Calendar:QA Chat:2006-05-17:Log

From MozillaWiki
Jump to: navigation, search
ssitter was promoted to operator by ChanServ.
[11:25am] ctalbert: Five minutes to the QA Chat
[11:34am] ctalbert: Howdy folks.
[11:34am] ctalbert: Might as well go on with the QA Chat
[11:36am] ctalbert: mschroeder: Thanks for attempting to to run the qa chat last
 week.  I don't know if I'll be super available in the future at this timeslot, so
  it's good for this small meeting to depend a bit less on my presence.
[11:36am] ctalbert: I hesitate to move the time since that's always difficult.
[11:37am] ctalbert: the news...we had a calendar test day on Tuesday
[11:37am] ctalbert: I think largely due to the fact that we posted on Mozilla Zine
 in addition to our normal places, we had a huge turnout.
[11:38am] ctalbert: I haven't had time to count everyone from the IRC logs but I 
think it was almost 20 people.
[11:38am] ctalbert: We ran 332 litmus tests, opened 9 non-dupe bugs. 
[11:39am] ctalbert: None of the bugs were serious enough to block the release.  We
 did get good coverage on the testcases, and on OS's.  Folks ran tests on Windows, 
 Linux, Mac PPC and Mac Intel.
[11:39am] ctalbert: So, this was a test day we can all be proud of.
[11:39am] ctalbert:
[11:39am] ctalbert: The winner of the test day is ssitter.  No one came close to
 catching him.
[11:40am] ctalbert: Congratulations.
[11:40am] mschroeder: ssitter: grats 
[11:40am] mschroeder: Moving the time should be no problem because we are always
 the same, few people.
[11:40am] ctalbert: I need to catch Marcia next week (she's out this week) and get
 you a prize.
[11:41am] ctalbert: mschroeder: The only time that would be better would be to either
 move this to Friday or earlier on Thursday.
[11:42am] ctalbert: I don't want to move it earlier, because that conflicts with folks'
 work schedules in Europe.
[11:42am] mschroeder: then it's more difficult than I thought
[11:42am] ctalbert:
[11:43am] ctalbert: My solution is for me to try my best to keep this slot free.
[11:43am] ctalbert:
[11:43am] ctalbert: Which has worked so far.
[11:43am] mschroeder: okay.
[11:44am] ctalbert: If it becomes more of a problem, then we'll look at a change. 
 Right now, it's manageable.
[11:44am] ssitter: thanks
[11:44am] ctalbert: Any comments or questions from the test day?
[11:44am] ctalbert: I have one: We should always post to mozilla zine
[11:45am] mschroeder: They should be forced to file UNCONFIRMED bugs 
[11:46am] ctalbert: What's the real reason behind that?  I told them it was for
 "tracking purposes", but I think it's because they often aren't really checking
  for dupes/confirming bugs themselves.  Is that right?
[11:48am] mschroeder: No. For me it's the tracking purpose and the workflow of
 confirming a bug by a second person.
[11:49am] ctalbert: Ok.  I think it never hurts to force two people to look at the
 bugs we find in the test day anyway.  Many times, they are dupes and people aren't 
 good enough at searching bz to find that out.
[11:50am] ctalbert: I'll add something to the Test Day Wiki Template to put that in there.
[11:50am] mschroeder: I have to drop out of the QA chat. Dinner is already on the table
 and my girlfriend is waiting. Bad timing today.
[11:50am] ctalbert: heh, Have a nice dinner
[11:51am] mschroeder: Thanks.
[11:51am] Archaeopteryx: send her on vacation the next time you need time for calendar 
;-). et bon appetite
[11:51am] ctalbert: However, I want to point out that so far from Tuesday's test day, 
we only decided that one of those bugs were dupes.  The 9 others seem to be valid.
