RPC Error

[Guna_2]

Hi,

Last night i was installing a software on my HP box(hostA) from a depot server. But it returned me the rpc error. and it was not able to connect to the depot.

Then i tried

HostA#rpcinfo -p

rpcinfo: can't contact portmapper: RPC: Rpcbind failure - RPC: Failed (unspecified error)

But if i try rpcinfo from my depot server to this client, i am able to get the registerd details. And my depot server is okay as every other clients are able to connect to the depot.

then i treid the following on my client
HostA#/sbin/init.d/swagend stop
HostA#/sbin/init.d/nfs.client stop

HostA#/sbin/init.d/nfs.client start
HostA#/sbin/init.d/swagend start

Still the same problem.

And even i am unable to mount a nfs share directory from this client.

Can any please guide me?

Thanks & Rgds
---MGP---

[Ivan Krastev]

Try also:

/sbin/init.d/nfs.core start

Can you post full error message?

regards,
ivan

[Guna_2]

Hi ivan

I tried to restart the nfs.core also

stop swagentd
stop nfs.client
stop nfs.core

start nfs.core
start nfs.client
start swagentd

It did not gave me any error. It just started the rpcbind program.

while i try to mount a nfs mount also i am reveing rpc bind time out error.

Thanks
---MGP---

[Guna_2]

Hi,

I am attaching the error log too


======= 07/27/07 02:46:10 MDT BEGIN swinstall SESSION
(non-interactive) (jobid=rich-0083)

* Session started for user "root@test".

* Beginning Selection
* Target connection succeeded for "test:/".
ERROR: RPC exception: "Communications failure (dce / rpc)" 07/27/07
02:48:03 MDT
ERROR: A Remote Procedure Call to a daemon has failed. Could not
start a management session for
"depotserver:/depot". Make sure the
host is accessible from the network, and that its daemon,
swagentd, is running.
* Source connection failed for
"depotserver:/depot".
WARNING: More information may be found in the daemon logfile on this
target (default location is depotserver:/var/adm/sw/swagentd.log).
* Selection had errors.

[Dave Olker]

Hi,

Is there any messages in the syslog.log file related to rpc? What name resolution mechanism are you using for RPC and SERVICES? Files? NIS?

Dave

[skt_skt]

Also check if that server is in a DMZ zone.

We do have dmz servers where they wont be able to communicate to source depot.WE create the local depots from them

[Guna_2]

Hi,

I am using DNS. I am able to resolve the ip address, and even able to ping with the depot server.


---MGP---

[Dave Olker]

Is there any messages in the syslog.log file related to rpc? What name resolution mechanism are you using for RPC and SERVICES? Files? NIS?

[Guna_2]

Hi Dave,

I am using files name resolution for rpc and services. No nis.....

The syslog messages

Jul 27 03:34:36 rich rpcbind: check_netconfig: Found CLTS loopback transport
Jul 27 03:34:36 rich rpcbind: check_netconfig: Found COTS loopback transport
Jul 27 03:34:36 rich rpcbind: check_netconfig: Found COTS ORD loopback transport
Jul 27 03:34:36 rich rpcbind: init_transport: udp: cannot bind: address in use
Jul 27 03:43:04 rich inetd[15879]: rpc.cmsd/udp: Connection from rich (10.3.220.22) at Fri Jul 27 03:43:04 2007
Jul 27 04:07:20 rich rpcbind: terminate: rpcbind terminating on signal. Restart with "rpcbind -w"
Jul 27 04:07:27 rich rpcbind: check_netconfig: Found CLTS loopback transport
Jul 27 04:07:27 rich rpcbind: check_netconfig: Found COTS loopback transport
Jul 27 04:07:27 rich rpcbind: check_netconfig: Found COTS ORD loopback transport
Jul 27 04:07:27 rich rpcbind: init_transport: check binding for udp
Jul 27 04:07:27 rich rpcbind: init_transport: check binding for tcp
Jul 27 04:07:27 rich rpcbind: init_transport: check binding for ticlts
Jul 27 04:07:27 rich rpcbind: init_transport: check binding for ticotsord
Jul 27 04:07:27 rich rpcbind: init_transport: check binding for ticots



#cat /etc/services | grep -i rpc
portmap 111/tcp sunrpc # SUN Remote Procedure Call
portmap 111/udp sunrpc #
courier 530/tcp rpc #


Thanks
Guna

[Safarali]

Hi
try to mount on IP basis
mount 120.x.x.x:/depot /hostA
if you can then it is the problem for host resolution else
do the following
got to depot server
do
#exportfs -a
check depot file has permission to mount all server if
then do
nfs.core stop
nfs.server stop
nfs.client stop and restart

if not please put your full error message

Regards
Safar

[Guna_2]

Hi

I tried to mount a nfs share via ip address also.

# mount 10.0.1.2:/depot /depot
nfs mount: get_fh: 10.0.1.2:: RPC: Rpcbind failure - RPC: Timed out
nfs mount: get_fh: 10.0.1.2:: RPC: Rpcbind failure - RPC: Timed out
nfs mount: retry: retrying(1) for: /depot after 5 seconds
nfs mount: retry: giving up on: /depot

The server setup is o.k. Becuase i am able to mount the same share from another client.

---MGP---

[Dave Olker]

Hi Guna,

What OS is the client system running? What NFS patch level is the client running? If you are on the client system and you type:

# rpcinfo -p localhost

do you get the same errors?

Regards,

Dave

[Guna_2]

Hi,

From client
rpcinfo -p client # Is working
rpcinfo -p depotserver # Error.
rpcinfo: can't contact portmapper: RPC: Rpcbind failure - RPC: Failed (unspecified error)


and From Server
rpcinfo -p client # Is working. and i am able to get all the details also.
rpcinfo -p depotserver # Is also working

This what i don't understand, the client rpcbind is registering all the rpc services and listening also.

Bye
Guna

[Ninad_1]

Hi,

I hope you have the entry
127.0.0.1 localhost loopback
in your /etc/hosts file.

Regards,
Ninad