[Linux-cluster] new dlm control/configuration

[David Teigland]

Hi,

We've recently done some re-engineering to the dlm to make it more easily
adaptable to other environments.

The dlm is now configured/controlled from user space using sysfs.  It
depends on nothing in the kernel that's not already there and it's
completely agnostic about who or what is controlling it from user space.
(Another unrelated change is that it now uses SCTP intead of TCP.)

Someone or thing in user space now just tells the dlm who the nodes are
who are using it.  This happens per-lockspace, of course, as different
groups of nodes may be using different lockspaces.

A new command line program, dlm_tool, can be used to set up the dlm
manually in which case it depends on no other software (much like using
dmsetup with device-mapper.)  Of course, some sort of user-level
membership system would usually be used to control the dlm instead of
dlm_tool, but that's completely variable now (like using either LVM2 or
EVMS over device-mapper).

Here are instructions on how to use dlm_tool.  It also illustrates what
the responsibilities of a membership system would be when dealing with the
dlm:
http://sources.redhat.com/cgi-bin/cvsweb.cgi/cluster/dlm/doc/dlm_tool.txt?cvsroo
t=cluster

We're interested in some feedback -- does this make the dlm more useful
for you and the general public?  If not, what might?  We'd like to see
this dlm be used beyond GFS and CLVM.  A roadblock in the past was the
disparate clustering and membership managers.  While consolidation
continues slowly on that front, the dlm should be able to move ahead by
remaining agnostic.

This also clears the way for submitting the dlm to be included in the main
kernel.  We're currently working on making the code suitable for that
level of inspection after which (very soon) we'll send it out as a set of
kernel patches.

new dlm kernel code:

http://sources.redhat.com/cgi-bin/cvsweb.cgi/cluster/dlm-kernel/src2/?cvsroot=cl
uster

dlm_tool to configure/control the dlm manually:

http://sources.redhat.com/cgi-bin/cvsweb.cgi/cluster/dlm/?cvsroot=cluster


GFS (which uses the in-kernel dlm locking API) will be configured and
controlled from user space in the same way, remaining agnostic about who
or what is driving things.  An external membership system would typically
be used to control both the dlm and gfs since they both require very
similar control input.  It's worth noting that the dlm remains a general
purpose, VMS-like dlm, completely independent from GFS or any other
application.

-- 
Dave Teigland  <teigland at redhat.com>