[11:53am] ctalbert: Alright, on to the real work.  QA Discussion Bugs.  Grab the URL 
from here:
[11:53am] ctalbert: sort them by severity
[11:53am] ctalbert: bug 357493 is first
[11:53am] firebot: ctalbert: Bug maj,
 --, ---,, UNCO, Lightning unable to install if profile is stored on a
  fat32/vfat partition on Linux
[11:54am] Archaeopteryx: i have no fat partition
[11:55am] ctalbert: Me either.  It sounds like this might be an issue with extension management,
 because he can do this with Sunbird without any issue (comment 9)
[11:57am] ctalbert: It's pretty old, but we've seen several similar bugs on this type of
 instal. Should this be a valid install path?
[11:57am] ctalbert: s/valid/supported
[11:59am] ctalbert: I think that is a question for Daniel.  I'll ask him. I recommend we
 leave this bug where it is for now.
[11:59am] ctalbert: Next up is bug 371946
[11:59am] firebot: ctalbert: Bug maj,
 --, ---,, UNCO, Cannot delete event on network calendar
[11:59am] Archaeopteryx: release notes of support calendars?
[12:00pm] Archaeopteryx: asking for example calendar?
[12:01pm] ctalbert: Yeah, we may want to relnote this.  I meant to ask mschroeder how it 
went with the install of webcalendar.
[12:01pm] • ctalbert should try to install it too, now that I have a working linux system
[12:02pm] Archaeopteryx: first bug looks like dupe of bug 343691
[12:02pm] firebot: Archaeopteryx: Bug
 nor, --, ---,, RESO DUPLICATE, import of .ics file fails with Lightning
[12:02pm] ctalbert: Archaeopteryx: which is a dupe of 343691?
[12:02pm] Archaeopteryx: 357493
[12:03pm] ctalbert: I don't follow your logic.  Why do you think so?
[12:05pm] Archaeopteryx: NS_ERROR_XPC_BAD_CID and lightning only, maybe lightning tries to
 access a non-existing calendar?
[12:05pm] Archaeopteryx: ok, please forget this and let us continue
[12:06pm] ctalbert: No, I think the bad CID message is because the Component Loader can't 
find the components/interfaces to register.
[12:06pm] ctalbert: ok
[12:07pm] ctalbert: looks like bug 379500 is  next
[12:07pm] firebot: ctalbert: Bug maj, 
--, ---,, UNCO, Calendar display on all views displaying incorrectly. See URL.
[12:08pm] Archaeopteryx: hm, trying to reproduce and if we are not able to do so, needsinfo
[12:08pm] ctalbert: seems like it ought to be QAWanted.  Even the reporter says he can't reproduce it.
[12:08pm] Archaeopteryx: because could also be a tb bug (remaining task)
[12:09pm] ctalbert: yeah, maybe some object died inside tb
[12:09pm] ctalbert: Ask for other information, and will move it to qawanted list.
[12:12pm] Archaeopteryx: next is 362505
[12:12pm] Archaeopteryx: bug 362505
[12:12pm] firebot: Archaeopteryx: Bug 
nor, --, ---,, NEW, Mutiple Error Windows Without Content
[12:12pm] mschroeder: I'm back.
[12:12pm] ctalbert: you're a fast eater.
[12:13pm] ctalbert: mschroeder:  What is the status of the  WebCalendar installation
[12:13pm] ctalbert: bug 362505 is up next...
[12:14pm] Archaeopteryx: ok, let me test this
[12:14pm] mschroeder: I only tried with the Webcalendar version not supporting timezones, 
only offsets. I could subscribe to a calendar, but I couldn't create/delete/edit. The server
 didn't report back an error, so after a reload all changes were gone.
[12:15pm] mschroeder: The events showed at the correct time.
[12:17pm] mschroeder: 362505 > INVALID (like all the other bugs with malformed or missing 
[12:19pm] ctalbert: Yes, I agree.  But is that error sufficient?  The user has no idea why the ICS is invalid
[12:19pm] ctalbert: (Hence the reason they keep filing these bugs)
[12:19pm] Archaeopteryx: "there has been an error reading data for calendar: basic"
[12:19pm] mschroeder: Wasn't there a bug about better error messages for ics parsing?
