Well, you've already spent more time trying to figure this out than you would if you just do a re-install. So, why not try to "do it right" all the way through so you _know_ for sure what you've got on your hands? (I sure wouldn't want to inherit a system that had been glued back together from day 1.)
If you use CDL DASD, which is the default these days, you must create at least one partition on the volume before doing the mkfs on it. No ifs, ands, or buts about it. In your situation, I would do the re-install. Of course, that's your choice to make, not mine. Mark Post -----Original Message----- From: Linux on 390 Port [mailto:[EMAIL PROTECTED] Behalf Of Sue Sivets Sent: Thursday, February 05, 2004 6:37 PM To: [EMAIL PROTECTED] Subject: Re: Ipl problem after instal on vm mdisk Mark, I did not run fdasd to partition the 151 disk. I didn't think I needed to if I was only going to have a single partition. Both the 150 & the 151 mdisks are on the same volume. 150 is supposed to be 150 cyls, and 151 is supposed to be 2560 cyls. The zeros at address 0 bother me too. I was expecting to see numbers and almost anything except zeros. One of the people I work with suggested zipping everything up and copying it to another vol, so I could re-format, copy back, and then re-zipl. I guess it's worth a shot, and will certainly take a lot less time then re-installing to find out if it's going to work. If it doesn't work, I've only lost an hour or two. If it works, I've still got the "what didn't get installed" problem, but this is only going to be a test system for a couple of people to play with, so anything they find missing that they want can always be installed later as long as I can get the system to boot. Besides with my luck, the way this thing has been going, I'm going to have to re-install the whole thing again anyway. If you've got more ideas, I'm willing to try them out. Thank you. Sue "Post, Mark K" wrote: > Sue, > > /dev/dasdb1 looks more to be the size of a swap partition than a data > partition. 127776 1K blocks comes out to about 125MB. The 150 disk is about > 140MB in size, also small enough to be a swap disk. > > Yes, -cdl is the default for dasdfmt these days. When you ran fdasd (you > _did_ run fdasd, right?), did you create more than one partition on 151? > The 151 volume looks to be about 1,800 MB in size (4096*460800/1024/1024), > so the fact that only 125MB of it is being used is suspicious. > > Also, the dd | od stuff looks somewhat different from what I see on my > system. The ending stuff is the same, but I have non-zero values starting > at address 0: > # dd if=/dev/dasda bs=512 count=12 | od > 0000000 144727 151761 000010 000000 100000 020010 003000 001030 > 0000020 060000 000220 004000 001050 000000 000000 162745 162745 > 0000040 162745 162745 162745 162745 162745 162745 162745 162745 > * > 0010000 144727 151762 000017 000013 001420 000000 000000 000000 > 0010020 000000 000000 003400 001240 060000 000006 030400 001242 > 0010040 060000 000005 004000 001060 000000 000000 003000 020000 > 0010060 020000 010000 003000 000000 060000 000000 003000 000000 > 0010100 060000 000000 003000 000000 060000 000000 003000 000000 > * > 0010200 060000 000000 003000 000000 020000 000000 000000 000000 > 0010220 000000 002000 162745 162745 162745 162745 162745 162745 > 0010240 162745 162745 162745 162745 162745 162745 162745 162745 > * > 0014000 > 12+0 records in > 12+0 records out > > That's not a good sign either, perhaps, but I'm not 100% sure. > > Mark Post > > -----Original Message----- > From: Linux on 390 Port [mailto:[EMAIL PROTECTED] Behalf Of Sue > Sivets > Sent: Thursday, February 05, 2004 3:56 PM > To: [EMAIL PROTECTED] > Subject: Re: Ipl problem after instal on vm mdisk > > Hi Mark, I ipl'ed from the VM reader, and I'm pretty sure I copied the right > files from the cd since I just about followed the Suse install instructions > word > for word. The one thing I did not do was to specify "-d cdl" on the dasdfmt > command because I thought it was supposed to be the default. It's beginning > to > look like I'm going to have to step back and punt. Boo Hissssssssss. > > >From the ramdisk system (after yast started and dasd addresses loaded): > inst-sys:~ # cat /proc/dasd/devices > 0150(ECKD) at ( 94: 0) is dasda : active at blocksize: 4096, 36000 > blocksB > > 0151(ECKD) at ( 94: 4) is dasdb : active at blocksize: 4096, 460800 > blockB > > Output from df command: > inst-sys:~ # df /boot > Filesystem 1K-blocks Used Available Use% Mounted on > /dev/loop0 85692 85692 0 100% /mounts/instsys > inst-sys:~ # df /dev/dasdb > Filesystem 1K-blocks Used Available Use% Mounted on > /dev/dasdb 127776 20540 107236 17% > inst-sys:~ # df /dev/dasdb1 > Filesystem 1K-blocks Used Available Use% Mounted on > /dev/dasdb1 127776 20540 107236 17% > > 17% seems an awful lot like too small a percent used for the Suse8.0 > default > install. I remember that the LPAR installs of Suse 8.0 have been