaska Hurry (rhe) isn't here at the moment, but it seems she has a good handle on the setup for this weeks preupgrade event 15:18
jlaska #link Test_Day:2010-09-02_Preupgrade 15:18
jlaska as mentioned on the list, the test day will request testing against the command-line interface 15:18
jlaska we've not previously done much with it 15:18
jlaska this should be a good event ... nice to shake out/debug any preupgrade issues in advance of the beta 15:19
fenris02 is there any way to determine what percentage of users actually use the cli vs. gui for preupgrade? 15:19
jlaska robatino: you've done some preupgrades recently, are there any show stoppers we need to address before the event? 15:19
adamw i'll get some publicity out for this one as we'll want to have as many people as possible test 15:19
jlaska fenris02: good question ... I'm not aware of a good method 15:19
robatino jlaska: i haven't tested preupgrade for F14 (problems with VMs not working) 15:20
Viking-Ice-Home I always thought that those that used cli would just run yum upgrade.. 15:20
robatino KVM works better but just getting used to it 15:20
jlaska Viking-Ice-Home: they might, that is however a completely different method for upgrading a system 15:20
fenris02 i wonder how preupgrade would compare now that yum has dist-sync ... 15:21
jlaska fenris02: I've had enough of your good questions! 15:21
* Viking-Ice-Home just keeps home on separate partition and does a fresh install.. 15:21
jlaska Viking-Ice-Home: +1 to that method :) 15:21
fenris02 jlaska, that's what i'm here for :) 15:21
jlaska #help i wonder how preupgrade would compare now that yum has dist-sync 15:22
jlaska #help is there any way to determine what percentage of users actually use the cli vs. gui for preupgrade? 15:22
skvidal okay 15:22
* jlaska readies for next #topic 15:22
skvidal so preupgrade just preps the system and downloads the pkgs for an anaconda install 15:22
skvidal so that if there is a significant change that only anaconda can do - then it can be done 15:22
skvidal if the change is more iterative then yum can do it 15:22
jlaska #info skvidal clarified the difference between preupgrade and yum upgrade 15:23
fenris02 skvidal, you do not think yum could dist-sync your system from say, f13 to f14 after you updated the -release package? 15:23
skvidal fenris02: there have been updates to major system pieces before that were impossible to do ON the running system 15:23
adamw fenris02: it's an impossible question, it depends on the upgrade 15:24
skvidal you had to be standing outside of the system to perform the upgrade 15:24
skvidal the lvm->lvm2 transition as a case in point 15:24
fenris02 point 15:24
skvidal you couldn't be RUNNING the kernel from the previous release and hope to survive it 15:24
wwoods or upgrading ext3->ext4, or (someday perhaps) ext4->btrfs 15:24
skvidal wwoods: istr ext3->ext4 wasn't the end of the world but you had to reboot RIGHT AWAY 15:24
wwoods there have also been incompatible differences in the rpmdb, if memory serves 15:24
jlaska yeah 15:25
skvidal wwoods: yes - most of those required: yum update yum rpm ; yum upgrade 15:25
wwoods or the metadata switchover, but you could handle those probably by upgrading yum/rpm/etc first 15:25
wwoods yeah 15:25
skvidal wwoods: nod 15:25
jlaska okay, so it seems preupgrade has everything lined up for this week 15:25
adamw this seems somewhat ot 15:25
jlaska good info, but yeah, we can move this outside the meeting 15:25
adamw for now, we're testing preupgrade, not debating its philosophical / practical merits =) 15:25
wwoods it's a tremendously complex problem that anaconda has already solved, and preupgrade takes advantage of that, and therefore, yes, there is still a useful niche for preupgrade 15:25
jlaska #topic F14 Test Days - Sep 09 Systemd 15:26
adamw ooh, this is one of mine 15:26
jlaska adamw: Viking-Ice-Home: I think you two are the closest to the state of systemd 15:26
jlaska how's this event shaping up? 15:26
adamw i have the test day mostly set up, i would appreciate review of the monster test cases (which are heavily based on notting's systemd 'acceptance criteria') 15:26
adamw i may add a few other test cases, i'll check in with lennart on what he thinks so far 15:27
jlaska ooh, I'd like to review ... I'm in need of improved systemd knowledge :) 15:27
Viking-Ice-Home I think we play close attention to upgrade case to see if that's ironed out 15:28
