This is the mail archive of the mailing list for the Cygwin project.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: Win2k and cygwin memory leak

Assuming you're right - and I have no reason to think otherwise, using
cygwin in a memory
intensive fashion does indeed fragment the memory. *cygwin's problem?*  OK
- I'm being
persuaded it isn't - as such. After playing with it some over the last few
hours, I found that the problem is MUCH
worse on NT4 than W2K. (NT4 was my only platform 11 months ago when the
problem was first
discoverd.) Basically NT4 degrades very quickly and I have yet to see it
recover without RAMpage.
W2K degrades also, but if left alone long enough, appears to recover on its
own "eventually".
The *eventually* is the key thing since if the memory fragments "fast
enough" - I can cripple the box
before the OS recovers on its own. This is where RAMpage is useful on W2K.
Perhaps there
is a registry or ini setting that one could tweak to force OS defragmenting
earlier. It's all pretty new
to me. The question is, is this fragmenting and defragging an unavoidable
and inevitable side effect, or can
it be managed from within an application? Can some of this be controlled
with the cygwin daemon of the future?
Anyhow, I too am weary of this thread and since I'm not a c/c++ programmer,
I've exhausted
most of the resources I can use on this question. Nevertheless, if others
find RAMpage helps them to
use cygwin more robustly than without, especially on older OS's like NT4
and win98, then a
mention in the FAQ I think would be quite helpful. But I would think more
and experience from a wider cross section of folks would be needed first.

Brian Kelly

"Christopher Faylor" <> on 08/08/2003 11:15:36

Please respond to

Sent by:

cc:     (bcc: Brian Kelly/WTC1/Empire)

Subject:    Re: Win2k and cygwin memory leak

On Fri, Aug 08, 2003 at 09:32:16AM -0400, wrote:
>>Looking at RAMpage's code and reading the description, I see nothing
>>that indicates it would solve this supposed "memory leak" problem.  All
>>that it does is allocate a huge chunk of memory and free it, forcing
>>any fragmented memory out onto disk.  I really don't see how that would
>>cause any improvement in anything related to a memory leak.
>I do not know much, but I do know this.  With RAMpage running 4 times a
>day, my server stays up.  Without it, it crashes.  The only thing
>running on the box other than the naked OS and backup software and
>Anti-virus, is cygwin and cygwin installed apps.  I agree with you to
>the extent that it certainly doesn't *solve* the *supposed* memory leak
>problem.  But it keeps my server up and I'm sure it will help others

If RAMpage is truly solving your problem then you should stop referring
to the problem as a memory leak.  AFAICT, although the web site is very
misleading, RAMpage does not do anything to fix "memory leaks".  It just
defragments memory.  It even says that it is not likely to fix problems
on newer OSes like Win2K.


Unsubscribe info:
Problem reports:

"WellChoice, Inc." made the following
 annotations on 08/08/2003 01:28:21 PM
Attention!  This electronic message contains information that may be legally
confidential and/or privileged.  The information is intended solely for the
individual or entity named above and access by anyone else is unauthorized.
If you are not the intended recipient, any disclosure, copying, distribution,
or use of the contents of this information is prohibited and may be unlawful.
If you have received this electronic transmission in error, please reply
immediately to the sender that you have received the message in error, and
delete it. Release/Disclosure Statement

Unsubscribe info:
Problem reports:

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]