Calendar:QA Chat:2006-01-04:Log
From MozillaWiki
[11:29] <xFallenAngel> I have to go get food, so I wont make it to the beginning of the meeting. Since Sebo is not here, this is the result of the FTP QA: The Problem with FTP is that multiple channels are used, one for upload and one for download. Therefore we get multiple dialogs. I closed bug 312533 as WFM, noting that we need a spinoff bug, but I haven't had time to make it and I [11:29] <firebot> xFallenAngel: Bug https://bugzilla.mozilla.org/show_bug.cgi?id=312533 maj, --, ---, nobody@mozilla.org, RESO WORKSFORME, ftp support is gone in sunbird-1.2 [11:30] <xFallenAngel> don't know if Sebo wants to do that. As soon as we have one channel for both up and download, this will also resolve the bugs regarding max connection limit being reached. [11:33] <ctalbert> Good [11:34] <ctalbert> Sounds good. I'm glad we've gotten this figured out. [11:34] <ctalbert> Let's go on and get started with the meeting then [11:34] <ctalbert> Here is the agendahttp://wiki.mozilla.org/Calendar:QA_Chat:2007-01-04 [11:34] <ctalbert> er: http://wiki.mozilla.org/Calendar:QA_Chat:2007-01-04 [11:35] <ctalbert> We had a good test day on Tuesday. I should get the statistics up on the wiki sometime today. [11:35] <ctalbert> So, let's jump into QAWanted defects. [11:35] <ctalbert> https://bugzilla.mozilla.org/buglist.cgi?query_format=advanced&product=Calendar&keywords=qawanted [11:36] <ctalbert> Any news on 340488? [11:38] <ctalbert> 339960 is openXChange. I don't believe this happens on other webdav servers, so I am leaning toward marking this as a server specific issue, and removing the QAWanted flag from it. I hesitate to mark it WFM since we don't have an openXChange installation to test it against. [11:40] <ctalbert> Bug 255995 is also a webdav specific issue, but I imagine that has probably been fixed by changes in apache or Sunbird since it was filed against old versions of both. [11:40] <firebot> ctalbert: Bug https://bugzilla.mozilla.org/show_bug.cgi?id=255995 nor, --, ---, nobody@mozilla.org, UNCO, first calendar reload stalls when Content-Encoding: gzip used [11:41] <ctalbert> Bug 270915 is pretty simple to verify [11:41] <firebot> ctalbert: Bug https://bugzilla.mozilla.org/show_bug.cgi?id=270915 min, --, ---, nobody@mozilla.org, NEW, 800x600 resolution not supported well [11:41] <ctalbert> It's probably fixed by 360947 [11:42] <ctalbert> 285779 is another webdav specific bug [11:43] <ctalbert> Looks like Bruno answered my questions regarding what they want for QA in 308569. Anyone have experience with wireshark? [11:44] <ctalbert> I'm thinking to let that one stay on the list until we see dmose again and get his input on it. [11:44] <ctalbert> (as well as the input we've had from Bruno) [11:45] <ctalbert> Looks like xFallenAngel has marked 363363 as WFM. I'm going to change its resolution and remove QAWanted. [11:47] <firebot> ctalbert.moz@gmail.com set the Resolution field on bug 363363 to WORKSFORME. [11:47] <firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=363363 nor, --, ---, nobody@mozilla.org, RESO WORKSFORME, password manager continues to pop up when set to remember ftp password [11:47] -->| mschroeder (mozilla@moz-9B00BD1A.dip0.t-ipconnect.de) has joined #calendar-qa [11:47] <ctalbert> Hi mschroeder [11:47] <mschroeder> hi [11:48] <ctalbert> We are currently looking at QAWanted bugs, and at bug 263920 in particular. [11:48] <firebot> ctalbert: Bug https://bugzilla.mozilla.org/show_bug.cgi?id=263920 nor, --, ---, nobody@mozilla.org, UNCO, deleting calendar question [11:48] -->| ulf_home (chatzilla@moz-8CE3E7E7.dip.t-dialin.net) has joined #calendar-qa [11:48] <ctalbert> Hi Ulf [11:49] <ctalbert> We are currently looking at QAWanted bugs, and at bug 263920 in particular. [11:49] [ERROR] No match for “mshro”. [11:50] <ulf_home> Hi, ctalbert, all (sorry I'm late - train delay) [11:50] <ctalbert> mschroeder: This bug actually