Memory Issues IJ2

[Rebecca Keller]

Dear List,

is anyone else having memory issues in IJ2? Clicking on the status bar does
not free up memory, and also, I've even sometimes not been able to close
IJ2, and had to resort to killing it in the task mgr. Anyone else having
these issues?

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

[ctrueden]

Hi Rebecca,

> I've even sometimes not been able to close IJ2, and had to resort to
> killing it in the task mgr. Anyone else having these issues?

This is a known issue, tracked at:
http://fiji.sc/bugzilla/show_bug.cgi?id=805

We made one serious effort to fix it, and AFAIK it is fixed on OS X, but
there have been subsequent reports of continued problems. I was finally
able to replicate the problem on a Windows VM on my machine, so hope to
have a fix available very soon. Very sorry for the problems.

> Clicking on the status bar does not free up memory

I cannot reproduce this problem. In my tests, clicking the status bar
seems to perform a garbage collection, on both OS X and Windows.

Regards,
Curtis



On Tue, Jul 8, 2014 at 2:27 PM, Rebecca Keller <[hidden email]> wrote:

> Dear List,
>
> is anyone else having memory issues in IJ2? Clicking on the status bar does
> not free up memory, and also, I've even sometimes not been able to close
> IJ2, and had to resort to killing it in the task mgr. Anyone else having
> these issues?
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

[vbindokas]

I have been watching the memory leak in both ImageJ AND Fiji for some time.
closing images and/or running garbage collection doesn't clear memory
holdings in Windows7-64, This gets to be a real problem after dealing
with large images/data. I've seen >14GB claimed by IJ after large stacks
with nothing left open. I also can have IJ/IJ2 not close-shutdown and
have to use the manager to kill it.
     With windows task manager running, I see a clean launcj of IJ
taking 328,236K on my system. After manipulating images, the value
remain higher than this after closing all and running garbage routine.
Try this test macro, it strands memory at 521.5k.

run("Mitosis (26MB, 5D stack)");
run("32-bit");
run("Duplicate...", "title=mitosis-1.tif duplicate channels=1-2
slices=1-5 frames=1-51");
run("Concatenate...", "  title=[Concatenated Stacks] image1=mitosis.tif
image2=mitosis-1.tif image3=[-- None --]");
close();
run("Collect Garbage");


On 7/8/2014 3:12 PM, Curtis Rueden wrote:

> Hi Rebecca,
>
>> I've even sometimes not been able to close IJ2, and had to resort to
>> killing it in the task mgr. Anyone else having these issues?
> This is a known issue, tracked at:
> http://fiji.sc/bugzilla/show_bug.cgi?id=805
>
> We made one serious effort to fix it, and AFAIK it is fixed on OS X, but
> there have been subsequent reports of continued problems. I was finally
> able to replicate the problem on a Windows VM on my machine, so hope to
> have a fix available very soon. Very sorry for the problems.
>
>> Clicking on the status bar does not free up memory
> I cannot reproduce this problem. In my tests, clicking the status bar
> seems to perform a garbage collection, on both OS X and Windows.
>
> Regards,
> Curtis
>
>
>
> On Tue, Jul 8, 2014 at 2:27 PM, Rebecca Keller <[hidden email]> wrote:
>
>> Dear List,
>>
>> is anyone else having memory issues in IJ2? Clicking on the status bar does
>> not free up memory, and also, I've even sometimes not been able to close
>> IJ2, and had to resort to killing it in the task mgr. Anyone else having
>> these issues?
>>
>> --
>> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html

--
__

Vytas Bindokas, Ph.D.
Research Assoc. (Assoc. Prof.),
Director, BSD Light Microscopy Core Facility
phone: [phone]