fairly > small, > but I don't think they were this small. Is it possible that the default > system > really didn't get comopletely installed? The 151 minidisk (dasdb) is 2560 > cyls > of a 3390-3. > > Output from dd command: > inst-sys:~ # dd if=/dev/dasdb bs=512 count=12 | od > 0000000 000000 000000 000010 000000 100000 020010 003000 001030 > 0000020 060000 000220 004000 001050 000000 000000 162745 162745 > 0000040 162745 162745 162745 162745 162745 162745 162745 162745 > * > 0010000 001000 000000 001754 000010 001420 000000 000000 000000 > 0010020 000000 000000 003400 001240 060000 000006 030400 001242 > 0010040 060000 000005 004000 001060 000000 000000 003000 020000 > 0010060 020000 010000 003000 000000 060000 000000 003000 000000 > 0010100 060000 000000 003000 000000 060000 000000 003000 000000 > * > 0010200 060000 000000 003000 000000 020000 000000 000000 000000 > 0010220 000000 002000 162745 162745 162745 162745 162745 162745 > 0010240 162745 162745 162745 162745 162745 162745 162745 162745 > * > 12+0 records in > 12+0 records out > 0014000 > > The virtual dasd # is 151, and that's what I'm using in the IPL command. > > Sue > > "Post, Mark K" wrote: > > > No, I wasn't joking, but that doesn't mean I was right, either. Sigh. If > > you can't tell, it's been a while since I did a SUSE install from the > > beginning. I was thinking of the "starter" kernel that you use, depending > > on whether you'll be IPLing from tape, or the VM reader. Sorry. Once > > you've gotten zipl run successfully, that characteristic of the kernel is > no > > longer relevant. > > > > I guess it's time to start asking more detailed questions. Show us the > > output of: > > cat /proc/dasd/devices > > df /boot > > dd if=/dev/dasdb bs=512 count=12 | od > > Substitute the device name of your IPL volume for this particular system > in > > place of /dev/dasdb, if necessary. Note that this is the device node, not > a > > partition node, so don't use /dev/dasdb1, or /dev/dasdb2, etc. > > > > Also, what virtual device number do you IPL from? > > > > Mark Post > > > > -----Original Message----- > > From: Linux on 390 Port [mailto:[EMAIL PROTECTED] Behalf Of Sue > > Sivets > > Sent: Wednesday, February 04, 2004 11:08 PM > > To: [EMAIL PROTECTED] > > Subject: Re: Ipl problem after instal on vm mdisk > > > > Hi Mark, when I first read your message I thought you were joking, but I'm > > almost hoping that you aren't. How do I tell, or can I tell if I'm using > an > > "LPAR" kernel? Does such a thing really exist? > > > > I copied the ramdisk boot files from the cdrom to a VM minidisk, punched > > them > > to the reader, and ipl'ed. I didn't get any error messages, or I guess I > > should > > say I didn't get any messages that looked like error messages. Then I did > a > > "normal" install, just like I the installs I've done in the LPARs. > > > > I can re-install if I have to, but needless to say, that's not my first > > choice > > of fun things on my long list of things to do. > > > > Sue > > > > "Post, Mark K" wrote: > > > > > Hmm, the other possibility (?) is that Sue is using an "LPAR" kernel, > and > > > not the one with VM support. > > > > > > Mark Post > > > > > > -----Original Message----- > > > From: Linux on 390 Port [mailto:[EMAIL PROTECTED] Behalf Of Rob > > > van der Heij > > > Sent: Wednesday, February 04, 2004 2:47 AM > > > To: [EMAIL PROTECTED] > > > Subject: Re: Ipl problem after instal on vm mdisk > > > > > > On Wed, 2004-02-04 at 08:08, Sue Sivets wrote: > > > > I just finished installing Suse Linux on a VM mini disk (previous > > installs > > > were > > > > in standalone lpars). When I tried to ipl the system for the first > time > > I > > > > received the following message: > > > > HCPGIR453W CP entered; program interrupt loop > > > > Since I wasn't at all sure the ipl text really was written, I linked > the > > > new mini > > > > disk to another linux guest and re-ran zipl. The system still doesn't > > ipl. > > > > Has anyone run into something like this before? Does anyone have any > > ideas > > > about > > > > what I can or should do next? > > > > > > Could it be that the userid is in XC mode (Q SET) now, or maybe short of > > > storage? If zipl did not write the IPL records you'd get different ones. > > > One option could be that the contents of /boot changed after zipl and it > > > now points the wrong way. You could make sure and start from the ramdisk > > > system again, load the dasd driver and re-run zipl. > > > If nothing else you could run TRACE I during IPL and see where it gets. > > > > > > Rob > > > > -- > > Suzanne Sivets > > Systems Programmer > > Innovation Data Processing > > 973-890-7300 > > Fax 973-890-7147 > > [EMAIL PROTECTED] > > -- > Suzanne Sivets > Systems Programmer > Innovation Data Processing > 973-890-7300 > Fax 973-890-7147 > [EMAIL PROTECTED] -- Suzanne Sivets Systems Programmer Innovation Data Processing 973-890-7300 Fax 973-890-7147 [EMAIL PROTECTED]
