How to start the NFS server on HP-UX 11.31 at start up (boot up) automatically

[senthil_kumar_2]

Hi All,

I want to start the NFS server on HP-UX 11.31 at start up (boot up) automatically.

For that i have done following configuration in /etc/rc.config.d/nfsconf
---------------------------------------
# vi /etc/rc.config.d/nfsconf

NFS_CORE=1

LOCKMGR=1

NFS_SERVER=1

START_MOUNTD=1

-------------------------------------------

But NFS server is not starting automatically.

where as i am able to start manually and access the NFS share.

I used following commands to start NFS manually

# /sbin/init.d/nfs.core start

# /sbin/init.d/lockmgr start

# /sbin/init.d/nfs.server start


How to solve this issue

[Patrick Wallek · Accepted solution]

That should start it automatically when you boot the system.

What does '/etc/rc.log' show? Any error messages for the NFS start processes?

[Vishu]

do you have Startup scripts in the /sbin/rc2.d directory.

if yes, check /etc/rc.log to see what is the error causing it to fail.

[senthil_kumar_2]

Hi All,

Please find the output from /etc/rc.log

# more /etc/rc.log | grep -i nfs
Start NFS core subsystem
Output from "/sbin/rc2.d/S400nfs.core start":
Starting NFS CORE networking
Start NFS IPv6 subsystem
Output from "/sbin/rc2.d/S401nfs.core-ipv6 start":
Start NFS IPv6 subsystem
Output from "/sbin/rc2.d/S402nfs.core2-ipv6 start":
Start NFS client subsystem
Output from "/sbin/rc2.d/S430nfs.client start":
Starting NFS CLIENT subsystem
"/sbin/rc2.d/S430nfs.client start" FAILED
Start NFS server subsystem
Output from "/sbin/rc3.d/S100nfs.server start":
"/sbin/rc3.d/S100nfs.server start" FAILED


# more /etc/rc.log | grep -i lock
Start lock manager subsystem
Output from "/sbin/rc2.d/S425lockmgr start":
Starting up the lock manager daemon
/usr/sbin/rpc.lockd
ERROR: rpc.lockd not running. Run "/sbin/init.d/lockmgr start" to start rpc.lockd, exiting
Start clock daemon
ERROR: rpc.lockd not running. Run "/sbin/init.d/lockmgr start" to start rpc.lockd, exiting



So as per above log, the "NFS server" and "lockmgr" are not started.

How to solve this issue.

[Patrick Wallek]

We need to see more of the rc.log file. See if there are any errors in that section of the file. Just grep'ing for NFS may not show you the errors it encountered.

[Vishu]

Hi senthil,

you grepped the nfs word, it will not show the error it gave while failed. So, can you attach the /etc/rc.log with your post or paste the error from the file here to further look into.

[James R. Ferguson]

Hi Senthil:

By any chance did you make a backup of '/etc/rc.config.d/nfsconf' when you changed its settings? If so, did you retain that backup copy in '/etc/rc.config.d'? There are ways to do this and ways to _not_ do this.

Regards!

...JRF...

[senthil_kumar_2]

Please find logs related to NFS services from the output of /etc/rc.log:


Start NFS core subsystem
Output from "/sbin/rc2.d/S400nfs.core start":
----------------------------
Starting NFS CORE networking

Starting up the rpcbind
/usr/sbin/rpcbind



Start lock manager subsystem
Output from "/sbin/rc2.d/S425lockmgr start":
----------------------------
Starting up the Status Monitor daemon
/usr/sbin/rpc.statd
Starting up the lock manager daemon
/usr/sbin/rpc.lockd


Start NFS client subsystem
Output from "/sbin/rc2.d/S430nfs.client start":
----------------------------
Starting NFS CLIENT subsystem

ERROR: rpc.lockd not running. Run "/sbin/init.d/lockmgr start" to start rpc.lockd, exiting
"/sbin/rc2.d/S430nfs.client start" FAILED



Start NFS server subsystem
Output from "/sbin/rc3.d/S100nfs.server start":
----------------------------
ERROR: rpc.lockd not running. Run "/sbin/init.d/lockmgr start" to start rpc.lockd, exiting
"/sbin/rc3.d/S100nfs.server start" FAILED



Is it enough?

[Pete Randall]

The NFS failure is because of the lockmgr not being started. Is there any more detail concerning lock mgr in rc.log? If not, we need to take a look at /etc/rc.config.d/lockmgr to see if there is something obviously wrong there.


Pete

[Steven Schweda]

> For that i have done following [...]

Using SAM / SMH to enable NFS might get
everything done right automatically. Not
using the easy-to-use tools may make more
sense if you know what you're doing, but if
you're asking here about what went wrong,
then I'd vote for using the easy-to-use
tools.

After it works, you can go back in and play
with exporting/sharing particular file
systems in particular ways.

