nofiles(descriptors) is set at 64 on my system, that's probably the
default. That may be the problem, but I'm not sure why I would have 64
file descriptors open. Each server instance is it's own process right?
One of my servers has 8 web apps and each web app reads two config files
(global and webapp) and the server would read web.xml for each webapp as
well as a couple of other config files, so I guess we're up to about 27
or so descriptors. What else would the server process be consuming file
descriptors for?

I'm leaning toward my box being memory starved, I've got 1GB in use and
1.5GB of swap space, eventually mostly in use. That doesn't mean much
though because of how Solaris 6 does memory management, the fact is
there isn't much swapping going on which shows I'm not out of memory,
but it isn't being reclaimed. But I did read an interesting article
about memory management deficiencies in 2.6. I haven't had this error
again, but I am still concerned about it's root cause.

Thanks for the input and I'll research more.

Lloyd

PS. I'll write mfield after I've tried 3.1. I want to give that one a
shot first. I think I'm the only person I've met that prefered the old
swing config tool. :) I guess remote management was too big a problem on
NT systems without X.



Scott Stirling wrote:
> 
> Can also be caused if you ran out of file descriptors for that process.  Not
> sure what the default limit is on Solaris, but if it happens again you can
> use ulimit to increase the number of open file descriptors allowed per
> process.
> 
> Scott Stirling
> JRun QA
> Macromedia
> 
> > -----Original Message-----
> > From: Mark Phelps [mailto:[EMAIL PROTECTED]]
> >
> > The error is an IO error that is occuring when you are
> > invoking a native
> > unix process.  The message says not enough space.  If I had
> > to make a guess
> > I would say that the server is trying to load the native process into
> > virtual memory, (memory on disk), and it is running out.  You
> > may be able to
> > configure your server to use more disk space for virtual memory.
> >
> > -----Original Message-----
> > From: Lloyd H. Meinholz [mailto:[EMAIL PROTECTED]]
> >
> >
> > I got an interesting stack trace yesterday and I'm at a loss exactly
> > what the cause is. I suspect I am running low on memory, but am not
> > sure. One of the actions that one of my Web applications
> > execute caused
> > this error. I restarted the JRun server that caused this
> > error and that
> > didn't seem to fix the problem, but when I restarted all JRun
> > instances,
> > the problem went away. I am running on Solaris 2.6 using j2se 1.3.0_02
> > and using JRun 3.0 sp1. I have since rebooted the entire box
> > and haven't
> > seen the problem since.
> >
> > Thanks,
> >
> >
> > Lloyd
> >
> >
> >
> > java.io.IOException: Not enough space
> >         at java.lang.UNIXProcess.forkAndExec(Native Method)
> >         at java.lang.UNIXProcess.<init>(UNIXProcess.java:54)
> >         at java.lang.Runtime.execInternal(Native Method)
> >         at java.lang.Runtime.exec(Runtime.java:551)
> >         at java.lang.Runtime.exec(Runtime.java:477)
> >         at
> > allaire.jrun.scripting.JavaCompilerService.outProcessCompile(J
> > avaCompilerSer
> > vice.java:177)
> >
> 
>
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Structure your ColdFusion code with Fusebox. Get the official book at 
http://www.fusionauthority.com/bkinfo.cfm
Archives: http://www.mail-archive.com/[email protected]/
Unsubscribe: http://www.houseoffusion.com/index.cfm?sidebar=lists

Reply via email to