OpenSSI Logo SourceForge Logo

project
 home page
 sourceforge page
 mailing lists
 feature list
 demos
 screenshots
 Bruce's corner
 related links
 wiki
downloads
 notes
 1.2 stable
 1.9 development
 CVS
documentation
 1.2 stable
 1.9 development
 roadmap
 wiki
work items
 task list
 bug database
 feature requests
 process mgmt hooks
  hide sidebar
Configuring Automounter(amd) in an OpenSSI Cluster
==================================================
Automounter(amd) on openssi cluster is very clusterwide, 
automatic, and transparent.  Automounter daemon will be running
only on init node and all other nodes of the cluster can access 
the files under auto mounted directory. It automatically mounts
on all nodes which are up, and by any node that subsequently 
joins.  
Initnode where amd server is running must have network access to 
the remote NFS server.  The NFS server must export to each node in 
the cluster. If either of these are not met, mounts will not succeed.  
Note that if a client node in the cluster fails, the only locks lost 
are those held by processes on that node at the time of the failure.

This is a general document for automounter setup on an OpenSSI cluster. 
It is written to apply to Fedora, as well as Debian, with indications
to explain differences between the distributions.

NOTE: On Fedora/Redhat, the equivalent commands for `invoke-rc.d` 
      and `update-rc.d` are `service` and `chkconfig`, respectively. 

Requirements:
	All cluster nodes must have unique host names that correspond 
	to an external interface
	NFS server must export file systems to all cluster nodes

Set-up steps:
-------------
1. Make sure the hostname of each cluster node must corresponds to an 
   external interface. It should not be cvip address. 

2. The portmap service is needed on all distributions.

3. On Debian, the "am-utils" service must be started on init node.
   On Fedora, the "amd" service must be started on init node.
   /etc/rc.nodeinfo should have an entry for starting portmap and 
   am-utils on init node of the cluster.

4. Review standard Debian/Fedora amd configuration documentation

The default file to be configured is "/etc/am-utils/amd.conf" on Debian
and "/etc/amd.conf" on Fedora. Please see the standard Fedora or  
Debian document to have different configuration file name

Example configuration entries are as shown below
/etc/am-utils/amd.conf(On Debian)
=============================
# At startup, do we "restart" existing mounts if we determine they could
# have been automounted ?
  restart_mounts = yes
  [/tek]
  map_name = /etc/am-utils/amd.tek
  map_type = file
=============================

In the above example configuration, /tek is an automount directory and
the map file is "/etc/am-utils/amd.tek" and type is file. The sample
map file looks like as follow

map file "/etc/am-utils/amd.tek"
==============================
#/defaults fs:=${autodir}/${rhost}/root/${rfs};opts:=nosuid,nodev
#*       rhost:=${key};type:=host;rfs:=/
/defaults type:=nfs;opts:=nosuid,quota,intr,rw
name-of-directory rhost:=server1.example.com;type:=nfs;rfs:=/name-of-remote-directory
==============================

How to test
==========
Upon execution of `onall mount` should display an entry for /tek in addition 
to other mount points

(node 1)
/dev/1/sdb1 on / type ext3 (rw,errors=remount-ro)
server1.example.com:(pid67481) on /tek type nfs (intr,rw,port=1023,
            timeo=8,retrans=110,indirect,map=/etc/am-utils/amd.tek)
proc on /proc type proc (rw,node=*)
(node 2)
/dev/1/sdb1 on / type ext3 (rw,errors=remount-ro)
server1.example.com:(pid67481) on /tek type nfs (intr,rw,port=1023,
            timeo=8,retrans=110,indirect,map=/etc/am-utils/amd.tek)
proc on /proc type proc (rw,node=*)

Troubleshooting:
---------------
1. Verify that portmapper is running on each node.
   (ps -ef --shownode | grep portmapper)
2. Verify that each node can access the remote nfs server;
3. Verify that the remote nfs server is exporting to each cluster node;
4. Verify that amd is running (`ps -ef --shownode | grep amd`)


NOTE : 
------
'Amd' failover is not supported well in this version of am-utils. In case 
you have your init node failed over and 'amd' is restarted on the new 
init node, 'amd' will not cleanly restart its own mount and also previous 
auto-mount points properly. Make sure you unmount stale/old 'amd' mounts 
manually and restart 'amd' afresh. (This holds true until we support 
am-utils 6.1 with a better 'amd' failover capability).

This page last updated on Thu Dec 15 17:17:43 2005 GMT
privacy and legal statement