On 22 July 2012 22:49, Chris <[email protected]> wrote:

>
>
> On Sunday, July 22, 2012 2:48:10 PM UTC-7, Chris wrote:
>>
>> I installed process hacker and watched it freeze.  I have no idea what
>> any of this means, or how to report it here.  Nothing really seemed to
>> change in the threads tab, except CPU usage went up to ~50%.  I discovered
>> if I double clicked on any of the threads, this would further bring up some
>> sort of stack of those threads, but they were constantly changing and still
>> working on something, so I don't know how to report them.  Attached are a
>> bunch of pictures during a freeze.  No idea if any of them are useful.
>
>
Great! The main mnemosyne thread stacktrace is most interesting - it
suggests that the sqlite database driver may be running out of memory
(Sqlite returns NOMEM if it can't dynamically allocate memory when
processing a statement)?

I'm not sure why sqlite would run out of memory even on small test decks,
but hopefully Peter will be able to track this down. It'd be nice if there
was a log of the SQL statements Mnemosyne tries to execute - this could be
a bug in the database driver, but there might be a workaround.

Oisín


Like I say, keep in mind in the pictures those secondary "STACK - THREAD
>> XXXX" windows were constantly changing, so the 4 print screens I took of
>> each of the TID entries in the window behind were not at the exact same
>> moment in time, so they may not have all been functioning at the same
>> time.  As Will Ferrel once said when describing The Matrix, "*Ergo*, *
>> concordantly*, *vis-a-vis*. You know what? I *have no idea* what the
>> hell *I'm saying*. "
>>
>>
>>
>> On Sunday, July 22, 2012 6:48:05 AM UTC-7, Oisín Mac Fhearaí wrote:
>>>
>>>
>>>
>>> On 22 July 2012 04:29, Scott Youngman <[email protected]> wrote:
>>>
>>>> I think there are two other indications that the problem is not in the
>>>> cards themselves. First, Peter has examined decks from some users who have
>>>> experienced the problem, and he hasn't reported anything wrong in their
>>>> structure. Second, decks which hang in one person's computer don't hang in
>>>> another user's computer.
>>>>
>>>
>>> Ah, okay. Well that's a shame, as it makes it so much harder to debug.
>>> Are you saying then that even if you create a blank database with 10 dummy
>>> cards, Mnemosyne will eventually hang even on that deck? Or only decks with
>>> some picture/audio cards, etc?
>>>
>>> Have you tried using a tool like Process Hacker (replacement for Windows
>>> task manager) to examine the hung Mnemosyne process?
>>> If you get it to hang, then double-click the Mnemosyne process in
>>> Process Hacker and go to the "threads" tab, you'll see a list of the
>>> different threads and can look at the call stack of each one.
>>>
>>> I'm not sure if you reported whether the hang was a CPU-hogging infinite
>>> loop, or a deadlock or some other bad blocking situation, but this could
>>> help to find out. You might find the main thread is waiting on a call to Qt
>>> or sqlite which never completes?
>>>
>>> The call traces aren't too long, so I'm sure it would be helpful to post
>>> them here.
>>>
>>> Oisín
>>>
>>  --
> You received this message because you are subscribed to the Google Groups
> "mnemosyne-proj-users" group.
> To post to this group, send email to [email protected]
> .
> To unsubscribe from this group, send email to
> [email protected].
> To view this discussion on the web visit
> https://groups.google.com/d/msg/mnemosyne-proj-users/-/tcGZW_6id1oJ.
>
> For more options, visit https://groups.google.com/groups/opt_out.
>
>
>

-- 
You received this message because you are subscribed to the Google Groups 
"mnemosyne-proj-users" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit https://groups.google.com/groups/opt_out.


Reply via email to