Re: [CFT] Autofs.

On 0804T1451, Rick Macklem wrote:
> Bjoern A. Zeeb wrote:
> > On 30 Jul 2014, at 07:19 , Edward Tomasz Napierała
> > <trasz_at_freebsd.org> wrote:
> > 
> > > At the link below you will find a patch that adds the new
> > > automounter.
> > > The patch is against yesterdays 11.0-CURRENT.
> > > 
> > > http://people.freebsd.org/~trasz/autofs-head-20140729.diff
> > 
> > I also just submitted
> > https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=192379 to allow -o
> > vers= mount_nfs compatibility, which makes it easier to integrate
> > with Linux/OSX/Solaris LDAP setups and mount options from LDAP.
> > 
> > 
> > > Testing is welcome.  Please start with manual pages, eg.
> > > automount(8).
> > 
> > 
> > I found one case now doing the aforementioned where when the initial
> > mount_nfs fails (e.g., for invalid options), then a later mount did
> > not succeed either, with the correct mount options;  I did try to
> > run automount -u in between tries, as well as service automountd
> > restart, but that did not make a change;  given I was short on time,
> > a reboot of my desktop made this go away.   Is there some “negative
> > caching” in the kernel module possibly that will not retry the mount
> > for another time or something—as in if I were more patient and
> > waited 5 minutes, would it maybe just have worked again?
> > 
> If I recall it correctly, it retries a mount attempt that fails after
> something like 1min (which seems like a long time when you are waiting
> for it;-). I think that was done so that clients don't flood a rebooting
> server with mount attempts, but that code was written long, long ago.

>From what I've seen mount_nfs(8) retries indefinitely; that's why
automountd disables that behaviour by default, by passing "retrycnt=1".