[SATLUG] NFS Mounts

Daniel J. Givens daniel at rugmonster.org
Fri Dec 21 15:01:13 CST 2007


Richard Maynard wrote:
> There are times where a soft mount is appropriate, it's not always going to
> cause corruption and woes for your systems. It just needs to be in the right
> place, and for the right reasons.

Thanks for the excellent explanation. It seemed that things worked like
a charm as long as the files weren't large (+500MiB).

Could the intr option serve as a workaround for the scenarios you gave
when using hard mounts? I think I know the answer (no), but would like
to confirm.

> What protocol do you use for NFS? Hard mounts in Linux using UDP have caused
> our engineering groups some grief because they seem to be prone to blocking.
> Using TCP for our mounts in many of these instances has allowed us to work
> around whatever the underlying problem was. 

I do use UDP. Here's an example of one of my exports.

/srv/net/media
10.20.0.0/16(rw,no_root_squash,no_subtree_check,async)

And here is the complementing automount (translated from ldap).

# under /net/storage/
media -rw,hard,intr,udp,wsize=32768,rsize=32768
nfs.home.rugmonster.org:/srv/net/media

The server is running CentOS 5.1 and the client I'm using is Ubuntu
7.10. I wonder if the switch between the two might be flaky or possibly
one of the NICs.

> While, not a fix for the issue,
> it can be a workaround until you can find out why UDP traffic is being lost
> on your network.

Any suggestions on how I might track down the issue? Considering this is
my home network, I don't think pouring over tcpdump or wireshark
captures is going to be worth my time except for the experience I could
possibly use down the road.

Thanks!

Daniel


More information about the SATLUG mailing list