kparal adamw: post here a link 15:28
jlaska Viking-Ice-Home: F13 -> F14 upgrades? 15:28
Viking-Ice-Home yup and systemd 15:28
adamw kparal: they're on the page 15:28
Viking-Ice-Home there was set back with 8.1 that got fixed in 8.3 15:28
adamw kparal: QA:Testcase_initialization_basic , QA:Testcase_initialization_tools 15:29
kparal adamw: ah, there's no summary page yet? 15:29
adamw kparal: a category page? 15:29
kparal well, a test day page 15:29
jlaska #link Test_Day:2010-09-09_Systemd 15:29
adamw yes 15:29
adamw jlaska linked to it already i thought 15:29
adamw oh no :) 15:29
adamw thanks 15:29
Viking-Ice-Home it would just be good to test systemd when preupgrade tests are performed 15:29
kparal adamw: it's not available from QA/Fedora_14_test_days 15:30
jlaska Viking-Ice-Home: okay 15:30
jlaska kparal: I just updated that page 15:30
adamw kparal: oh yeah, i'll fix that 15:30
kparal thanks 15:30
jlaska at one point we were working on a debugging page for systemd ... is that still in progress? 15:31
adamw think that's viking's department 15:31
kparal adamw: there should be clear instructions how to setup F14 machine, since installing F14 Alpha and updating will break the boot process. to avoid confusion and reporting that as bugs, it should be covered in installation instructions 15:32
jlaska not a bad idea ... from the list at least, it seems there is significant confusion around what to test, and how to get there 15:32
jlaska this could help clear that up 15:32
adamw kparal: it shouldn't by the time we run the event 15:33
adamw in fact it shouldn't now, since updates-testing is enabled by default and 8-3 is in updates-testing 15:33
jlaska #info Test cases available, will request review shortly 15:33
kparal adamw: alright, you're always ahead of me :) 15:33
adamw actually, it's gone stable now...https://admin.fedoraproject.org/updates/systemd-8-3.fc14,initscripts-9.17-2.fc14 15:34
adamw though /me wonders about that unpush by lennart before the push to stable...better check with him on that 15:34
jlaska anything else to consider to prepare for this event? 15:35
adamw not that i can think of 15:35
jlaska feel free to #info,#action,#{favorite-tag} as needed 15:35
jlaska cool ... I'm sure this will be another well attended event given the buzz :) 15:35
jlaska #topic Proventester (metrics and sponsor upgrade) 15:36
jlaska I wanted to raise this for brief discussion 15:36
jlaska I know this has come up previously, but I received a request from a proventester to upgrade them to a sponsor 15:36
jlaska Wanted to ask if we've done this before, what we'd need to consider for this etc... 15:37
Viking-Ice-Home jlaska: Ah yes I need to finish that one up 15:37
Viking-Ice-Home systemd debugging page that is 15:37
jlaska and it seemed like a good flow into the previous topic of gathering proventester metrics 15:37
adamw what I wrote in the policy was just "Any proven tester can become a mentor. Simply let any existing mentor or group administrator - those listed as administrator or sponsor in the group member list - know you would like to become a mentor, and they will upgrade you to sponsor level, which will allow you to accept applications to the group. " 15:37
jlaska Viking-Ice-Home: cool thanks, I'm happy to review if you need input 15:37
adamw so basically the process to become a mentor is...ask to be a mentor 15:37
jlaska adamw: this is too simple! 15:38
adamw going with the priniciple of least resistance we agreed on for the whole proven tester process 15:38
fenris02 no trac ticket? 15:38
* adamw hates bureaucracy 15:38
jlaska we need more obfuscation in the way :) 15:38
adamw a trac ticket would give a record of the process i guess. i just made it up as i was going along. 15:38
jlaska fenris02: ah, yeah I can recommend the trac ticket route just for transparency 15:38
adamw but for now according to what we have written down, you can go ahead and make 'em a mentor, jlaska 15:38
adamw if someone wants to revise the policy to say 'file a trac ticket' i'm fine with that 15:39
fenris02 jlaska, CYAP 15:39
jlaska okay, thanks 15:40
jlaska anything else we want to talk through on the metrics side 15:41
adamw on metrics, um, not done anything yet. 15:41
jlaska adamw: you reached out to lmacken, I forget what the plan is here 15:41
adamw it should probably go on my todo list 15:41
