[lkml]   [1996]   [Sep]   [10]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: NFS as a module / kerneld question
On Tue, 10 Sep 1996, Alan Cox wrote:

> > Since nfs.o forks 4 useless processes when loaded (nfsiod), it will never
> > be auto-removed. Mounting doesn't use any of those processes, a new one
> > is started so the use count is never lower than 4. I have figured out
> > a 'post-install' command for kerneld that kills these useless processes
> > and allows auto-unload. What kind of fate am I tempting using this
> > command?
> You'll totally break and ruin your NFS apart from that its fine. The 4
> processes are not useless. Kill them as a pre-removal thing if you need to

pre-removal doesn't work since the use count has to be 0 for kerneld to
trigger pre-removal commands and then remove the module. And if these
processes are needed, how come they just go to sleep and never wake up? And
when another mount command is executed, a new one is forked to handle the new
mount rather than using one of the existing processes? I can see nfs.o
forking one process since it gets loaded due to a mount command and that
one nfsiod is necessary to handle that mount. But what about the other 3?
And when every nfs mount is unmounted, the original 4 don't terminate. They
just hang around. As a result the use count of nfs.o is always some number
that is equal or greater than 4.

As pointed out in a private response from another reader, 'is memory that
tight?'. Well it isn't on most of my machines but I don't like stuff
hanging around when it isn't needed. That is the main reason that I
started using modules and kerneld.

Also, I blanketed the original message to 3 mailing lists. Whis one is
the correct one (or are all 3 incorrect) so that this discussion can
just be posted to that list?


 \ /
  Last update: 2005-03-22 13:37    [W:0.069 / U:0.564 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site