ImageJ launches slowly on Lion

[Philip Ershler]

Hi,
        I have noticed a problem with the first launch of 32 bit ImageJ after a reboot. I am running 1.46a and Java 1.6.0_29 [32bit] under OS X 10.7.2. That first launch takes tens of seconds. Subsequently, ImageJ will launch almost instantly. Is anyone else noticing this issue?

Thanks, Phil

Philip R. Ershler Ph.D.
University of Utah
Cardiovascular Research and Training Institute
95 South 2000 East
Salt Lake City, UT 84112-5000

phone: [phone]
alt ph: [phone]
fax: [phone]
e-mail: [hidden email]

[Gabriel Landini]

On Saturday 26 Nov 2011 22:30:04 you wrote:
> Hi,
> I have noticed a problem with the first launch of 32 bit ImageJ after a
> reboot. I am running 1.46a and Java 1.6.0_29 [32bit] under OS X 10.7.2.
> That first launch takes tens of seconds. Subsequently, ImageJ will launch
> almost instantly. Is anyone else noticing this issue?

I think launching Java for the first time is slow in any platform. It is in
win-xp as well as linux at least.
Cheers
G.

[rajava]

It's probably the disk caching in the OS. For the first time i believe the
OS reads from the disk to memory and keeps it there even after you closed
your program until any other program requires the memory. So when you
launch next time it would just use the program in the memory which is why
it's faster, quiet a smart thing to do.

The way to prove is once you launch and stop the program, run purge to
flush the cache from the memory and re-run it and see if it makes any
difference. Note as the purge command sounds scary it will only purge the
cache from the memory and not from the disk :). The command will take few
second to run deponding on the size of the memory kind of freezes ur system
for the duration.



On Sat, Nov 26, 2011 at 2:30 PM, Philip Ershler <[hidden email]>wrote:

> Hi,
>        I have noticed a problem with the first launch of 32 bit ImageJ
> after a reboot. I am running 1.46a and Java 1.6.0_29 [32bit] under OS X
> 10.7.2. That first launch takes tens of seconds. Subsequently, ImageJ will
> launch almost instantly. Is anyone else noticing this issue?
>
> Thanks, Phil
>
> Philip R. Ershler Ph.D.
> University of Utah
> Cardiovascular Research and Training Institute
> 95 South 2000 East
> Salt Lake City, UT 84112-5000
>
> phone: [phone]
> alt ph: [phone]
> fax: [phone]
> e-mail: [hidden email]
>

[Philip Ershler]

I have found the issue with the slow launching of either ImageJ32 or ImageJ64 on OS X 10.7. I have been evaluating Sophos Anti-Virus Free Version. If the On-Access Scanning is set to scan inside archive files (.jar files are actually just  archive files), a routine called InterCheck takes over and causes the delay. Sophos must keep a cache of recently scanned files so that subsequent launches show no delay. As is often the case, anti-virus programs on OS X often cause more trouble than they are worth. Unchecking scan inside archive files resolves the problem.

Phil

On Nov 27, 2011, at 9:51 AM, Raja Kannan wrote:

> It's probably the disk caching in the OS. For the first time i believe the
> OS reads from the disk to memory and keeps it there even after you closed
> your program until any other program requires the memory. So when you
> launch next time it would just use the program in the memory which is why
> it's faster, quiet a smart thing to do.
>
> The way to prove is once you launch and stop the program, run purge to
> flush the cache from the memory and re-run it and see if it makes any
> difference. Note as the purge command sounds scary it will only purge the
> cache from the memory and not from the disk :). The command will take few
> second to run deponding on the size of the memory kind of freezes ur system
> for the duration.
>
>
>
> On Sat, Nov 26, 2011 at 2:30 PM, Philip Ershler <[hidden email]>wrote:
>
>> Hi,
>>       I have noticed a problem with the first launch of 32 bit ImageJ
>> after a reboot. I am running 1.46a and Java 1.6.0_29 [32bit] under OS X
>> 10.7.2. That first launch takes tens of seconds. Subsequently, ImageJ will
>> launch almost instantly. Is anyone else noticing this issue?
>>
>> Thanks, Phil
>>
>> Philip R. Ershler Ph.D.
>> University of Utah
>> Cardiovascular Research and Training Institute
>> 95 South 2000 East
>> Salt Lake City, UT 84112-5000
>>
>> phone: [phone]
>> alt ph: [phone]
>> fax: [phone]
>> e-mail: [hidden email]
>>