jlaska this is something included in an upcoming bodhi release? 15:41
adamw luke asked me to file a trac ticket with a list of desired metrics 15:41
adamw i think i did that...lemme see 15:41
adamw yeah, https://fedorahosted.org/bodhi/ticket/456 15:42
adamw seems we're waiting on luke there, we could give him a poke 15:42
jlaska cool, I forgot about the ticket, thanks adamw 15:42
jlaska okay, we can check-in with lmacken later 15:43
jlaska okay, I need to move my focus to another meeting :( 15:43
adamw #action adamw contact lmacken about proventester metrics https://fedorahosted.org/bodhi/ticket/456 15:43
jlaska kparal or adamw can you guys push throug the remaining autoqa and open discussion topics? 15:44
adamw kparal: do you want to take autoqa? it's your area 15:44
adamw #topic AutoQA 15:44
kparal adamw: ok 15:44
adamw kparal: wwoods: go for it 15:44
wwoods whoosh 15:45
adamw is it a bird? is it a plane? no, it's...autoqa! 15:45
kparal last week there were some big documentation updates 15:45
kparal the main culprit is jskladan, so I'll let him talk about it 15:45
kparal praise yourself, jskladan! :) 15:45
wwoods hooray! 15:45
jskladan okay 15:46
jskladan last week, i finished the "how to write autoqa tests" wiki draft 15:46
fenris02 autoqa should be a massive WIN for fedora :) 15:46
Viking-Ice-Home jskladan: got link 15:46
jskladan User:Jskladan/Sandbox:Writing_autoqa_tests 15:46
kparal #link User:Jskladan/Sandbox:Writing_autoqa_tests 15:46
jskladan so, feel free to comment/edit... 15:47
adamw nice stuff 15:47
jskladan and at the end of this week, i'll replace the "official" page with the content of mine sandbox 15:47
adamw might be worth mentioning to the docs team? see if they have any special skills to improve it? 15:47
jskladan adamw: great idea? could you throw any names? 15:48
jskladan (so i can abuse their knowledge and creativity :-D) 15:48
Viking-Ice-Home I think looks good 15:48
kparal #info jskladan updated AutoQA documentation 15:48
kparal #info jskladan will replace the official page with the content of his sandbox at the end of the week 15:48
adamw jskladan: mailing list - docs@lists.fp.o 15:48
* Viking-Ice-Home sneaks in 'it' 15:48
adamw jskladan: or irc #fedora-docs i think 15:48
jskladan ok, will try, thx adamw 15:49
jskladan other than that - i have nothing 15:49
jskladan i started to dig into the mash 15:49
jskladan but that thing is crazy :-D 15:49
wwoods yeah, mash is kind of.. black magic 15:49
wwoods but it's black magic we need 15:49
wwoods so that's a nice segue to depcheck! 15:50
jskladan i was thinking - can we (maybe) add/change tags on packages in koji? 15:50
jskladan you know - add a custom (unique) tag on the one package, create repo from that that tag 15:50
wwoods I updated the autoqa depcheck tickets to give a bit more information - including the "make it use mash" ticket 15:50
jskladan and then remove the respective tag? :) 15:50
wwoods that's https://fedorahosted.org/autoqa/milestone/Package%20Update%20Acceptance%20Test%20Plan%20-%20depcheck 15:51
wwoods and the ticket: https://fedorahosted.org/autoqa/ticket/201 15:51
kparal #info wwoods added comments to depcheck tickets to explain more about mash support 15:51
wwoods jskladan: yes, we could do something like that - but remember that we need to run depcheck using *all* the packages that have requested moving 15:52
wwoods not just the latest one 15:52
jskladan sure, but that could be arranged :) (you know, tagging one package, or several packages... same thing...) 15:52
wwoods so the simplest process would be: when a maintainer requests an update get moved to stable/testing, the builds get a new tag (like -blah-pending) 15:52
wwoods then when we run the test, we use that tag to get ourselves a proper multilib repo of proposed packages 15:53
wwoods I seem to remember Oxf13 made tags for this purpose 15:53
wwoods but bodhi doesn't tag packages like that 15:53
adamw so, summary here - you need to work with mash for the depcheck tests? 15:53
wwoods so either we can a) make the autoqa bodhi watcher handle the tagging, or b) figure out a way to run mash without needing a tag 15:53
jskladan so, maybe the "post koji build" watcher could fire some "tag this package" routine? 15:53
wwoods adamw: yes. 15:53
adamw and you're working on how to do that 15:53
adamw okay 15:53
wwoods or, at least, we need the multilib-solving bits of mash 15:54
