NFS - RPC Port mapper failure - RPC: unable to send

NFS - RPC Port mapper failure - RPC: unable to send

Post by ralnw » Wed, 04 Dec 2002 08:55:40



Hello,
  I have 3 linux computers. A redhat 72 server, redhat 72 workstation
and a redhat 73 notebook. I setup NFS on the server using the HOWTO from
ldp.org. Everthing went fairly smooth. Then I setup the client on the
redhat 72 workstation and have it working fine. Now the notebook is the
problem. Only on bootload I get the error message in the subject line of
my post. Here is the syslog edited:
[snip]

Quote:> Dec  1 23:50:35 dfi cardmgr[819]: watching 2 sockets
> Dec  1 23:50:35 dfi kernel: cs: IO port probe 0x0c00-0x0cff: clean.
> Dec  1 23:50:35 dfi cardmgr[819]: initializing socket 1
> Dec  1 23:50:35 dfi kernel: cs: IO port probe 0x0100-0x04ff: excluding 0x4d0-0x4d7
> Dec  1 23:50:35 dfi cardmgr[819]: socket 1: Xircom IIps Ethernet
> Dec  1 23:50:35 dfi kernel: cs: IO port probe 0x0a00-0x0aff: clean.
> Dec  1 23:50:35 dfi kernel: cs: memory probe 0xa0000000-0xa0ffffff: clean.
> Dec  1 23:50:35 dfi mount: mount: RPC: Port mapper failure - RPC: Unable to send
> Dec  1 23:50:35 dfi netfs: Mounting NFS filesystems:  failed
> Dec  1 23:50:35 dfi cardmgr[819]: executing: 'modprobe xirc2ps_cs'
> Dec  1 23:50:36 dfi kernel: xirc2ps_cs.c 1.31 1998/12/09 19:32:55 (dd9jn+kvh)
> Dec  1 23:50:36 dfi kernel: eth0: Xircom: port 0x300, irq 9, hwaddr 00:80:C7:26:CD:81
> Dec  1 23:50:36 dfi cardmgr[819]: executing: './network start eth0'
> Dec  1 23:50:36 dfi /etc/hotplug/net.agent: invoke ifup eth0
> Dec  1 23:50:36 dfi netfs: Mounting other filesystems:  succeeded

[snip]

note the ./network start after failure. If I look at the portmap script
I think it exits if  networking is not up..

[snip]

# Check that networking is up.
if [ "$NETWORKING" = "no" ]
then
        exit 0
fi

[ -f /sbin/portmap ] || exit 0

[snip]

I can mount the share from the command line after bootload no problem.
I think this indicates the exports file on the server is good and
firewall rules (local and server) plus the hosts.allow file are okay
The fstab is exactly the same as on the good workstation
the rpcinfo -p <servername> shows all the services okay
Here is the fstab entry from the notebook:
 > gateway:/home/share       /mnt/share              nfs     rw,hard,intr    0 0
When I traced the traffic with ethereal it looked like dhcp was
assigning the ip address of the notebook after the port mapper failure
so I set a static address - no help -
I also tried a faster pcmcia 10/100 card again - no change -
I googled for solutions and did not find any.
I can get it to work on bootload with a mount command in the rc.local file
Below is the /etc/mtab with the share mounted:

Quote:> gateway:/home/share /mnt/share nfs rw,addr=192.168.100.1 0 0

This the /var/lib/nfs/rmtab from the server: (x.x.x.04 is the notebook)

Quote:> 192.168.100.9:/home/share:0x00000046
> 192.168.100.4:/home/share:0x00000001

Here is the /var/lib/nfs/etab from the server:
Quote:> /home/share        192.168.100.0/255.255.255.0(rw,async,wdelay,hide,secure,

 > > root_squash,no_all_squash,subtree_check,secure_locks,mapping=identity,anonuid=-2,anongid=-2)

And the /var/lib/nfs/xtab from the server:

Quote:> /home/share        192.168.100.4(rw,async,wdelay,hide,secure,root_squash,no_all_squash,
> subtree_check,secure_locks,mapping=identity,anonuid=-2,anongid=-2)
> /home/share        192.168.100.9(rw,async,wdelay,hide,secure,root_squash,no_all_squash,
> subtree_check,secure_locks,mapping=identity,anonuid=-2,anongid=-2)

Any help to get this working with the fstab mount is most welcomed.

The notebook OS version is:

version 2.96 20000731 (Red Hat Linux 7.3 2.96-112)) #1 Wed Nov 13
20:29:30 EST 2002

Best regards,

 
 
 

