NFS problem

[Alastair Neil]

On Mon, Mar 24, 2008 at 3:17 PM, Gerhard Magnus <magnus at agora.rdrop.com> wrote:
>
>  On Mon, 2008-03-24 at 13:49 -0500, Roger Heflin wrote:
>  > Gerhard Magnus wrote:
>  > > On Mon, 2008-03-24 at 12:55 -0500, Roger Heflin wrote:
>  > >> Gerhard Magnus wrote:
>  > >>> On Mon, 2008-03-24 at 10:38 -0600, Aaron Konstam wrote:
>  > >>>> No route to host sounds more like a connection problem. You can ssh between the machines?
>  > >>>> --
>  > >>> ssh works fine. I've been googling this problem and found that other
>  > >>> people have had it and it may be a serious bug. Could it be that NFS
>  > >>> doesn't work in fedora and that everybody uses samba anyway?
>  > >>>
>  > >> Nope, NFS works fine in FC8, both the default kernel, and a number of the
>  > >> upgrade kernels, no one in their right mind uses SAMBA within a group of
>  > >> Linux/Unix machines, Samba is typically only used when exporting Linux/Unix disk
>  > >> to machines that don't have NFS support.
>  > >>
>  > >> "no route to host" *IS* a connection problem, run the commands "netstat -r" and
>  > >> "ifconfig -a" on both the server and client machine, and return the output.
>  > >>
>  > >>                               Roger
>  > >>
>  > > OK... Here's the server:
>  > >
>  >
>  > >
>  > > ...and the client:
>  > >
>  In trying to fix this, I've since run "service iptables stop" on both
>  boxes to shut down the firewalls. Then I was able to mount the shared
>  directory on the client without problems. I started iptables again and
>  can still see the shared directory on the client.
>
>
>  > I don't see anything in either that stands out as wrong, both machines are
>  > plugged into the "LAN" ports of the router?
>  yes
>
> >
>  > Try "arp -a" on both the client and server, and see what returns.
>  Here's the server:
>  magnusg at PuteF Mon Mar 24 12:04:28
>  [270] ~ $ arp -a
>  PuteB.SMAssociates.com (192.168.1.13) at 00:13:20:8C:8D:D2 [ether] on
>  eth0
>  ...and the client:
>  magnusg at PuteB Mon Mar 24 12:11:05
>  [583] /mnt/PuteF $ arp -a
>  PuteF.SMAssociates.com (192.168.1.14) at 00:16:76:C2:73:01 [ether] on
>  eth0
>  ? (192.168.1.1) at 00:06:25:09:6C:22 [ether] on eth0
>
>
>  > If the arp command does not show the other machine, try pinging the other
>  > machine and rerun the "arp -a".
>  "arp -a" on the server shows the client, while "arp -a" on the client
>  shows the server.
>
>
>  > And can you both ping and ssh both directions?  client -> server and server ->
>  > client?
>  yes, both ping and ssh work fine. Is there some other port I need to
>  open to get this to work?
>
>
>
>
>
>  --
>  fedora-list mailing list
>  fedora-list at redhat.com
>  To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list
>

from the looks of it you are using NFSv3 exports, so enabling NFS4 on
the firewall tool buys you nothing.  Here is one solution, first it
tells the NFS daemons what ports to listen on, the the iptables rules
open up those ports:

[ajn at pensieve2 ~]$ sudo cat /etc/sysconfig/nfs
RPCNFSDCOUNT=25
STATD_PORT=10002
STATD_OUTGOING_PORT=10003
MOUNTD_PORT=10004
RQUOTAD_PORT=10005
[ajn at pensieve2 ~]$ sudo cat /etc/sysconfig/iptables
<snip>
-A RH-Firewall-1-INPUT -p tcp -m tcp --dport 111 -j ACCEPT
-A RH-Firewall-1-INPUT -p udp -m udp --dport 111 -j ACCEPT
-A RH-Firewall-1-INPUT -p tcp -m tcp --dport 2049 -j ACCEPT
-A RH-Firewall-1-INPUT -p udp -m udp --dport 2049 -j ACCEPT
-A RH-Firewall-1-INPUT -p tcp -m tcp --dport 10000 -j ACCEPT
-A RH-Firewall-1-INPUT -p udp -m udp --dport 10001 -j ACCEPT
-A RH-Firewall-1-INPUT -p tcp -m tcp --dport 10002:10005 -j ACCEPT
-A RH-Firewall-1-INPUT -p udp -m udp --dport 10002:10005 -j ACCEPT
<snip>

note: the 2049 ports are for nfs4