adamw we got 6 minutes left so maybe the practical aspects can go outside of the meeting :) 15:54
jskladan ok ok 15:54
* jskladan out 15:54
adamw meetings aren't for practical stuff, everyone should know that! 15:54
adamw ;) 15:54
* jskladan lurks into the shadows :) 15:54
skvidal wwoods: you could declare multilib forbidden and never touch it again, that would also work :) 15:54
* skvidal kids 15:54
wwoods heh. the point I'm making is a) we require this for proper multilib testing, and b) it's kind of tricky and we're still trying to determine the best way to accomplish that 15:54
wwoods yes, we could also just decide we aren't going to test multilib 15:55
wwoods and/or have depcheck 0.1 just.. not handle multilib 15:55
adamw #info depcheck tests need to use mash to handle multilib cases, team is investigating how best to do this 15:55
adamw #info possibility to release initial set of depcheck tests without multilib handling 15:55
wwoods and work on handling multilib/mash in parallel with setting up depcheck to handle karma etc 15:55
skvidal wwoods: probably unwise 15:56
skvidal once you get it mostly working 15:56
skvidal no one will want to disturb it only to make the whole world break with multilib 15:56
wwoods at this point I'm thinking about the latter, but I *REALLY DO NOT* want depcheck to be active and enforcing policy until we're really sure it's doing everything right 15:56
wwoods skvidal: yeah, exactly 15:56
wwoods once it's in Production nobody will want to change it, so it can't go into Production until it's feature-complete 15:56
adamw Google solution - endless beta! 15:57
wwoods hah yes. 15:57
* kparal notes there is one more update from vaschenb about upgradepath 15:57
adamw yeah, let's get to that 15:58
kparal vaschenb: go! 15:58
wwoods onward! 15:58
vaschenb i updated upgradepath, the new test... now it has improved output and is automaticaly started with post-bodhi-update hook... 15:59
kparal that means it's now included in the set of tests run on always bodhi update 15:59
kparal always->every 16:00
kparal and that reminds me one more thing, I asked jlaska whether we should deploy a new version of autoqa to our production server 16:00
kparal #info vaschenb updated upgradepath - now it has improved output and is automaticaly started with post-bodhi-update hook 16:01
kparal but I guess jlaska is busy right now 16:01
kparal wwoods: jskladan: what do you think? 16:01
kparal I think the current autoqa version is quite outdated 16:01
wwoods If memory serves there are some big changes and bugfixes etc. that we'll want on the production systems soon 16:02
jlaska kparal: I'm happy to build and deploy as needed 16:02
jlaska I can do a autoqa-0.4-pre1 or something? 16:02
kparal it will take some time until depcheck is ready, right? I think now it quite a good time to make a snapshot 16:02
jskladan jlaska: cool :) i'd like to know, if my puppy can survive in the real world of production servers :) 16:02
wwoods agreed - it'd be good to get some testing on the new code before we start adding new tests on top of it 16:03
wwoods if need be we can take this to the list and review what (if anything) we need to fix/finish before pushing new code 16:03
wwoods but I think kparal's right and this is an opportune time 16:03
wwoods I feel like it's ready for a snapshot - if there are no concerns let's just do it 16:04
kparal #info autoqa team agrees now is a good time to make a new autoqa release and deploy it 16:04
* jskladan +1 16:04
wwoods otherwise, send send mail to the list and we'll talk about what we need 16:04
wwoods anyone have any concerns about trying to push the new autoqa code live? 16:05
* wwoods assumes silence == "let's do it!" 16:05
wwoods so.. let's do it 16:05
kparal #action jlaska build and deploy new autoqa release 16:05
jlaska I'm just going to build from master ... unless there are specific things you guys want me to wait for 16:06
* kparal hopes he's not too daring to assign action items like that :) 16:06
jlaska kparal: assign away! :) 16:06
kparal yes, from master, that should be fine 16:06
kparal I think that's all from autoqa for today, don't you think? 16:07
adamw okay then 16:08
adamw #topic open discussion 16:08
adamw any other business? 16:08
adamw otherwise, closing meeting in 3... 16:09
adamw 2.. 16:09
adamw 1... 16:09
adamw thanks, everyone! 16:09
adamw #endmeeting 16:10

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!