[senthil_kumar_2]

Hi Steven,

I tried following:

smh --> Networking and Communications --> Network Services Configuration ---> Networked File Systems --> Automounted Remote File Systems

but it gives the below error:

Command Unsuccessful
press enter Key to continue

[Steven Schweda]

> [...] --> Automounted Remote File Systems

And then what? "Run"? Around here, it says:

Run Command /usr/sam/lbin/samx -C -s nfs_client /usr/sam/lib/C/nfs.ui

Try that command from a shell?

In SMH, is the "Display" setting correct?
On my system, it seems always to point to the
server itself, even when I had DISPLAY set to
some other system when I ran SMH.

Anything under the SMH "stderr" tab when the
command is finished?

[Dave Olker]

I think Pete is asking the right questions. Rpc.lockd appears to be starting as part of the lockmgr script but when the nfs.client script tries to start it finds rpc.lockd is not running. So the real issue to figure out is why does rpc.lockd not continue running after it's started.

My first question is what version of ONCplus are you running? Issuing the command:

# swlist ONCplus

will tell you the exact version. If you're not running the latest version of ONCplus you can try installing the latest bits available for download from:

https://software.hp.com/portal/swdepot/displayProductInfo.do?productNumber=ONCplus

Once you're running a version of ONCplus that includes debug KLM logging (we added that feature in ONCplus B.11.31.01), I'd recommend you enable debug KLM logging via the command:

# kctune klm_log_level=9

Once you've enabled debug logging, try the following:

# /sbin/init.d/lockmgr stop
# /sbin/init.d/lockmgr start

If it says rpc.lockd started, use the ps command to see if it is still running or if it died:

# ps -ef | grep rpc.lockd

If the daemon died, look for a core file either in /, or in /usr/sbin or in the directory you executed the "lockmgr start" command. I'd then examine any debug log messages from rpc.lockd when it tried to come up. These should be logged to dmesg or to the /var/adm/syslog/syslog.log file.

The entries generated by rpc.lockd and the KLM subsystem will all begin with the string "KLM:"

Hopefully either updating ONCplus or enabling debug KLM logging will help you make some forward progress.

Regards,

Dave

[senthil_kumar_2]

Hi All,

I am new to HP-UX.

1)I have find that the follwoing services only have been started after system reboot.

1.1)nfs.core:

# ps -ef | grep -i rpcbind
root 1702 1 0 09:03:39 ? 0:00 /usr/sbin/rpcbind


1.2)rpc.statd only has been started from lockmgr but not rpc.lockd:

# ps -ef | grep -i rpc.statd
daemon 1713 1 0 09:03:46 ? 0:00 /usr/sbin/rpc.statd

# ps -ef | grep -i rpc.lockd
(no output)


As per above outpiut the file /sbin/init.d/lockmgr has been called during boot up.
but it has started only "rpc.statd" and not "rpc.lockd".




2)I have used following commands to start manually:

2.1)Stopped nfs.core and lockmgr:

# /sbin/init.d/nfs.core stop
killing rpcbind


# /sbin/init.d/lockmgr stop
killing rpc.statd

(here it is killing only rpc.statd since rpc.lockd is not started.)



2.2)Starting nfs.core, lockmgr and nfs.server:

# /sbin/init.d/nfs.core start
Starting NFS CORE networking

Starting up the rpcbind
/usr/sbin/rpcbind


# /sbin/init.d/lockmgr start
Starting up the Status Monitor daemon
/usr/sbin/rpc.statd
Starting up the lock manager daemon
/usr/sbin/rpc.lockd

Now /rpc.lockd" has been started:

Ex:

# ps -ef | grep -i rpc.lockd
root 1719 1 0 09:03:46 ? 0:00 /usr/sbin/rpc.lockd


2.3)# /sbin/init.d/nfs.server start
Starting NFS SERVER subsystem

Reading in /etc/dfs/dfstab
Starting up the mount daemon
/usr/sbin/rpc.mountd
Starting up the NFS server daemon
/usr/sbin/nfsd
Starting up nfsmapid daemon
/usr/sbin/nfsmapid

Now "nfs.server" is starting successfully.


my understanding is:

1) There is no issues with file "/sbin/init.d/lockmgr", since it is starting "rpc.statd" and "rpc.lockd" successfully while trying manual.


2) There might be issues with file "/sbin/init.d/nfs.core" or while starting this file at boot up,
since i have tried to restart (stop and start) the "/sbin/init.d/lockmgr" with out restarting (stop and start) nfs.core.
That time also the "rpc.lockd" is not started.

But when i restarted (stop and start) the "/sbin/init.d/lockmgr" after restarting (stop and start) nfs.core. now the "rpc.lockd" is started successfully.

So i believe that there may be issues while starting nfs.core at boot up.

And i believe there is no issues with /etc/rc.config.d/nfsconf since it starting the "lockmgr" at boot up.


Pls help me to resolve this since i am facing this issue for all my new hp-ux 11.31 running on integrity VM.


