[ Back on list ]
Wondering if this is a hardware issue. Switch negotiation still going on while machine is booting... NIC start fails... home doesn't get mounted, etc.
I assume you've watched the console for failures during startup? The reason is sounds a bit like a switch is because of the inconsistency. Some nodes work on a restart, some don't... and it's not consistent. Sounds like a switch, no?
Jeremy
At 02:00 PM 8/26/2002 -0700, David Dustan wrote:
Your understanding is correct. Let me tell you what I am doing..... 1. We boot up the Master. 2. Clients are booted up. 3. After 5-10 minutes we are able to ssh some of the nodes. We run a home made script to mount nfs (home). It fails on nodes that do not have a working network connection. 4. On the non-working nodes we run /etc/init.d/network restart Strangely, it takes two times of doing this! At this point we have access to all 47 client nodes. We have finished our cluster setup and testing but need to do these steps.The clients are using a 3com 3c996B-T and the master is using the same. The ifconfig -a and the ifcfg-ethx (from the master or client?) will follow. We are buttoning up the cabinets....but will be doing more testing/configuring. David David Dustan Senior Systems Engineer Puget Sound Data Systems, Inc. 10236 E. Riverside Dr. Bothell, WA 98011 (425) 488-0710 mailto:[EMAIL PROTECTED] -----Original Message----- From: Thomas Naughton [mailto:[EMAIL PROTECTED]] Sent: Monday, August 26, 2002 11:38 AM To: David Dustan Cc: 'Jeremy Enos' Subject: RE: [Oscar-users] SSH Information Needed David, Just to make sure I understood your response. * The nodes come up and the machine is not SSH'able. * So, you type 'service network restart', and it is then SSH'able. Additionally: + What type of NIC are you using in the nodes? + Could you paste the results from 'ifconfig -a' [on offending node(s)] + Also, attach the contents of '/etc/sysconfig/network-scripts/ifcfg-eth*' As mentioned previously, sounds like your network interfaces aren't behaving properly (thus the mount/ssh/ping problems). thanks, --tjn _________________________________________________________________________ Thomas Naughton [EMAIL PROTECTED] Research Associate (865) 576-4184
------------------------------------------------------- This sf.net email is sponsored by: OSDN - Tired of that same old cell phone? Get a new here for FREE! https://www.inphonic.com/r.asp?r=sourceforge1&refcode1=vs3390 _______________________________________________ Oscar-users mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/oscar-users </x-flowed>