sounds like it is a good match for the "Delete item dialog" discussion that took place on the newsgroup the a couple of weeks ago. [11:50] <ctalbert> ulf_home: no problem. [11:51] <ctalbert> I don't recall if a bug was ever logged for that discussion or not.... [11:51] <mschroeder> I'm on the phone. [11:51] <ctalbert> ok [11:54] <ctalbert> Okay. The redesign of the delete functionality is referenced in two bugs: 363932 and 320178 [11:54] <ctalbert> that's more UI specific. [11:56] <ctalbert> I think we can simply dupe 263920 against 351499, since 351499 is more recent and addresses the underlying issue a little better. [11:56] <ctalbert> Thoughts? [11:57] <ctalbert> Ok. That's what I did. [11:58] <firebot> ctalbert.moz@gmail.com set the Resolution field on bug 263920 to DUPLICATE of bug 351499. [11:58] <firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=351499 nor, --, Sunbird 0.5, nobody@mozilla.org, NEW, Sort out "delete calendar" vs. "unsubscribe" UI hilarity [11:58] <xFallenAngel> sounds good to me [11:58] <ctalbert> thanks [11:58] <mschroeder> I'm back. [11:59] <ctalbert> Cool [11:59] <ctalbert> Next on the list is 271402 [11:59] <ctalbert> I agree with mschroeder and Omar --> invalid [12:01] <firebot> ctalbert.moz@gmail.com set the Resolution field on bug 271402 to INVALID. [12:01] <firebot> Bug https://bugzilla.mozilla.org/show_bug.cgi?id=271402 nor, --, ---, nobody@mozilla.org, RESO INVALID, Modify calendar dialog to include new features (autoreload/shared) [12:01] <xFallenAngel> same here [12:01] <ctalbert> bug 295637 gets to the heart of what we were talking about on the test day [12:01] <firebot> ctalbert: Bug https://bugzilla.mozilla.org/show_bug.cgi?id=295637 maj, --, ---, nobody@mozilla.org, UNCO, Calender will not automatically publish when a change is made to it. [12:02] <ctalbert> Whether Publish is a one time upload or whether publish links your local calendar to some online store and continuously updates that online store. [12:02] <ctalbert> The current design is that Publish is a one time upload. [12:03] <ctalbert> oh wait, I'm wrong, aren't I. This bug has to do with a checkbox for "automatically publish". I don't think that exists anymore. Does it? [12:04] <xFallenAngel> no, its gone afaik [12:04] <xFallenAngel> I checked in 2007010305 [12:05] <ctalbert> yep, it's gone. Ok. This is invalid. [12:05] <jminta> yeah, it disappeared when the pref-rewrite landed [12:05] <ctalbert> thx [12:06] <firebot> ctalbert.moz@gmail.com set the Resolution field on bug 295637 to INVALID. [12:06] <ctalbert> Bug 312533 is our favorite FTP bug. It's been marked as WFM, but still has qawanted listed. Was that on purpose or can I clear QAWanted status. [12:06] <firebot> ctalbert: Bug https://bugzilla.mozilla.org/show_bug.cgi?id=312533 maj, --, ---, nobody@mozilla.org, RESO WORKSFORME, ftp support is gone in sunbird-1.2 [12:07] <xFallenAngel> no, not on purpose. You can clear qawanted. I would like a new bug on that [12:07] <ctalbert> done [12:07] <xFallenAngel> since that bug is pretty messy. First it talks about ftp not working, then about usernames with dots. [12:07] <ctalbert> Yes, a new bug would be very helpful for that issue. [12:08] <ctalbert> Thanks for working with it and Sebo to get that thing closed. It's been on the list a while. [12:08] <xFallenAngel> sure thing :) [12:08] <mschroeder> Do we need two follow-up bugs, one on usernames with dots and one on multiple dialogs? [12:09] <ctalbert> I think both those issues are WFM now [12:09] <ctalbert> The only remaining issue is the double channel problem, as I understand it. xFallenAngel -- is that correct? [12:10] <xFallenAngel> I believe so, yes. [12:10] <xFallenAngel> Wait I think Sebo mentioned something else that remains. Let me see if I have a history on irc [12:11] * ctalbert wishes he could have been online for that discussion yesterday [12:12] <xFallenAngel> Hmm no history :/ I know that the max connections