I have attached my "/sbin/init.d/nfs.core" and "lockmgr"

[Pete Randall]

And what about all the diagnostics that Dave Olker suggested? Did you run any of those?

Trust me, Dave is the #1 source for networking wisdom. You would be very wise to take his words as gospel and follow his suggestions explicitly.


Pete

[sujit kumar singh]

hi


what does this file /etc/default/nfs contain on the system.

regards
sujit

[sujit kumar singh]

hi also


on the NFS server can you please run the following commnd and give the O/p

/sbin/init.d/nfs.client start
/sbin/init.d/nfs.client stop



to see what NFS services are satrtted and stopped due to this.



also in the server in /etc/rc.config.d/nfsconf try using
NFS_CLIENT=1

regards

[James R. Ferguson]

Hi (again):

I had asked, "did you make a backup of '/etc/rc.config.d/nfsconf' when you changed its settings? If so, did you retain that backup copy in '/etc/rc.config.d'?"

I can't see where you responded to this question. Please answer this.

Regards!

...JRF...

[senthil_kumar_2]

Hi All,

I am finding below mentioned errors in /var/adm/syslog/syslog.log, this logs are being created while starting "lockmgr" with out stopping "nfs.core" immediately after reboot.


--------------------------------------------
Dec 21 11:33:11 emdlaghpvm04 /usr/sbin/rpc.lockd[1735]: Cannot get address for transport udp host \1 service lockd
Dec 21 11:33:11 emdlaghpvm04 /usr/sbin/rpc.lockd[1735]: Cannot establish NLM service over /dev/udp: transport setup problem.
Dec 21 11:33:11 emdlaghpvm04 /usr/sbin/rpc.lockd[1735]: Cannot get address for transport tcp host \1 service lockd
Dec 21 11:33:11 emdlaghpvm04 /usr/sbin/rpc.lockd[1735]: Cannot establish NLM service over /dev/tcp: transport setup problem.
Dec 21 11:33:11 emdlaghpvm04 /usr/sbin/rpc.lockd[1735]: Could not start NLM service for any protocol. Exiting.
Dec 21 11:33:32 emdlaghpvm04 /usr/sbin/rpc.lockd[1765]: Cannot get address for transport udp host \1 service lockd
Dec 21 11:33:32 emdlaghpvm04 /usr/sbin/rpc.lockd[1765]: Cannot establish NLM service over /dev/udp: transport setup problem.
Dec 21 11:33:32 emdlaghpvm04 /usr/sbin/rpc.lockd[1765]: Cannot get address for transport tcp host \1 service lockd
Dec 21 11:33:32 emdlaghpvm04 /usr/sbin/rpc.lockd[1765]: Cannot establish NLM service over /dev/tcp: transport setup problem.
Dec 21 11:33:32 emdlaghpvm04 /usr/sbin/rpc.lockd[1765]: Could not start NLM service for any protocol. Exiting.
-------------------------------------------

Hi sujit,

I am getting below error message when starting "nfs.client"

# /sbin/init.d/nfs.client start
Starting NFS CLIENT subsystem

ERROR: rpc.lockd not running. Run "/sbin/init.d/lockmgr start" to start rpc.lockd, exiting


Hi James,

I checked, i did not take any backup of nfsconf.

[sujit kumar singh]

Hi


a template copy of nfsconf can be found at /usr/newconfig/etc/rc.config.d/nfsconf,

you shall need to edit that when using as /etc/rc.config.d/nfsconf

Senthil,

you can add as earlier said to /etc/rc.config.d/nfsconf on the server
NFS_CLIENT=1


also please post the O/p of /etc/default/nfs

regards

[senthil_kumar_2]

Hi Sujit,

Please find attachment of /etc/default/nfs

[Dave Olker]

Here's what I find troubling in those syslog error messages:

> Dec 21 11:33:11
> emdlaghpvm04 /usr/sbin/rpc.lockd[1735]:
> Cannot get address for transport udp
> host \1 service lockd

The "host \1" is what concerns me. Are you absolutely certain your hostname resolution mechanisms are configured properly? The /etc/hosts file doesn't have any bogus information in it? Which ever hostname resolution database you're using (dns, nis, ldap, /etc/hosts) contains correct information for this system? I'd also check the /etc/services file to ensure it contains correct information and no bogus characters.

Regards,

Dave

[senthil_kumar_2]

Hi Dave,

I am having the same issue for all of newly installed five servers (HP-UX 11.31) on Integrity VM.

[OldSchool]

"I am having the same issue for all of newly installed five servers (HP-UX 11.31) on Integrity VM."

you might consider answering the questions he asked.

Eliciting relevant information from you has been somewhat less than enlighteing at this point.

[Steven Schweda]

> I am having the same [...]

> Dec 21, 2009 19:12:14 GMT 8 pts

It's nice that you've awarded points, but it
might be even nicer if you would ANSWER THE
QUESTIONS. The non-psychics among us can't
get any information from your system(s)
unless _you_ provide it.