MathGroup Archive 2012

[Date Index] [Thread Index] [Author Index]

Search the Archive

Re: Mathematica 8 + OS X + McAfee = trouble; help?

  • To: mathgroup at smc.vnet.net
  • Subject: [mg124643] Re: Mathematica 8 + OS X + McAfee = trouble; help?
  • From: Murray Eisenberg <murray at math.umass.edu>
  • Date: Fri, 27 Jan 2012 06:16:01 -0500 (EST)
  • Delivered-to: l-mathgroup@mail-archive0.wolfram.com
  • References: <jfm05k$jt3$1@smc.vnet.net> <201201251205.HAA06033@smc.vnet.net> <201201260826.DAA23929@smc.vnet.net>
  • Reply-to: murray at math.umass.edu

I've now tracked down the Java runtime on OS X that's causing the 
hang-up for JLink due to McAfee on-access scan:

    /System/Library/Java/JavaVirtualMachines/1.6.0.jdk

Now, excluding both that package _and_ $InstallationDirectory and 
$BaseDirectory from on-access scans circumvents the JLink problems.

It's still strange that: (1) I have to exclude the Mathemtica files, too 
(excluding only the Java runtime does not suffice); and (2) this did not 
always happen (which suggests that something in McAfee updates changed 
things).


On 1/26/12 3:26 AM, Murray Eisenberg wrote:
> OK, tracking down and excluding the Java run-time jar may help.
>
> But on OS X, there seems to be no file named rt.jar. The closest
> candidate I've found is:
>
>
> /System/Library/Frameworks/JavaVM.framework/Versions/A/Frameworks/JavaRuntimeSupport.framework/Versions/A/Resources/Java/JavaRuntimeSupport.jar
>
> However, excluding that from the McAfee on-access scan does not cure the
> problem.
>
> I'll investigate further.
>
> On 1/25/12 7:05 AM, Oleksandr Rasputinov wrote:
>> On Tue, 24 Jan 2012 10:12:04 -0000, Murray Eisenberg
>> <murray at math.umass.edu>   wrote:
>>
>>> Under Mac OS X 10.6.8 (Snow Leopard) I've been having persistent
>>> troubles getting the Mathematica 8.0.4 front end to connect with the
>>> kernel when I start the program, and to start the PacletManager, too.
>>> Evidently, the problem is caused by JLink not starting or running
>>> properly.
>>>
>>> After substantial direct help from Wolfram tech support, we've traced
>>> the problem to McAfee Security, specifically, Version 1.1.0 (1309), with
>>> its anti-malware version 9.1.0 (4478), engine version 5400.1158.
>>>
>>> Has anybody else encountered this?
>>>
>>> What happens is that with normal on-access scan enabled, the JLink
>>> problem occurs; with on-access scan disabled, the problem disappears.
>>>
>>> I'd like to be able to keep McAfee Security and simply specify the
>>> needed file exclusions from on-access scan. We've tried to exclude the
>>> entire Mathematica.app package (i.e., $InstallationDirectory) as well as
>>> /Library/Mathematica (i.e., $BaseDirectory), and even the entire
>>> $UserBaseDirectory. But that doesn't do the trick.
>>>
>>> Perhaps the exclusions need to include specific files in those
>>> directories, or even some Java files/directories. The Mac in question
>>> has the usual Apple 64- and 32-bit Javas, version 1.6.0_29-b11-402.
>>>
>>> Any suggestions?
>>>
>>
>> A while ago I encountered a similar problem (probably the same, in fact).
>> McAfee's software seems to have a problem with scanning the large .jar
>> files needed by the Java runtime, which of course is needed for the
>> Mathematica front end. Specifically, McAfee takes a very long time (with
>> certain definition versions) to scan rt.jar (of certain versions of Java)
>> and eventually gives up, but in the mean time J/Link times out. I'm not
>> very familiar with Macs or how the Mathematica installation is laid out on
>> that platform, but I would think you can use an exclusion for rt.jar,
>> wherever that may reside. Actually, to be frank, I always found McAfee to
>> be extremely slow in general and to use an unjustifiably large amount of
>> memory, which caused me some difficulty in using Mathematica even without
>> the added annoyance of this particular problem. As there are many other
>> anti-virus programs available and pretty much every one of them (except
>> Norton/Symantec) is better in every respect than McAfee, I'd certainly
>> encourage you not to rule out that solution.
>>
>

-- 
Murray Eisenberg                     murray at math.umass.edu
Mathematics & Statistics Dept.
Lederle Graduate Research Tower      phone 413 549-1020 (H)
University of Massachusetts                413 545-2859 (W)
710 North Pleasant Street            fax   413 545-1801
Amherst, MA 01003-9305



  • Prev by Date: Re: Mapping a curve to a surface using Manipulate
  • Next by Date: Re: Compile function and AppendTo for lists (vrs. 8.0.4)
  • Previous by thread: Re: Mathematica 8 + OS X + McAfee = trouble; help?
  • Next by thread: Re: Mathematica 8 + OS X + McAfee = trouble; help?