and multiple dialog think has the same origin, but I think there was something else. [12:12] <ctalbert> ok. We can ask Sebo about that. [12:12] <xFallenAngel> mschroeder: I haven't been able to confirm the usernames with dots thing. It works fine with proftpd as a server [12:12] <ctalbert> What about this last bug: bug 347128. We have some input from the reporter (which is nice!). [12:12] <firebot> ctalbert: Bug https://bugzilla.mozilla.org/show_bug.cgi?id=347128 nor, --, ---, nobody@mozilla.org, UNCO, Changes to a remote calendar using ftp (without providing the username/password in the URL) are not [12:13] <ctalbert> This one looks like it has to do with the settings on his specific FTP server. [12:14] <mschroeder> xFallenAngel: okay [12:14] <xFallenAngel> I think so too. I am using proftpd and couldn't reproduce his steps [12:14] <xFallenAngel> but it is strange indeed [12:15] <ctalbert> Is there a set of steps we can have him run that would proove that it is his server configuration that is incorrect? [12:16] <ctalbert> Is there a way to replicate what Sunbird should be doing in this case with just FTP commands from a prompt? [12:17] <ctalbert> If we could ask him to run those, and it still fails with the same error, then we know that sunbird is not the culprit. If those commands succeed, then we know that sunbird (or necko, more likely) is doing something odd [12:18] -->| damian (chatzilla@moz-A6C96649.kosson.com) has joined #calendar-qa [12:18] <mschroeder> good idea [12:18] <ctalbert> xFallenAngel: Can you put a test case like that together by having your FTP server log what happens when you attempt the steps in this bug from Sunbird? [12:19] <xFallenAngel> I'm sure there is. I can do that, yes. [12:19] <ctalbert> Thanks. I think that would be a good thing to do, since we have a reporter that is responsive. And that will conclusively tell us what is going on. [12:20] <ctalbert> Cool. Now we only have 7 QAWanted bugs!! :-D [12:21] <ctalbert> We still have 200+ Unconfirmed bugs. Any ideas on how to address those? [12:21] <ctalbert> Force the seneca students to do more QA? [12:21] <ctalbert> (that's one idea) [12:22] <xFallenAngel> hehe. I guess theres no other possibility than to do more testdays or let the seneca students do it. [12:22] <ctalbert> I can probably address some of those after lilmatt and I get iTIP submitted for review. [12:22] <xFallenAngel> but I dont really think we will have more people here if we do more testdays [12:22] <lilmatt> :) [12:22] <ctalbert> Yeah, that's a good idea. [12:23] <lilmatt> It depends on the publicity. When we hit slashdot and digg we had a ton of people here [12:23] <ctalbert> That's true. I need to get more of the people from the test days to be more involved on non-test days. [12:23] <xFallenAngel> like me, I came from slashdot :P [12:23] <ctalbert> True. But, it's hard to get publicity without doing something different. [12:23] <lilmatt> If you want the prize, you need to be at the meeting! [12:23] <ctalbert> There's an idea [12:24] <ctalbert> heh [12:24] <damian> there are also many confirmed that are becomming as a duplicate soon or later - so resolving confirmed bugs are also quite nice to redece new issiues [12:24] <mschroeder> maybe I find some more unconfirmed bugs for fast qa :) [12:24] <ctalbert> good ideas [12:25] <damian> ctalbert: are you happy of result after one of the latest test day when we cleaned up bugzilla and unconfirmed bugs? good progress? [12:25] <damian> what i mean is the question - is it worth to spend time for cleaning up bugzilla? [12:26] <lilmatt> Yes [12:26] <ctalbert> Yes, we've always made good progress. The problem is that there are always many more unconfirmed bugs going into bugzilla than we clean out. [12:26] <damian> yes was the answer for which question? ;) [12:26] <ctalbert> We've actually done a great job holding the number constant around 200 or so. [12:26] <ctalbert> And it's