      [address for letters ONLY (see shipping addr below):]
Dept Pharmacol Physiol Sci MC0926
947 E 58th Street
The University of Chicago
Chicago IL 60637
Room Abbott 129


shipping address (main KCBD site):
V. Bindokas
900 E 57th Street
KCBD room 1250, Microscopy Core
The University of Chicago
Chicago IL 60637


email [hidden email]
web site for LMCF:
http://digital.uchicago.edu/index.html

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

[ctrueden]

Hi everyone,

> closing images and/or running garbage collection doesn't clear memory
> holdings in Windows7-64

Actually, that is a characteristic of Java. The Task Manager will never
report memory decreasing. See:

http://imagej.net/FAQ#Why_does_ImageJ_not_release_any_memory_back_to_the_system.3F

Regards,
Curtis

P.S. I am still actively working on the "ImageJ2 won't quit" issue, getting
very close to a resolution. We will get the fix uploaded ASAP and I'll
follow up with a blog post explaining the nature of the difficulty, for
those curious.


On Tue, Jul 8, 2014 at 5:22 PM, vbindokas <[hidden email]> wrote:

> I have been watching the memory leak in both ImageJ AND Fiji for some time.
> closing images and/or running garbage collection doesn't clear memory
> holdings in Windows7-64, This gets to be a real problem after dealing with
> large images/data. I've seen >14GB claimed by IJ after large stacks with
> nothing left open. I also can have IJ/IJ2 not close-shutdown and have to
> use the manager to kill it.
>     With windows task manager running, I see a clean launcj of IJ taking
> 328,236K on my system. After manipulating images, the value remain higher
> than this after closing all and running garbage routine. Try this test
> macro, it strands memory at 521.5k.
>
> run("Mitosis (26MB, 5D stack)");
> run("32-bit");
> run("Duplicate...", "title=mitosis-1.tif duplicate channels=1-2 slices=1-5
> frames=1-51");
> run("Concatenate...", "  title=[Concatenated Stacks] image1=mitosis.tif
> image2=mitosis-1.tif image3=[-- None --]");
> close();
> run("Collect Garbage");
>
>
>
> On 7/8/2014 3:12 PM, Curtis Rueden wrote:
>
>> Hi Rebecca,
>>
>>  I've even sometimes not been able to close IJ2, and had to resort to
>>> killing it in the task mgr. Anyone else having these issues?
>>>
>> This is a known issue, tracked at:
>> http://fiji.sc/bugzilla/show_bug.cgi?id=805
>>
>> We made one serious effort to fix it, and AFAIK it is fixed on OS X, but
>> there have been subsequent reports of continued problems. I was finally
>> able to replicate the problem on a Windows VM on my machine, so hope to
>> have a fix available very soon. Very sorry for the problems.
>>
>>  Clicking on the status bar does not free up memory
>>>
>> I cannot reproduce this problem. In my tests, clicking the status bar
>> seems to perform a garbage collection, on both OS X and Windows.
>>
>> Regards,
>> Curtis
>>
>>
>>
>> On Tue, Jul 8, 2014 at 2:27 PM, Rebecca Keller <[hidden email]>
>> wrote:
>>
>>  Dear List,
>>>
>>> is anyone else having memory issues in IJ2? Clicking on the status bar
>>> does
>>> not free up memory, and also, I've even sometimes not been able to close
>>> IJ2, and had to resort to killing it in the task mgr. Anyone else having
>>> these issues?
>>>
>>> --
>>> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>>>
>>>  --
>> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>>
>
> --
> __
>
> Vytas Bindokas, Ph.D.
> Research Assoc. (Assoc. Prof.),
> Director, BSD Light Microscopy Core Facility
> phone: [phone]
>
>      [address for letters ONLY (see shipping addr below):]
> Dept Pharmacol Physiol Sci MC0926
> 947 E 58th Street
> The University of Chicago
> Chicago IL 60637
> Room Abbott 129
>
>
> shipping address (main KCBD site):
> V. Bindokas
> 900 E 57th Street
> KCBD room 1250, Microscopy Core
> The University of Chicago
> Chicago IL 60637
>
>
> email [hidden email]
> web site for LMCF:
> http://digital.uchicago.edu/index.html
>
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

[ctrueden]

Hi everyone,

> I am still actively working on the "ImageJ2 won't quit" issue, getting
> very close to a resolution. We will get the fix uploaded ASAP

The quitting problems, I sincerely hope, have finally been fixed if you run
Help > Update.

