[OLPC New Zealand] [Testing] Testing Summary: 3 March 2012 - Auckland, New Zealand

Deepak Muddha deepak at laptop.org.au
Mon Mar 12 22:40:09 EDT 2012


Hi all

For the defect

   -  socialcalc - started entering data, tried to do some formatting and
   got the spinning wheel of death on the cursor and data can no longer be
   entered.

Here is the link for the above defect raised:
http://dev.laptop.org.au/issues/1137

Please go through 4th and 6th comments in that link.Those are the testing
steps followed to test this defect. Please confirm whether the testing
steps followed were correct or not according to the defect raised.

If not correct, then please explain in a brief on what exactly the defect
meant.

Regards

Deepak
On 3 March 2012 16:35, Tabitha Roder <tabitha at tabitha.net.nz> wrote:

> Testing Summary: 3 March 2012 - Auckland, New Zealand
> Who: Alana, Charlotte, Eliot, John, Fabiana, Oliver, Leslie, Tabitha, Tom
>
> This week we tested build "29 customized" on XO-1.75 and "188 11.3.1-au"
> on XO-1.5
>
> Broken in os29 on XO-1.75 (details below):
>
>    - Raised #11668 for record high quality video problems
>    - We had several XO-1.75s hang while idle
>    - Distance is still broken on XO-1.75
>    - Yum complains the rpmdb has problems
>    - wake-on-lan works well but prevents laptops from going to sleep,
>    power logs are attached
>
>
> Broken in au188 on XO-1.5 (details below):
>
>    - Speak’s toolbar is squashed, see photo on picasa
>
>
>    - Musicpainter on au188 still doesn’t work
>    - Font in terminal on au188 still to big
>
>
> Google Docs is still unstable on Browse on both builds but it’s better on
> Firefox 10 on au188.
>
> Last week we successfully installed a new wifi card in Tank, thank you
> Sridhar for sending it to us! Next week we’ll do Poppy Copy.
>
> Last week we tested au183 and among other things, we used Screencast to
> record some Physics. Screencast seems to make the display update somewhat
> jerkely. Encoding the resulting screencast took a long time and seemed to
> stop at 35%, I don’t know if it aborted or the progress bar was mesleading.
> Something appeared in the journal, but I couldn’t play it -- error could
> not determine type of stra3. - gstypefindelement.c(8-984) …. Can’t typefind
> empty stream. We will follow this up next week.
>
> With os28 and several times in the last few months, I’ve noticed that the
> wifi password window steals the focus but does not appear at the front,
> worse, clicking on other windows does not restore the focus to the
> front-most application. It appears to be an invisible modal dialog.
> Switching to it with alt-tab brings it to the front and everything returns
> to normal.
>
> os29 on Clyde:
>
>    - Physics- still does that lack of update thing when I draw a line - I
>    see the cursor line but no line is drawn. I don’t seem to have this problem
>    with the other shapes. The response time also becomes really really long if
>    there are too many shapes!
>    - Measure: waveform ok, fft ok. Tried sampling - I get the constant
>    waveform and then  every 30 seconds it appears to freeze but only for a
>    couple of seconds and then returns to the live stream. The sampled frame
>    appears to be recorded in the journal. This means that one cannot start
>    looking at the data until the experiment ends. It would make more sense to
>    let the captured image frozen on screen and replaced every 30 seconds so
>    children can do measurements on them or have a split screen. Can’t really
>    tell if increase amplitude (bar on the right) is actually working
>    - Paint: froze completely, with no screen refresh when switching
>    applications. GDB doesn’t seem to work, it found two threads, but claimed
>    both of them had corrupt stacks.
>    - yum complains
>
>
> Warning: RPMDB altered outside of yum.
>
> ** Found 1 pre-existing rpmdb problem(s), 'yum check' output follows:
>
> mesa-libGL-7.9-6.fc14.armv5tel has missing requires of
> mesa-dri-drivers(armv5tel-32) = ('0', '7.9', '6.fc14')
>
>    - When you install gcc you get kernel-headers-2.6.40.3-0.fc14.armv5tel
>    while the running kernel is version 3.0.19. Aftering writing a simple
>    program in C with a couple of functions and loops, we find gdb behaves
>    normally on this .
>
>
>
> os29 on Sundance:
>
>    - Record - photo ok. audio ok, movie on low ok for the full 2 minutes
>    but quality very poor. Movie on high quality has noticeably slowed down by
>    10 seconds and eventually stops. Raised olpc #11668.
>    - Memorise 36 can play, create games save and play them- did not test
>    collaboration.
>    - Measure seems to work ok (same as in clyde).