NFS - RPC Port mapper failure - RPC: unable to send

Post by CrayzeeWul » Wed, 04 Dec 2002 13:41:01



> Hello,
>   I have 3 linux computers. A redhat 72 server, redhat 72 workstation
> and a redhat 73 notebook. I setup NFS on the server using the HOWTO from
> ldp.org. Everthing went fairly smooth. Then I setup the client on the
> redhat 72 workstation and have it working fine. Now the notebook is the
> problem. Only on bootload I get the error message in the subject line of
> my post. Here is the syslog edited:
> [snip]
>> Dec  1 23:50:35 dfi cardmgr[819]: watching 2 sockets
>> Dec  1 23:50:35 dfi kernel: cs: IO port probe 0x0c00-0x0cff: clean.
>> Dec  1 23:50:35 dfi cardmgr[819]: initializing socket 1
>> Dec  1 23:50:35 dfi kernel: cs: IO port probe 0x0100-0x04ff: excluding 0x4d0-0x4d7
>> Dec  1 23:50:35 dfi cardmgr[819]: socket 1: Xircom IIps Ethernet
>> Dec  1 23:50:35 dfi kernel: cs: IO port probe 0x0a00-0x0aff: clean.
>> Dec  1 23:50:35 dfi kernel: cs: memory probe 0xa0000000-0xa0ffffff: clean.
>> Dec  1 23:50:35 dfi mount: mount: RPC: Port mapper failure - RPC: Unable to send
>> Dec  1 23:50:35 dfi netfs: Mounting NFS filesystems:  failed
>> Dec  1 23:50:35 dfi cardmgr[819]: executing: 'modprobe xirc2ps_cs'
>> Dec  1 23:50:36 dfi kernel: xirc2ps_cs.c 1.31 1998/12/09 19:32:55 (dd9jn+kvh)
>> Dec  1 23:50:36 dfi kernel: eth0: Xircom: port 0x300, irq 9, hwaddr 00:80:C7:26:CD:81
>> Dec  1 23:50:36 dfi cardmgr[819]: executing: './network start eth0'
>> Dec  1 23:50:36 dfi /etc/hotplug/net.agent: invoke ifup eth0
>> Dec  1 23:50:36 dfi netfs: Mounting other filesystems:  succeeded

Hi ralnw5,

Here is what I think is happening during the startup (assuming your default
init level is 5, but the answer is the same even if it is 3):

(a) Sometime during the startup, /etc/rc.d/rc5.d/S24pcmcia is executed.
This script starts the PCMCIA services. In particular, it starts "cardmgr"
which takes care of setting up your PCMCIA devices.

(b) While "cardmgr" is trying to figure its way through the system and
detect PCMCIA devices (in background), /etc/rc.d/rc5.d/S25netfs is
executed. This script is responsible for mounting (among other things) NFS
filesystems. In your case, the S25netfs attempts to mount NFS filesystes
_before_ cardmgr has had a chance to configure your network card.
Obviously, it fails. Eventually, cardmgr sets up your PCMCIA ethernet card
and everything works fine afterwards.

The simplest solution to your problem is to put the following at the end
of /etc/rc.d/rc.local:

/bin/mount -a -t nfs

This will provide cardmgr enough time to configure your PCMCIA ethernet
card and will attempt to mount all the NFS filesystems listed in your
/etc/fstab file at the very end of the startup procedure.

Enjoy,
--
CrayzeeWulf

 
 
 

NFS - RPC Port mapper failure - RPC: unable to send

Post by ralnw » Thu, 05 Dec 2002 02:39:54




> The simplest solution to your problem is to put the following at the end
> of /etc/rc.d/rc.local:

> /bin/mount -a -t nfs

> This will provide cardmgr enough time to configure your PCMCIA ethernet
> card and will attempt to mount all the NFS filesystems listed in your
> /etc/fstab file at the very end of the startup procedure.

Thanks,

  your solution works fine. Now I have the correct parameters on the
remote mount.

 
 
 

1. RPC: Port mapper failure - RPC: Unable to send

Hello, I'm having some NFS problems which result in the error in the
subject line.

I'm running Linux on a P-90 with a 3Com 3c509 ethernet card.
I have been trying to mount a disk "-t nfs" over ethernet at work.
My mount command shows the following:


/dev/hda2 on / type ext2 (rw)
/dev/hda1 on /dosc type msdos (rw)
none on /proc type proc (rw)

The disk I want to mount is called "/export/home" on a machine called
"vader."  Sometimes I am only able to mount it for A SINGLE ACCESS.
Afterwards, the mount remains, but I am no longer able to access it.



dave    fred    george
stuff   in      home

        That was how /home was supposed to look.
        Then, some time later...



        /home has disappeared!


/dev/hda2 on / type ext2 (rw)
/dev/hda1 on /dosc type msdos (rw)
none on /proc type proc (rw)
vader:/export/home on /home type nfs (rw,addr=192.83.225.35)

        Hmmm... NFS thinks it's still there!



can't create lock file /etc/mtab~: File exists

        oops.



mount clntudp_create: RPC: Port mapper failure - RPC: Unable to send

No matter what I do, until I boot linux again I get this same RPC
error.  I have tried killing rpc.nfsd and re-statring it, and this
SOMETIMES works.  I have tried dmesg, and this is what I see


NFS: notice message: result=-11
nfs_statfs: statfs error = 11
NFS: notice message: result=-11
nfs_statfs: statfs error = 11
NFS: notice message: result=-11
nfs_statfs: statfs error = 11
NFS: notice message: result=-11
nfs_statfs: statfs error = 11
etc. etc. etc.

How can I get NFS to mount robustly?  Why should I be able to  access
the mount once or twice before I start getting errors?

Thanks for any advice you can give me,
Dave

2. Can't mount ISO9660 CD-ROM in OSE 5.0

3. NFS: RPC: Port mapper failure - RPC: Unable to receive

4. Synaptic Touchpad Taps Giving Enter Keys

5. RPC: Port mapper failure - RPC: Unable to receive

6. Making Mozilla the default browser problem

7. clnttcp_create: RPC: Port mapper failure - RPC: Unable to receive

8. ESS ES688 (Shark Multimedia MAKO) sound card

9. clnttcp_create: RPC: Port mapper failure - RPC: Unable to connect

10. mount clntudp_create: RPC: Port mapper failure - RPC: Unable to receive

11. clnttcp_create: RPC: Port mapper failure - RPC: Unable to receive

12. mount clntudp_create: RPC: Port mapper failure - RPC: Authentication error

13. RPC: Port mapper error RPC: Unable to recieve