If anyone notices continued problems with ImageJ not shutting down as it
should, please let us know.

Thanks,
Curtis


On Thu, Jul 10, 2014 at 10:44 AM, Curtis Rueden <[hidden email]> wrote:

> Hi everyone,
>
> > closing images and/or running garbage collection doesn't clear memory
> > holdings in Windows7-64
>
> Actually, that is a characteristic of Java. The Task Manager will never
> report memory decreasing. See:
>
> http://imagej.net/FAQ#Why_does_ImageJ_not_release_any_memory_back_to_the_system.3F
>
> Regards,
> Curtis
>
> P.S. I am still actively working on the "ImageJ2 won't quit" issue,
> getting very close to a resolution. We will get the fix uploaded ASAP and
> I'll follow up with a blog post explaining the nature of the difficulty,
> for those curious.
>
>
> On Tue, Jul 8, 2014 at 5:22 PM, vbindokas <[hidden email]>
> wrote:
>
>> I have been watching the memory leak in both ImageJ AND Fiji for some
>> time.
>> closing images and/or running garbage collection doesn't clear memory
>> holdings in Windows7-64, This gets to be a real problem after dealing with
>> large images/data. I've seen >14GB claimed by IJ after large stacks with
>> nothing left open. I also can have IJ/IJ2 not close-shutdown and have to
>> use the manager to kill it.
>>     With windows task manager running, I see a clean launcj of IJ taking
>> 328,236K on my system. After manipulating images, the value remain higher
>> than this after closing all and running garbage routine. Try this test
>> macro, it strands memory at 521.5k.
>>
>> run("Mitosis (26MB, 5D stack)");
>> run("32-bit");
>> run("Duplicate...", "title=mitosis-1.tif duplicate channels=1-2
>> slices=1-5 frames=1-51");
>> run("Concatenate...", "  title=[Concatenated Stacks] image1=mitosis.tif
>> image2=mitosis-1.tif image3=[-- None --]");
>> close();
>> run("Collect Garbage");
>>
>>
>>
>> On 7/8/2014 3:12 PM, Curtis Rueden wrote:
>>
>>> Hi Rebecca,
>>>
>>>  I've even sometimes not been able to close IJ2, and had to resort to
>>>> killing it in the task mgr. Anyone else having these issues?
>>>>
>>> This is a known issue, tracked at:
>>> http://fiji.sc/bugzilla/show_bug.cgi?id=805
>>>
>>> We made one serious effort to fix it, and AFAIK it is fixed on OS X, but
>>> there have been subsequent reports of continued problems. I was finally
>>> able to replicate the problem on a Windows VM on my machine, so hope to
>>> have a fix available very soon. Very sorry for the problems.
>>>
>>>  Clicking on the status bar does not free up memory
>>>>
>>> I cannot reproduce this problem. In my tests, clicking the status bar
>>> seems to perform a garbage collection, on both OS X and Windows.
>>>
>>> Regards,
>>> Curtis
>>>
>>>
>>>
>>> On Tue, Jul 8, 2014 at 2:27 PM, Rebecca Keller <[hidden email]>
>>> wrote:
>>>
>>>  Dear List,
>>>>
>>>> is anyone else having memory issues in IJ2? Clicking on the status bar
>>>> does
>>>> not free up memory, and also, I've even sometimes not been able to close
>>>> IJ2, and had to resort to killing it in the task mgr. Anyone else having
>>>> these issues?
>>>>
>>>> --
>>>> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>>>>
>>>>  --
>>> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>>>
>>
>> --
>> __
>>
>> Vytas Bindokas, Ph.D.
>> Research Assoc. (Assoc. Prof.),
>> Director, BSD Light Microscopy Core Facility
>> phone: [phone]
>>
>>      [address for letters ONLY (see shipping addr below):]
>> Dept Pharmacol Physiol Sci MC0926
>> 947 E 58th Street
>> The University of Chicago
>> Chicago IL 60637
>> Room Abbott 129
>>
>>
>> shipping address (main KCBD site):
>> V. Bindokas
>> 900 E 57th Street
>> KCBD room 1250, Microscopy Core
>> The University of Chicago
>> Chicago IL 60637
>>
>>
>> email [hidden email]
>> web site for LMCF:
>> http://digital.uchicago.edu/index.html
>>
>>
>> --
>> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>>
>
>