Philip R. Ershler Ph.D.
University of Utah
Cardiovascular Research and Training Institute
95 South 2000 East
Salt Lake City, UT 84112-5000

phone: [phone]
alt ph: [phone]
fax: [phone]
e-mail: [hidden email]

[Michael Schmid]

Hi everyone,

it seems that there are also other problems on OS X, apart from  
loading the JVM (a few seconds at most) and problems due to some anti-
virus software:

At least with OS X 10.6 (Snow Leopard), I think that the 'Repair  
permissions' command of the Disk Utilities was the reason for an  
awfully slow start of ImageJ (> 1 minute):
I have never seen the slow start before running 'Repair permissions'  
for the first time, thereafter it came from time to time.

The problem is not well reproducible; sometimes it's the normal  
speed, sometimes slow.
I have the impression that it is related to 'Repair permissions  
removing the 'x' (execute or search directory) permission of  
several .jar files in the Java installation: adding the 'x' flags to  
several .jar files of the Java distribution has improved it, it is  
slow only one of a few dozen times that I launch ImageJ.

Anyhow, it is a nuisance that Apple has not fixed the permissions  
problem, instead they have created a list 'Mac OS X: Disk Utility's  
Repair Disk Permissions messages that you can safely ignore':  http://
support.apple.com/kb/ts1448

Michael
________________________________________________________________

On 28 Nov 2011, at 03:19, Philip Ershler wrote:

> I have found the issue with the slow launching of either ImageJ32  
> or ImageJ64 on OS X 10.7. I have been evaluating Sophos Anti-Virus  
> Free Version. If the On-Access Scanning is set to scan inside  
> archive files (.jar files are actually just  archive files), a  
> routine called InterCheck takes over and causes the delay. Sophos  
> must keep a cache of recently scanned files so that subsequent  
> launches show no delay. As is often the case, anti-virus programs  
> on OS X often cause more trouble than they are worth. Unchecking  
> scan inside archive files resolves the problem.
>
> Phil
>
> On Nov 27, 2011, at 9:51 AM, Raja Kannan wrote:
>
>> It's probably the disk caching in the OS. For the first time i  
>> believe the
>> OS reads from the disk to memory and keeps it there even after you  
>> closed
>> your program until any other program requires the memory. So when you
>> launch next time it would just use the program in the memory which  
>> is why
>> it's faster, quiet a smart thing to do.
>>
>> The way to prove is once you launch and stop the program, run  
>> purge to
>> flush the cache from the memory and re-run it and see if it makes any
>> difference. Note as the purge command sounds scary it will only  
>> purge the
>> cache from the memory and not from the disk :). The command will  
>> take few
>> second to run deponding on the size of the memory kind of freezes  
>> ur system
>> for the duration.
>>
>>
>>
>> On Sat, Nov 26, 2011 at 2:30 PM, Philip Ershler  
>> <[hidden email]>wrote:
>>
>>> Hi,
>>>       I have noticed a problem with the first launch of 32 bit  
>>> ImageJ
>>> after a reboot. I am running 1.46a and Java 1.6.0_29 [32bit]  
>>> under OS X
>>> 10.7.2. That first launch takes tens of seconds. Subsequently,  
>>> ImageJ will
>>> launch almost instantly. Is anyone else noticing this issue?
>>>
>>> Thanks, Phil
>>>
>>> Philip R. Ershler Ph.D.
>>> University of Utah
>>> Cardiovascular Research and Training Institute
>>> 95 South 2000 East
>>> Salt Lake City, UT 84112-5000
>>>
>>> phone: [phone]
>>> alt ph: [phone]
>>> fax: [phone]
>>> e-mail: [hidden email]
>>>
>
> Philip R. Ershler Ph.D.
> University of Utah
> Cardiovascular Research and Training Institute
> 95 South 2000 East
> Salt Lake City, UT 84112-5000
>
> phone: [phone]
> alt ph: [phone]
> fax: [phone]
> e-mail: [hidden email]

[Walter Wolf]

Michael and others:

I have had the same problem with "Repair Permissions" when running on OS
X 10.6. The recommendation I received from an Apple technician is not to
run Repair Permissions, it does not perform any useful function anymore.

