HPE Community read-only access December 15, 2018
This is a maintenance upgrade. You will be able to read articles and posts, but not post or reply.
Hours:
Dec 15, 4:00 am to 10:00 am UTC
Dec 14, 10:00 pm CST to Dec 15, 4:00 am CST
Dec 14, 8:00 pm PST to Dec 15, 2:00 am PST
Operating System - OpenVMS
cancel
Showing results for 
Search instead for 
Did you mean: 

VMS NFS Clinet, dismount fails if server has failed

 
Brian Reiter
Valued Contributor

VMS NFS Clinet, dismount fails if server has failed

Hi Folks,

 

OK on an RX2800 riunning OpenVMS 8.4 Update v7.0, TCPIP version is 5.7 ECO 3 the NFS client fails to dismoun t the NFS share if the link to the server has gone, i.e. server has been rebooted or a cable has been disconnected. Apparently the /ALL switch on the TCPIP DISMOUNT should dismount the share even if the server is not available.

 

From the account of the software attempting to perform the dismount I get:

 

$ tcpip dism dnfs999: /all
%TCPIP$DNFSDISMOUNT-E-DISMOUNTFAIL, error dismounting DNFS999:[000000]
-RMS-E-FND, ACP file or directory lookup failed
-SYSTEM-F-TIMEOUT, device timeout

Presumably the user account doesn't have the privilege to use the /ALL switch. Using the system account I get:

 

(WMRCC1)SYSTEM>tcpip dism dnfs999: /all
%TCPIP$DNFSDISMOUNT-E-PARSEFAIL, error parsing DNFS999:
-RMS-E-ACC, ACP file access failed
-SYSTEM-F-TIMEOUT, device timeout

In both cases the NFS share remains. I can get rid of the shares (assuming no attempt to dismount them has been made) by restarting the NFS client software. The NFS shares are removed on the startup command.

 


So, anyone any ideas how to remove the shares?

 

 

Regards

 

Brian Reiter

6 REPLIES
Volker Halle
Honored Contributor

Re: VMS NFS Clinet, dismount fails if server has failed

Brian,

 

did you try an OpenVMS $ DISM/ABORT DNFSx: ?

 

Volker.

Brian Reiter
Valued Contributor

Re: VMS NFS Clinet, dismount fails if server has failed

Hi Volker, The DISM/ABORT marjs the share as being "Mounted dismount" but its sill present. cheers Brian
Volker Halle
Honored Contributor

Re: VMS NFS Clinet, dismount fails if server has failed

Brian,

 

there do not seem to be any 'special' commands in TCPIP$NFS_CLIENT_STARTUP.COM, which look like cleaning up old DNFS: devices. It's just re-loading the driver.

 

Maybe you were not patient enough and the DNFS: devices would have disappeared 'after a while' even without stopping and starting NFS client ?

 

Volker. 

Brian Reiter
Valued Contributor

Re: VMS NFS Clinet, dismount fails if server has failed

Hi Volker,

 

Are, I think we've come to that conclusion. The only issue really is working out jhow long it will take.

 

cheers

 

 

Brian

Volker Halle
Honored Contributor

Re: VMS NFS Clinet, dismount fails if server has failed

Brian,

 

use TCPIP SHOW MOUNT/FULL DNFSx: and look at the various values reported (things like RPC timeout, RPC retry limit, Directory time). Then shutdown the NFS server or remove the network cable and start a little DCL procedure to loop checking for the DNFSx: device every second with F$GETDVI("DNFSx:","MNT"). If the DNFS device has disappeared, report the time.

 

That should give you an idea, how long it takes for the DNFSx: device to disappear. Maybe you can figure out from the TCPIP SHOW MOUN/FULL DNFSx: output, which parameters are involved here.

 

Volker.

Brian Reiter
Valued Contributor

Re: VMS NFS Clinet, dismount fails if server has failed

Hi Volker,

 

Well it looks as though only the TCPIP DISMOUNT/ALL command actually does what I expected.

 

Although this is cpmplicated by the fact that our system uses aliases, there are two faolure modes (at least), one where the server goes down leaving the client with the NFS mount, the second where the alias used by the client no longer exists on the machie running the client software.

 

 

cheers

 

Brian