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

[Niko Ehrenfeuchter]

Dear Curtis et al,

On 10.07.2014 21:20, Curtis Rueden wrote:

> Hi everyone,
>
>> I am still actively working on the "ImageJ2 won't quit" issue, getting
>> very close to a resolution. We will get the fix uploaded ASAP
>
> The quitting problems, I sincerely hope, have finally been fixed if you run
> Help > Update.
>
> If anyone notices continued problems with ImageJ not shutting down as it
> should, please let us know.

unfortunately I still do after applying all updates. How can I help in
debugging?

Thanks,
Niko

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

[ctrueden]

Hi Niko,

> unfortunately I still do after applying all updates. How can I help in
> debugging?

Please post details to bug #805 at:
http://fiji.sc/bugzilla/show_bug.cgi?id=805

Of specific interest:

* Help > Report a Bug stuff (especially OS, version of Java, and which
update sites are enabled)

* Exact minimal steps to reproduce, and what goes wrong. E.g.: "I start
Fiji, open the Blobs sample image by pressing Shift+B, then try to quit by
clicking the main window's X button. Nothing happens, the program does not
quit, and no windows disappear."

In particular: how are you trying to quit? X button? File > Quit? This
actually matters. Also, what happens exactly? Do the windows disappear, or
do they stay? This also matters.

Thanks,
Curtis


On Thu, Jul 10, 2014 at 4:09 PM, Niko Ehrenfeuchter <
[hidden email]> wrote:

> Dear Curtis et al,
>
>
> On 10.07.2014 21:20, Curtis Rueden wrote:
>
>> Hi everyone,
>>
>>  I am still actively working on the "ImageJ2 won't quit" issue, getting
>>> very close to a resolution. We will get the fix uploaded ASAP
>>>
>>
>> The quitting problems, I sincerely hope, have finally been fixed if you
>> run
>> Help > Update.
>>
>> If anyone notices continued problems with ImageJ not shutting down as it
>> should, please let us know.
>>
>
> unfortunately I still do after applying all updates. How can I help in
> debugging?
>
> Thanks,
> Niko
>
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

[Paul van Schayck]

Dear Curtis,

On Fri, Jul 11, 2014 at 5:38 AM, Curtis Rueden <[hidden email]> wrote:
> Please post details to bug #805 at:
> http://fiji.sc/bugzilla/show_bug.cgi?id=805

The problem I reported a while a go on this list, where in headless
mode ImageJ doses after about 60 seconds, but Fiji closes immediately,
is also still there after the recent updates.

http://imagej.1557.x6.nabble.com/Headless-ImageJ2-rc2-Analyze-Particles-returns-no-result-tp5008127p5008391.html

And the stack dump: http://pastebin.com/uF9AMKNH

Do you also want additional information on that in that bug report, or
is this problem unrelated? I can replicate it on a wide variety of
systems (Ubuntu, Debian, RHEL, ScientificLinux) in 32 and 64 bit mode.

Thanks,

Paul

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

[ctrueden]

Hi everyone,

> The quitting problems, I sincerely hope, have finally been fixed if
> you run Help > Update.