[12:20pm] Archaeopteryx: error number: Invalid_timezone
[12:20pm] Archaeopteryx: it has changed
[12:21pm] Archaeopteryx: hex is now invalid_timezone
[12:21pm] ctalbert: I'll try this in a bit, and if I don't like the new message I'll 
file a bug/add to an existing one.
[12:21pm] ctalbert: let's close 362505 as invalid
[12:21pm] damian joined the chat room.
[12:21pm] ctalbert: Hi damian
[12:22pm] damian: hi
[12:22pm] ctalbert: next bug is bug 364385
[12:22pm] firebot: ctalbert: Bug
 nor, --, ---,, NEW, Creating a repeating event with start date in
  past cloning the event in view
[12:24pm] ctalbert: Damian, lucky for you to join us.  Do you think that the steps 
in bug 364385 comment 7 are the same as your bug 352024?
[12:24pm] firebot: ctalbert: Bug 
nor, --, ---,, NEW, calendar alarm dialog contains the same events
[12:25pm] • ctalbert thinks we can make 364385 a dupe of 352024
[12:25pm] mschroeder: But the original bug was another one.
[12:26pm] mschroeder: and are WFM per comment#6
[12:26pm] damian: ctalbert: yes they looks the same
[12:26pm] ctalbert: true.  So should 364385 go to WFM or dupe?
[12:27pm] ctalbert: Note to all: this is why it's good to only have one issue per bug
[12:27pm] • ctalbert votes WFM
[12:27pm] ctalbert: (my logic: we maintain that the bug is about the original issue raised).
[12:27pm] mschroeder: +1
[12:28pm] chiller2 joined the chat room.
[12:28pm] ctalbert: And we already have a bug tracking the extra issue
[12:28pm] chiller2: Hello.
[12:28pm] chiller2: <- Kelv
[12:28pm] Archaeopteryx: hi kelv
[12:29pm] ctalbert: hi
[12:29pm] firebot: set the Resolution field on bug 364385 to 
[12:29pm] chiller2: I responded about that bug 379500, you probably saw. Not sure 
how useful that was.
[12:29pm] firebot: chiller2: Bug 
maj, --, ---,, UNCO, Calendar display on all views displaying 
incorrectly. See URL.
[12:29pm] Archaeopteryx: seems as we can make your bug a worksforme, feel free to
 reopen is it happens again
[12:29pm] ctalbert: We were just talking about your bug
[12:30pm] mschroeder: +1
[12:30pm] ctalbert: chiller, what do you think about that?
[12:30pm] ctalbert: er.. chiller2: ^^
[12:30pm] JasnaPaka left the chat room. (Quit: Leaving)
[12:30pm] chiller2: ?
[12:31pm] Archaeopteryx: seems as we can make your bug a worksforme, feel free to 
reopen is it happens again
[12:31pm] ctalbert: chiller2: We mark your bug as Works For Me, since no one can 
reproduce it.  If it happens again, we'll need you to re-report it.  Be sure to grab the error console output.
[12:31pm] chiller2: That sounds fair enough.
[12:31pm] ctalbert: ok.  Thanks for dropping in here.  I wish everyone would do that. 
[12:32pm] Archaeopteryx: thank you for the quick response
[12:32pm] chiller2: I'm happy to.
[12:32pm] firebot: set the Resolution field on bug 379500 to WORKSFORME.
[12:33pm] ctalbert: One more: bug 375397
[12:33pm] firebot: ctalbert: Bug 
nor, --, ---,, NEW, Wrong Redo keyboard shortcut in Edit menu
[12:33pm] chiller2: I still can't figure out what is going on with bug 380291 though.