On 11/28/11 5:56 AM, Michael Schmid wrote:

> Hi everyone,
>
> it seems that there are also other problems on OS X, apart from
> loading the JVM (a few seconds at most) and problems due to some
> anti-virus software:
>
> At least with OS X 10.6 (Snow Leopard), I think that the 'Repair
> permissions' command of the Disk Utilities was the reason for an
> awfully slow start of ImageJ (> 1 minute):
> I have never seen the slow start before running 'Repair permissions'
> for the first time, thereafter it came from time to time.
>
> The problem is not well reproducible; sometimes it's the normal speed,
> sometimes slow.
> I have the impression that it is related to 'Repair permissions
> removing the 'x' (execute or search directory) permission of several
> .jar files in the Java installation: adding the 'x' flags to several
> .jar files of the Java distribution has improved it, it is slow only
> one of a few dozen times that I launch ImageJ.
>
> Anyhow, it is a nuisance that Apple has not fixed the permissions
> problem, instead they have created a list 'Mac OS X: Disk Utility's
> Repair Disk Permissions messages that you can safely ignore':  
> http://support.apple.com/kb/ts1448
>
> Michael
> ________________________________________________________________
>
> On 28 Nov 2011, at 03:19, Philip Ershler wrote:
>
>> I have found the issue with the slow launching of either ImageJ32 or
>> ImageJ64 on OS X 10.7. I have been evaluating Sophos Anti-Virus Free
>> Version. If the On-Access Scanning is set to scan inside archive
>> files (.jar files are actually just  archive files), a routine called
>> InterCheck takes over and causes the delay. Sophos must keep a cache
>> of recently scanned files so that subsequent launches show no delay.
>> As is often the case, anti-virus programs on OS X often cause more
>> trouble than they are worth. Unchecking scan inside archive files
>> resolves the problem.
>>
>> Phil
>>
>> On Nov 27, 2011, at 9:51 AM, Raja Kannan wrote:
>>
>>> It's probably the disk caching in the OS. For the first time i
>>> believe the
>>> OS reads from the disk to memory and keeps it there even after you
>>> closed
>>> your program until any other program requires the memory. So when you
>>> launch next time it would just use the program in the memory which
>>> is why
>>> it's faster, quiet a smart thing to do.
>>>
>>> The way to prove is once you launch and stop the program, run purge to
>>> flush the cache from the memory and re-run it and see if it makes any
>>> difference. Note as the purge command sounds scary it will only
>>> purge the
>>> cache from the memory and not from the disk :). The command will
>>> take few
>>> second to run deponding on the size of the memory kind of freezes ur
>>> system
>>> for the duration.
>>>
>>>
>>>
>>> On Sat, Nov 26, 2011 at 2:30 PM, Philip Ershler
>>> <[hidden email]>wrote:
>>>
>>>> Hi,
>>>>       I have noticed a problem with the first launch of 32 bit ImageJ
>>>> after a reboot. I am running 1.46a and Java 1.6.0_29 [32bit] under
>>>> OS X
>>>> 10.7.2. That first launch takes tens of seconds. Subsequently,
>>>> ImageJ will
>>>> launch almost instantly. Is anyone else noticing this issue?
>>>>
>>>> Thanks, Phil
>>>>
>>>> Philip R. Ershler Ph.D.
>>>> University of Utah
>>>> Cardiovascular Research and Training Institute
>>>> 95 South 2000 East
>>>> Salt Lake City, UT 84112-5000
>>>>
>>>> phone: [phone]
>>>> alt ph: [phone]
>>>> fax: [phone]
>>>> e-mail: [hidden email]
>>>>
>>
>> Philip R. Ershler Ph.D.
>> University of Utah
>> Cardiovascular Research and Training Institute
>> 95 South 2000 East
>> Salt Lake City, UT 84112-5000
>>
>> phone: [phone]
>> alt ph: [phone]
>> fax: [phone]
>> e-mail: [hidden email]
>

--
Professor Walter Wolf, Ph.D.
Distinguished Professor of Pharmaceutical Sciences
Director, Pharmacokinetic Imaging Program
Department of Pharmaceutical Sciences, School of Pharmacy
Chair, Biomedical Imaging Science Initiative
University of Southern California
1985 Zonal Ave., Los Angeles, CA 90089-9121
E-Mail: [hidden email]
Telephone: [phone]
Fax: [phone]
http://www.usc.edu/research/initiatives/bisi/
http://www.macnis.org/
http://www.usc.edu/schools/pharmacy/faculty_directory/detail.php?id=59