We have pushed another slew of updates to address the quitting problems.
You should see this as version 2.0.0-rc-9 after running the updater. I
really, really think (hope!) we've nailed it this time.

For those curious, I wrote a blog post with lots of juicy technical details
about what all the fuss is about:

    http://imagej.net/2014-07-11_-_Fiji_won%27t_quit!

Enjoy,
Curtis


On Thu, Jul 10, 2014 at 10:38 PM, Curtis Rueden <[hidden email]> wrote:

> Hi Niko,
>
> > unfortunately I still do after applying all updates. How can I help in
> > debugging?
>
> Please post details to bug #805 at:
> http://fiji.sc/bugzilla/show_bug.cgi?id=805
>
> Of specific interest:
>
> * Help > Report a Bug stuff (especially OS, version of Java, and which
> update sites are enabled)
>
> * Exact minimal steps to reproduce, and what goes wrong. E.g.: "I start
> Fiji, open the Blobs sample image by pressing Shift+B, then try to quit by
> clicking the main window's X button. Nothing happens, the program does not
> quit, and no windows disappear."
>
> In particular: how are you trying to quit? X button? File > Quit? This
> actually matters. Also, what happens exactly? Do the windows disappear, or
> do they stay? This also matters.
>
> Thanks,
> Curtis
>
>
> On Thu, Jul 10, 2014 at 4:09 PM, Niko Ehrenfeuchter <
> [hidden email]> wrote:
>
>> Dear Curtis et al,
>>
>>
>> On 10.07.2014 21:20, Curtis Rueden wrote:
>>
>>> Hi everyone,
>>>
>>>  I am still actively working on the "ImageJ2 won't quit" issue, getting
>>>> very close to a resolution. We will get the fix uploaded ASAP
>>>>
>>>
>>> The quitting problems, I sincerely hope, have finally been fixed if you
>>> run
>>> Help > Update.
>>>
>>> If anyone notices continued problems with ImageJ not shutting down as it
>>> should, please let us know.
>>>
>>
>> unfortunately I still do after applying all updates. How can I help in
>> debugging?
>>
>> Thanks,
>> Niko
>>
>>
>> --
>> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>>
>
>

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html

[ctrueden]

Hi Paul,

> The problem I reported a while a go on this list, where in headless
> mode ImageJ doses after about 60 seconds, but Fiji closes immediately,
> is also still there after the recent updates.

I am guessing that problem is unrelated, although it is hard to be certain
without further investigation.

Regardless, could you please file a bug in the Bugzilla for it? Otherwise I
am afraid it may get lost in the shuffle.

Thanks,
Curtis


On Fri, Jul 11, 2014 at 10:26 AM, Paul van Schayck <[hidden email]>
wrote:

> Dear Curtis,
>
> On Fri, Jul 11, 2014 at 5:38 AM, Curtis Rueden <[hidden email]> wrote:
> > Please post details to bug #805 at:
> > http://fiji.sc/bugzilla/show_bug.cgi?id=805
>
> The problem I reported a while a go on this list, where in headless
> mode ImageJ doses after about 60 seconds, but Fiji closes immediately,
> is also still there after the recent updates.
>
>
> http://imagej.1557.x6.nabble.com/Headless-ImageJ2-rc2-Analyze-Particles-returns-no-result-tp5008127p5008391.html
>
> And the stack dump: http://pastebin.com/uF9AMKNH
>
> Do you also want additional information on that in that bug report, or
> is this problem unrelated? I can replicate it on a wide variety of
> systems (Ubuntu, Debian, RHEL, ScientificLinux) in 32 and 64 bit mode.
>
> Thanks,
>
> Paul
>
> --
> ImageJ mailing list: http://imagej.nih.gov/ij/list.html
>

--
ImageJ mailing list: http://imagej.nih.gov/ij/list.html