Richard A Downing wrote:
David Fix wrote:
You know, I know this is late for a comment... But I just thought of
something... Could it be as simple as doing something like this:
kill `pgrep nfsd`
That's just an inelegant equivalent of:
rpc.nfsd -- 0
I think Tor Olav is trying to confirm that the kernel threads have
stopped before he shuts down. Hence his script.
That's spot on.. Probably should add some comments on that script. I'm
trying to do the same thing as the original script does, but without the
delay, and with correct result output.
As I said in my first post, I've experienced this on all my BLFS builds.
First I thought I had some bad configuration, or maybe permissions
problem because I'm using package users. But several builds later, the
issue persist. Got a feeling either not much people use NFS anymore, or
noone really care too much about it..
The general flow of the script would be:
1. If process is running, then stop it, else issue warn
2. If still running, then kill it, else issue ok
3. If still running, then issue fail
I'll use this script for awhile, and see how it performs. Especially is
a process is active or locked, to see what happens. There shouldn't be
any data-loss or corruption caused by the script. I'm wondering if
that's the actual reason for the delay in the original blfs-script.
Again, all comments or better solutions appreciated. It's never too late :)
--Tor Olav
--
http://linuxfromscratch.org/mailman/listinfo/blfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page
