Hi,
I run a java based chat, running a java server, using wrapper from
http://wrapper.tanukisoftware.org, and a java applet as a client. The
server runs on a Debian based system on port 23.
The java server runs 2 threads for each new connection, one as a main
thread who receives messages and send
Hi,
I run a java based chat, running a java server, using wrapper from
http://wrapper.tanukisoftware.org, and a java applet as a client. The
server runs on a Debian based system on port 23.
The java server runs 2 threads for each new connection, one as a main
thread who receives messages and send
We are using Redhat Advanced Server 2.1
kernel 2.4.9-e.3smp
java version 1.3.1_07
Bea Weblogic 6.2 sp4
4 processors
3GB RAM
The problem we are running into is that when we invoke weblogic with
-mx1536m jvm option flag we hit a native thread limit of 199 threads.
If we invoke the JVM with no
Joseph Shraibman wrote:
Juergen Kreileder wrote:
[EMAIL PROTECTED] writes:
On Fri, 14 Feb 2003, Juergen Kreileder wrote:
BTW: If you use another setting than CONFIG_1GB
or if you're on x86-64
you want to use the new release. Older HotSpot versions have
problems with the other values.
So
Juergen Kreileder wrote:
[EMAIL PROTECTED] writes:
On Fri, 14 Feb 2003, Juergen Kreileder wrote:
BTW: If you use another setting than CONFIG_1GB
or if you're on x86-64
you want to use the new release. Older HotSpot versions have
problems with the other values.
So which ones are good? Is t
On Sun, 16 Feb 2003, Dan Kegel wrote:
> [EMAIL PROTECTED] wrote:
> > But they have no logical reason to share stack memory. I was assuming
> > the os did things logically. My mistake.
>
> Just curious: do you know of any operating systems that don't
> share the stack memory for different thread
[EMAIL PROTECTED] wrote:
On Fri, 14 Feb 2003, Dan Kegel wrote:
So I just have to set that line in /boot/config-2.4.18-14 and reboot?
No; you have to install your kernel source,
read the file /usr/src/linux-2.4/README,
copy that config file to /usr/src/linux-2.4/.config,
edit it not manually b
On Fri, 14 Feb 2003, Dan Kegel wrote:
> >
> > I was under the impression that since linux made ever thread a seperate
> > process they each had their own stack and only shared the heap memory.
>
> Threads share the entire memory space. That's what makes them threads
> instead of processes.
But
On Fri, Feb 14, 2003 at 04:22:31PM -0800, John Neffenger wrote:
> Further note to readers: I got past this 16,000 native thread limit by
> specifying the "-green" user-level thread option to the Blackdown
> virtual machine. So my 20,000 threads were all mapped onto one Li
Hi Dan,
Note to readers: he's using a smaller thread stack than the
previous poster (64k vs. 100k). At 1GB address space limit
(or 2^30 bytes), that's an upper limit of 2^(30-16 = 14) = 16000
or so threads.
Further note to readers: I got past this 16,000 native thread limit by
[EMAIL PROTECTED] writes:
> On Fri, 14 Feb 2003, Juergen Kreileder wrote:
>>
>> BTW: If you use another setting than CONFIG_1GB
or if you're on x86-64
>> you want to use the new release. Older HotSpot versions have
>> problems with the other values.
>
> So which ones are good? Is the latest
Hi Joseph,
(See also http://www.kegel.com/c10k.html )
That points to http://www.volano.com/linux.html which doesn't exist
anymore. The wayback machine has a verion from Oct 31 2001 that doesn't
say how to recompile the kernel to increase the limit. It does point me
to /etc/security/limits.c
On Fri, Feb 14, 2003 at 03:38:49PM -0500, [EMAIL PROTECTED] wrote:
> > $ grep CONFIG_.GB /boot/config-2.4.18-14
> > # CONFIG_3GB is not set
> > # CONFIG_2GB is not set
> > CONFIG_1GB=y
>
> So I just have to set that line in /boot/config-2.4.18-14 and reboot? Is
> that limit per process or system
On Thu, 13 Feb 2003, Dan Kegel wrote:
> Man Chi Ly wrote:
> >>>Physical memory, maybe, but I bet you a nickel you got
> >>>up near 1GB of *virtual* memory. Go learn about how
> >>>thread stacks work. You're not running out of RAM; you're
> >>>running out of address space.
> >>
> >>I guess I don'
On Fri, 14 Feb 2003, Juergen Kreileder wrote:
>
> BTW: If you use another setting than CONFIG_1GB you want to use the
> new release. Older HotSpot versions have problems with the other
> values.
>
So which ones are good? Is the latest sun jdk good or only the
forthcoming blackdown?
Man Chi Ly <[EMAIL PROTECTED]> writes:
> On Thu, 13 Feb 2003, Joseph Shraibman wrote:
>
>> Dan Kegel wrote:
>> > Joseph Shraibman wrote:
>>
>> The one that comes with redhat 8.0 (updated), which is
>> kernel-smp-2.4.18-19.8.0
>>
>> > CONFIG_1GB
>> > CONFIG_2GB
>> > CONFIG_3GB
>> > are set in your
On Thu, 13 Feb 2003, Man Chi Ly wrote:
>
> Dan's probably one of the authorities on this performance issue; but isn't
> it generally considered bad form to be spawning so many threads in an
> application?
Nature of the application. I need to send out a lot of email in parallel.
>
On Thu, 13 Feb 2003, Joseph Shraibman wrote:
> Dan Kegel wrote:
> > Joseph Shraibman wrote:
> >
> >> I never got anywhere near 1gb of memory usage.
> >
> >
> > Physical memory, maybe, but I bet you a nickel you got
> > up near 1GB of *virtual* memory. Go learn about how
> > thread stacks work. Y
Dan Kegel wrote:
Joseph Shraibman wrote:
I never got anywhere near 1gb of memory usage.
Physical memory, maybe, but I bet you a nickel you got
up near 1GB of *virtual* memory. Go learn about how
thread stacks work. You're not running out of RAM; you're
running out of address space.
I gues
Dan Kegel wrote:
The -Xss100k helps because each thread stack immediately occupies
its full range of address space (discontiguous stacks not being supported).
If the OS defaults to a limit of 1GB address space for user processes,
at 100KB per stack, that's a hard limit of 10,000 threads.
I nev
Hi Joseph,
Here's a good explanation about the space Java reserves for each thread
stack. It's written about Windows, but applies to Solaris, Linux and
FreeBSD as well:
Connection scaling in Java
http://www.javaworld.com/javaworld/jw-03-1999/jw-03-volanomark_p.html#sidebar
It's just a matter
On Thu, 13 Feb 2003, John Rousseau wrote:
> The number of threads is limited by kernel config (as Nathan pointed
> out)
I'm not running into that limit. That is something over 14k.
and by available memory. By decreasing the stack size used by each
> thread (via -Xss), you made each thread use
. I've discovered that I get get somewhere between 3600
> and 3700 threads in a jvm without any arguments. If I pass in -Xss100k I
> can get that up to 8190 threads, but I'm not sure why. How can I maximize
> the number of threads per jvm?
Have you checked your thread li
I'm running:
java version "1.4.1_01"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.4.1_01-b01)
Java HotSpot(TM) Client VM (build 1.4.1_01-b01, mixed mode)
... on redhat 8.0. I've discovered that I get get somewhere between 3600 and 3700 threads
in a jvm without any arguments. If I p
24 matches
Mail list logo