been decreasing. It's now something like 250 or so, I think. [12:27] <ctalbert> So, we're doing a great job. I think we just need more people to help out. [12:27] <damian> how/can we reduce new unconfirmed bugs? not cleaning up but prevent [12:28] <ctalbert> The fact that we are holding that number pretty constant, and the fact that everyone on the calendar QA team are volunteers is a truly amazing accomplishment. [12:28] <mschroeder> how many of these unconfirmed bugs are enhancement requests? [12:28] <ctalbert> Bunches [12:28] * ctalbert queries [12:28] <damian> good question [12:28] <ctalbert> 266 total [12:29] <ctalbert> 118 enhancement [12:29] <ctalbert> That's what makes it hard, enhancement bugs are not easy to confirm. [12:29] <damian> many, many [12:29] <bbbrowning> the enhancement requests won't all be marked as such, either, witness the one we just INVALIDed where they wanted autopublish [12:29] <damian> well, some of them are just request for new feature so it's easy [12:30] <ctalbert> bbbrowning: true [12:31] <xFallenAngel> Isnt it enought to make sure the non-enhancement bugs are confirmed? I mean to confirm the enhancements, we need to know if that feature is something thats wanted. [12:31] <ctalbert> Yes, I'd be happy if all the unconfirmed bugs were enhancement requests. [12:32] <xFallenAngel> and I think enh bugs should only be confirmed if they are planned for the next release, but thats only my opinion, I dont know what the official take is on that. [12:32] <damian> well this one we can verify quite easy, don't we? [12:32] <damian> you suggest to have a look what's in the road map? Agreed [12:33] <ctalbert> xFallenAngel: I tend to think we should confirm them if they will be included in 1.0. [12:33] <ctalbert> damian: Yes, we can use the roadmap for that. [12:34] <damian> I guess it's not dificult but decision should be taken by someone how manages road map - just to be sure... [12:34] <ctalbert> These are good ideas. Thanks for your help. [12:34] <ctalbert> We're just about out of time. Is there anything else that folks want to discuss? [12:35] <xFallenAngel> Yes I agree with damian. [12:35] <ctalbert> Me too. In order to confirm enh bugs you almost need to have everyone in the same room, so you can discuss each one easily. [12:35] <damian> so the question is more how to dispatch it and less who is going to do it [12:36] <damian> true [12:36] * ctalbert wonders if he can use this type of logic to get a trip to Europe. [12:36] <ctalbert> :-p [12:36] <ulf_home> :-) [12:38] <ctalbert> I agree with you Damian (dispatch...) [12:38] <damian> I will have a look which bugs can be confirmed - hope many [12:38] <ctalbert> Ok thanks. If you run into something you're not sure about, leave it unconfirmed and mark it QAWanted [12:38] <ctalbert> That way we can take a look at it during this meeting. [12:39] <damian> it's not the best idea [12:39] <damian> it would be nice to see differences betwen QAWanted and unconfirmed bugs just to know priority - first qawanted then if you have time have a look the rest [12:40] <ctalbert> That's a good point [12:40] <damian> otherelse you may see next week 30 qawanted but only a few of them need your urgent help [12:40] <damian> can we use whiteboard? [12:40] <damian> something like [litmus testcase wanted] [12:41] <ctalbert> Yeah, that's true. Either that, or add a comment to them with my nick in there, and I'll see them in my bugmail, and keep a tally of things for the next meeting. [12:42] <damian> sorry, not clear what is your intention [12:42] <ctalbert> Let's use the whiteboard, that would be easiest. How about [qa discussion needed] [12:43] <damian> ok, makes sense, sounds better ;) [12:43] <ctalbert> Alright. I should get the stats posted from Tuesday's test day a little later. [12:43] <ctalbert> Thanks for all your help. [12:44] <ctalbert> Let's make this the end of the QA chat (since we're a bit over our ending time).