 I've been installing each nightly and trying the same, with the same result. I'm sure 
 it's not my dav server though as the same ics file works fine in the other profile 
 using Lightning 0.3.1
[12:33pm] firebot: chiller2: Bug maj,
 --, ---,, UNCO, Event movement failure on local or remote events
[12:35pm] chiller2: I'm happy to try anything you want to help figure it out. Am even
 prepared to give access to it if need be so you can try it with your own client.
[12:35pm] Archaeopteryx: i get another error code
[12:35pm] Archaeopteryx: Error: [Exception... "Component returned failure code: 0x804a0001
 [calIIcalComponent.startTime]"  nsresult: "0x804a0001 (<unknown>)"  location: "JS
[12:36pm] mschroeder: Looks like a timezone problem.
[12:36pm] ctalbert: Archaeopteryx:  you're trying 380291?
[12:36pm] Archaeopteryx: yes, let me test
[12:37pm] ctalbert: mschroeder: If there are timezone issues between 0.3.1 and 0.5,
 that would be very bad.
[12:37pm] ctalbert: chiller2: Did you use this ICS calendar with 0.3.1?
[12:37pm] Archaeopteryx: oh forget it, it was from the basic. ics from the second bug
[12:37pm] ctalbert: k
[12:38pm] chiller2: ctalbert: yes, it's a .ics I use daily on 0.3.1.
[12:38pm] ctalbert: chiller2:  So, did this situation ever happen:
[12:38pm] ctalbert: Access with 0.5
[12:38pm] ctalbert: Access with 0.3.1
[12:38pm] ctalbert: Access with 0.5?
[12:39pm] ctalbert: Basically, accessing with one version of the calendar and then
 a different version of the calendar.
[12:39pm] chiller2: Yes.
[12:39pm] ctalbert: I'm going to add that to the bug and mark it as QAWanted.
[12:39pm] • ctalbert worries this may be an upgrade issue.
[12:40pm] ctalbert: Although, not much changed in the ICS provider between those 
two versions
[12:40pm] chiller2: Each time I went to a new nightly to see if the problem had
 gone away I essentially started in testing profile w 0.5, tried to drag test event
  on that cal from one date to another. It'd fail with that message in error log. 
  I'd go back to my 0.3.1 profile and carry on.
[12:42pm] ctalbert: I'll bet that version change might have something to do with 
the failure.
[12:42pm] chiller2: And each time before I can go back to 0.3.1 I have to go to task
 manager, kill the TB process, and restart TB.
[12:42pm] ctalbert: ouch
[12:43pm] ctalbert: I updated 380291
[12:43pm] ctalbert: We're over our hour for the QA Chat meeting.  Does anyone have
 anything else they'd like to bring up?
[12:44pm] Archaeopteryx: can you attach a calendar with the test event, please?
[12:44pm] mschroeder: there are already some bugs with "nsIHttpChannel.requestSucceeded" in comments
[12:44pm] ctalbert: chiller2: Thanks again for dropping in.  You're always welcome.
[12:44pm] mschroeder: maybe some of them describe the same problem
[12:45pm] ctalbert: mschroeder: link to the search?
[12:45pm] mschroeder: chiller2: Have a look at those ( bugs.
[12:45pm] ctalbert: thanks
[12:45pm] chiller2: Looking now.
[12:45pm] mschroeder: I'm fast 
[12:46pm] ctalbert:
[12:49pm] ctalbert: Seems to be a kind of general error that the ICS provider uses
[12:52pm] ctalbert:
[12:53pm] ctalbert: I don't see that any of these bugs are due to mixing calendar 
versions while using the same ICS
[12:54pm] ctalbert: chiller2: If you could attach a sample event from your ICS calendar 
(please take the personal data out) to bug 380291, that would help us get it tested easier.
[12:54pm] firebot: ctalbert: Bug maj,
 --, ---,, UNCO, Event movement failure on local or remote events