Boy. I solved my problem. The 1 RU units we are using for slave nodes have 2 NICs in them. This may not solve Ed's problem, but if he does have dual NICs, try moving the ethernet cable to the other one!!

The install worked fine. The chassis of the slave nodes labels the NICs 1 and 2, but I guess the sysimager doesn't really care about eth0 and eth1. Once the RH 7.3 kernel was installed it configured eth0, which was not the interface the install ran on. Of course I assumed the labeled NIC 1 would correspond to eth0 in Linux world. Ha. A bit like Solaris changing the way they number SCSI devices between I think 2.5 and 2.6.

I am just hoping that the MAC addresses listed for the nodes which no longer match the interfaces used for the cluster doesn't come back to haunt me. May be another install to be safe!


**********************************************************************
Henry Johnson Ph.D., CIRES Computing Facility,
Cooperative Institute for Research in Environmental Sciences
Campus Box 216,
Boulder, CO 80309-0216
phone (303) 735-2377, fax (303) 492-1149
email: [EMAIL PROTECTED]
Web: http://cires.colorado.edu
**********************************************************************



-------------------------------------------------------
This SF.NET email is sponsored by:
SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
http://www.vasoftware.com
_______________________________________________
Oscar-users mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/oscar-users

Reply via email to