--- In "j.chitte" <> wrote:
> Thanks for your comments.
>
> I found the problem in fstab on the board. I had a nfs mount in the other
> direction that was getting pulled in by 'mount -a' during boot-up.
>
> The mount was not showing up in mount output that's why I missed it but it
> seems it caused this problem before failing.
>
> so humble apologies to the nfs team and much relief to me , I was really
> starting to doubt how reliable nfs was.
>
> My control loop is now doing what it should.
>
> /js
>
Oh no. I spoke too soon. I retract my apology , it sucks.
nfs seems unable to even keep changes made on the client visisble to the client
in any reproducible way.
Mostly , changes done on the server don't get seen on the client. Presumably
some kind of caching as Jason said.
Changes made on the client may or may not still be there 10s later. Some stick,
some get sort of merged with an earlier copy and loads of ^@ get appended to
the file , some other changes just get ignored if I reopen the file with vi 10
seconds after a save and quit.
It's damn near impossible to debug anything with this sort of shit going on.
I may try building nfs4 into the kernel and see if it's more reliable.
This is not a viable dev. environment.
------------------------------------
Yahoo! Groups Links
<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/ts-7000/
<*> Your email settings:
Individual Email | Traditional
<*> To change settings online go to:
http://groups.yahoo.com/group/ts-7000/join
(Yahoo! ID required)
<*> To change settings via email:
<*> To unsubscribe from this group, send an email to:
<*> Your use of Yahoo! Groups is subject to:
http://docs.yahoo.com/info/terms/
|