lkml.org 
[lkml]   [1999]   [Feb]   [6]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
SubjectRe: Linux-2.2.2-pre2..
Date
Hello!

> > Then I run memory hog and expect some of unused inodes to be shrinked, to
> > free some memory, but alas!
> > mordor:~# cat /proc/sys/fs/inode-state
> > 2198 50 1 0 0 0 0
> > Number of inodes is not decreased.
> > Or do I misunderstood these numbers somehow?
> The inode numbers can easily grow past the "maximum", but once it reaches
> the maximum the growth should be stunted and controlled. The above is
> exactly the kind of behaviour you should expect: inodes freely grow until
> they hit the max number, and then the growth should slow down quite
> noticeably. Think of "max" as a soft limit rather than a hard one.
Yes. I can understand that. But why number of inodes only grows up?!
Ain't when we need some memory we must shrink inode pool and give freed memory
to those who need it?
Here is the example: Let's imagine I ran inode hog and now I have
a lot of inodes in cache. After some time I run memory hog(s), and a lot
of inodes in cache is now unused. One can expect that memory form that pool
can be returned back to be used by programs and another part of kernel,
is not it? But I got this:
mordor:~# cat /proc/sys/fs/inode-state
2058 1885 0 0 0 0 0
1885 inodes in freeable pool, but not freed. Is there reason for it?
And when we need more inodes, this number grows. For now it grows slowly,
but memory got eaten anyway => "NT syndrome", not as fast as with 2.2.[01],
but here it is.


> The thing is, that inodes are so critical that having a hard limit for
> them can result in serious problems (and the machine dies completely). In
> 2.0.x, for example, when this happened the machine usually died and
> printed out a kernel message like "No more inodes, contact Linus", and too
> many people have...
For now we have quite opposite situation "Too many inodes, contact Linux" :-\

Bye,
Oleg

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-03-22 13:50    [W:0.088 / U:0.168 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site