>    - Typing turtle: seems to be working ok.
>    - Does not respond to shutdown through right click on cetral XO figure
>    in the ring
>
>
>
> os29’s Distance on Kid and Bonnie
> Still very broken as reported earlier in #11596
>
> os29 on Kid
>
>    - paint works - tested opening from journal and activity ring, tested
>    some tools
>
>
>    - record works - tested picture, can open from journal and activity
>    ring
>    - maze works
>    - fototoon works
>
>
>
> os29’s wake on lan seems to be working well, but prevents the laptops from
> suspending for very long. Two laptops on an adhoc network do go to sleep
> but wake up frequently, 5 laptops on infrastructure wifi with internet
> access don’t really go to sleep at all. Looking in the power log is
> difficult because it seems to be written only occasionally? It also doesn’t
> clearly explain that a wakeup is due to lan. Power logs are attached but we
> didn’t take care to record what was happening when. Would such a test be
> useful? My power log collector does not work on XO-1.75 :(
>
> au188 on Tank  XO-1.5 C2
>
>    - speak and robot works but the toolbar is squashed, see
>    - maze works (didn’t test collaboration)
>    - record works - high, low quality, video, audio and image tested
>    - memorize - tested preloaded games, created a game but couldn’t save
>    it so it is lost (very disappointing to lose your work, it’s not in the
>    journal)
>    - ruler works
>    - music painter - no sound or play button doesn't work - cpu usage is
>    now 100% and csoundxo is using it - as reported previously
>    - moon works
>    - clock works
>    - poll works. improvements - ability to edit the questions and
>    possible answers after you have saved it but before someone votes. good
>    that you can hide the results as people answer so not influenced by other
>    voters, would be good if you can view them without all votes being
>    collected, and would be good if you could change number of votes required
>    after voting has begun (so if you forgot to change from 20 you can fix the
>    mistake). would be good if you could export your graphs with the title and
>    column/row headers as images for use in portfolio or write to make reports
>    and presentations.
>    - arithmetic - did all levels and tested using addition and
>    subtraction with an 11 year old, seemed to work, teacher in the background
>    didn't like it as she didn't like the panic it created in answering
>    questions
>    - socialcalc - started entering data, tried to do some formatting and
>    got the spinning wheel of death on the cursor and data can no longer be
>    entered
>    - spirolaterals -  tested on an 11 year old, she got it pretty quick
>    with a bit of help, teacher likes this one and how you can experiment until
>    you figure out the pattern, the child is doing some thinking aloud
>    - system resources - what is the * for?
>    - feedback - where does this go? how do i choose between personalized
>    (american spelling) and anonymous submit? submitted something but don't
>    know what that did, I saw the pulsing icon in the top left corner for few
>    seconds
>
>
> We’re using Browse to do collaborative editing of the testing report in
> google docs, it’s not working so well. I’ve successfully done collaborative
> editing for about 3 hours on a different wifi network with a different
> document and different collaborators on an XO-1.75 with Browse, but
> consistently, the testing report eventually encounters a slow script
> warning and usually browse appears to freeze shortly after. This happens
> both on the XO-1.5 with the Australian builds and on the XO-1.75. Perhaps
> the slow script timeout is too long? Firefox 10 on Rosella (XO-1.5) seems
> to operate google docs a lot better than browse. Not only is it much more
> responsive but is much more stable. This isn’t an option on the XO-1.75
> because Firefox don’t provide an arm build for Linux (they do for Android
> but that does not help).
>
> Kind regards
> Tabitha Roder
>
>
> _______________________________________________
> Testing mailing list
> Testing at lists.laptop.org
> http://lists.laptop.org/listinfo/testing
>
>


-- 
*_______________________________________________________
*

*Deepak Muddha
Software Test Engineer
*
*
One Laptop per Child Australia*
*
M: 91 8897367762
E: deepak at laptop.org.au
W: **www.laptop.org.au* <http://www.laptop.org.au/>* <deepak at laptop.org.au>*
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.laptop.org/pipermail/olpc-nz/attachments/20120313/813ef008/attachment-0001.html>


More information about the OLPC-NZ mailing list