[Kst] Out of Memory Error

[Peter Kümmel]

Am 20.07.2014 07:29, schrieb Ben Lewis:
> Hi Barth,
>
> That makes a lot of sense. Every 2..3 minutes after the first error the
> plot will temporarily "glitch" (jump to what appears to be a random
> value). It freezes in the glitched state for a few seconds and then
> everything resumes as normal. Including the glitches which disappear.
> This has never bothered me too much because it recovers with no
> permanent harm.
>
> These glitches must be due to the same error that is initially reported.
>
> I agree that the first error should be removed (just like all subsequent
> errors) because it automatically recovers.

OK, shipped this with 2.0.8.

>
> It would be great if the source of the problem could be rectified also.
> This will get rid of my regular "glitches". Is it possible that I am
> doing something wrong in my application or is this most likely a Kst
> issue? When I write to the data file I open it once and do not close the
> file between write operations. Should I be closing the file after each
> write operation?
>
> This problem occurs when plotting data from a remote PC but a colleague
> of mine has tired it with local data and gets the same result. So I
> don't think it is related to the location of the file.
>
> Regards, Ben
>
>
> On 20/07/2014 3:27 AM, Barth Netterfield wrote:
>> Interesting.
>>
>> This error triggered when the reader fails to read any points from the
>> file,
>> which sounds like some sort of file locking issue with the file
>> system.  I fixed
>> something like this elsewhere (that had a high cross section for
>> messing up)
>> but this one still exists.
>>
>> When this happens, it does reset the buffer.
>> Also, the first time something happens it prints the warning message.
>> After
>> this. the error probably keeps happening occasionally - but it just
>> does so
>> silently, and recovers.
>>
>> I think this should be changed.
>>
>> -It isn't clear that the warning message is particularly helpful -
>> except for
>> debugging.  Can we just remove it?  Or perhaps disable it after at
>> least one
>> successful read.   Either is safe and trivial.
>>
>> -Guessing that this is a file system access issue, I think that the
>> reader
>> should fail quietly, and just try again soon (without resetting) if
>> the read
>> fails here - at least for a couple times.
>>
>> Ben: what is the file system situation?  Remote disk? Other things
>> that might
>> be relevant?
>>
>> Peter: can you decide if you want to disable the error message for
>> 2.0.8.  I
>> can co either way on this.  As long as 2.0.8 goes out Monday or sooner.
>>
>> cbn
>>
>>
>>
>> On July 19, 2014 1:52:09 PM Ben Lewis wrote:
>>
>>> Hi Barth and Peter,
>>>
>>> I can send you instructions on how to reproduce this bug. You will
>>> need a
>>> Windows PC (32-bit) and a demo version of TwinCAT
>>> <http://www.beckhoff.com.au/english.asp?download/default.htm>. I can
>>> then
>>> send you my code to run within TwinCAT. You can them run Kst on the
>>> same PC
>>> or a remote PC.
>>>
>>> Regards, Ben
>>>
>>> On 16/07/2014 10:29 PM, Ben Lewis wrote:
>>>> Hi Peter,
>>>>
>>>> This is the error I get, usually within the first two or three
>>>> minutes of
>>>> plotting live data. It happens every time but only ever once per
>>>> file. It
>>>> appears to trigger a complete reread of the data source.
>>>>
>>>>
>>>>
>>>> Regards, Ben
>>>>
>>>> On 15/07/2014 10:16 PM, Ben Lewis wrote:
>>>>> Hi Peter,
>>>>>
>>>>> I haven't got access to my data logger tonight. I'll check tomorrow
>>>>> and
>>>>> get back to you.
>>>>>
>>>>> Regards, Ben
>>>>>
>>>>> On 15/07/2014 3:54 PM, Peter Kümmel wrote:
>>>>>> On 14.07.2014 13:39, Ben Lewis wrote:
>>>>>>> Hi Kst,
>>>>>>>
>>>>>>> I've just tested the latest build of Kst and I'm still getting
>>>>>>> the "out
>>>>>>> of memory" error. See
>>>>>>> http://kde.6490.n7.nabble.com/Out-of-memory-error-td1555215.html for
>>>>>>> details.>>>
>>>>>> Now there is no message with "out of memory" any more. Which one
>>>>>> did you
>>>>>> see now?>>>
>>>>>>> I know Peter has spent some time looking into this but was not
>>>>>>> able to
>>>>>>> resolve this problem. I thought the problem was
>>>>>>> unique to my paricular set up so I thought it would be ok to let
>>>>>>> it go.
>>>>>>> But a college of mine has just tried Kst for the
>>>>>>> first time, on a different computer, and is getting the same error.
>>>>>>> Also he was using a local data source, whereas I was
>>>>>>> using a remote data source. We are both using the 32-bit version
>>>>>>> of Kst
>>>>>>> on a Windows PC.
>>>>>>>
>>>>>>> I thought I should mention this just in case this should be fixed
>>>>>>> before releasing 2.0.8.
>>>>>>>
>>>>>>> I'm happy help out in any way I can.
>>>>>>>
>>>>>>> Regards, Ben
>>>>>>> _______________________________________________
>>>>>>> Kst mailing list
>>>>>>> Kst at kde.org
>>>>>>> https://mail.kde.org/mailman/listinfo/kst
>>>>>> _______________________________________________
>>>>>> Kst mailing list
>>>>>> Kst at kde.org
>>>>>> https://mail.kde.org/mailman/listinfo/kst
>
> _______________________________________________
> Kst mailing list
> Kst at kde.org
> https://mail.kde.org/mailman/listinfo/kst
>