==> On Mon, 29 Jan 2007 13:14:17 -0500, Jeff Moyer <[EMAIL PROTECTED]> said:
Jeff> From: "Murata, Dennis" <[EMAIL PROTECTED]> Jeff> Subject: FW: Advise for multiple directory/multi-map mounts Jeff> To: <[EMAIL PROTECTED]> Jeff> Date: Mon, 29 Jan 2007 09:57:03 -0800 Jeff> Jeff, Jeff> I am sorry about sending this directly to you, but I have not been able Jeff> to post anything to the autofs mailing list in quite awhile. I think I Jeff> will unsubscribe then resubscribe. Would you please post this for me? Jeff> Thanks Jeff> Wayne Jeff> > ______________________________________________ Jeff> > From: Murata, Dennis Jeff> > Sent: Friday, January 26, 2007 2:06 PM Jeff> > To: [email protected] Jeff> > Subject: Advise for multiple directory/multi-map mounts Jeff> > Jeff> > After reading the FAQ and various postings from this group, there are Jeff> > a couple of options that I need advise with the approved syntax and Jeff> > any recommendations for either one of the options. Jeff> > Jeff> > auto.master entry: Jeff> > /foo /etc/auto.foo Jeff> > indirect map auto.foo: Jeff> > * -fstype=nfs,rsize=32768,tcp server1:/foo1/& server2:/foo2/& Jeff> > Jeff> > Server1 and server2 may or may not be the same server, but the server Jeff> > directories foo1 and foo2 are definitely distinct. The subdirectories Jeff> > under foo1 and foo2 do not have overlapping names, but may be moved Jeff> > from one server foo1, to the other server directory foo2. I want it Jeff> > to appear transparently to the client. Because directories are going Jeff> > to be moved, should no_subtree_check be used? This should work, but only in autofs v5. You don't mention anywhere what version of autofs you are using, or under what kernel. There is a chance you can get this to work with v4, if you use the rhel4 packages from my people page: http://people.redhat.com/jmoyer/autofs/rhel4/ These are only test packages, so the "may eat your data" disclaimer applies. Once I've put the patches through the proper testing, I'll of course submit them for 4.1. However, the right way to move forward is to test out autofs v5. -Jeff _______________________________________________ autofs mailing list [email protected] http://linux.kernel.org/mailman/listinfo/autofs
