James, >From what I can tell, the SLES7 initrds all had a problem with passing the OSA portname parameter to the driver (and doing it correctly). I always had to issue the commands manually to get the driver up and running. I never took the time to figure out just what the problem was. One option would be to manually tweak the initrd so that /etc/modules.conf and /etc/chandev.conf had the correct information in it. Then, you might be able to get the network up and running with only a couple of commands.
Mark Post -----Original Message----- From: James Melin [mailto:[EMAIL PROTECTED] Sent: Monday, July 07, 2003 9:28 AM To: [EMAIL PROTECTED] Subject: SuSE 7.2 recovery/install system CD #3 and OSA GBE I attempted to IPL the installation/recovery system from the SuSE 7.2 Patch CD (the one with the kernal on it that supports 3390-9's properly) and found that there was a problem with attempting to allocate the Gigabit OSA card. It asks me if I am above service level 4139, I believe it was, and I said yes. It then asked me for the port name to use. I supplied the port as specified on my running/working production system. The process then went through and attempted to match against all OSA hardware and that port even though I'd said 8B0 as the device number to start with. It claimed that it could not find an osa/portname combination that matched. I ended up doing my DR volume backups by hand. Mostly I was doing this because I wanted a procedure to give to the Sungard operator to follow to get up to a point where I can mount my dasd, edit configuration files and otherwise prepare a system to come up in LPAR mode (no VM) on the DR hardware. I am also going to attempt this with VM, however since I don't have an exact match of networking hardware, I'm not sure what the thing is going to do when we try it with VM. That's neither here nor there at the moment, and simply will be a nice plus.
