Re: [expert] followup to boot failure message
On Sunday 27 July 2003 04:45, Todd Lyons wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 David E Fox wanted us to know: installing a new DVD/CDRW combo drive last weekend. I rebooted the system (which had been rebooted before, successfully) and the system hung completely at finding module dependencies (I think). My cooker box did the same thing a couple weeks ago. Only way to get a good bootup was failsafe mode. It gets better, I promise :) This probably better belongs in Cooker ML. - -- Blue skies... Todd On a simular lines: Since 2.4.21-0.13 booting my laptop hangs if my pcmcia CD-Rom is inserted. Not starting pcmcia services and starting manually and/or inserting the card after boot solves that. On top every now and then the CD will not unmount requiring me to stop devfsd first. The latter happens on my desktop system (supermount=disabled) too. -- Good luck, HarM Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
[expert] followup to boot failure message
I think I found a problem in /etc/lilo.conf Originally I had /dev/ram3 in the 2.4.19-35 image, and that didn't work. Also I noticed boot= set to /dev/hda, which is /var, and my boot is on /dev/hdb. However, I have a boot.0300 file. I'm going to switch it back to /dev/hda for now. I think my lilo is stored in /dev/hda for some reason, and I don't want to fiddle. The only thing I can think of to explain the 'finding module dependencies' issue is the presence of 'append devfs=nomount' vs 'yes' in the appended lilo.conf. One stanza says yes (mount), the other says no. But does that explain not being able to use the cdrecorder? And I was incorrect on the kernel issue. Apparently both are using linux 2.4.19-35mdk, since vmlinuz points there. I've set it up so that one will boot 2.4.19-16, the other 2.4.19-35. But I don't think that will make much difference. I'll try and reboot the 'linux' image now and report back. boot=/dev/hdb map=/boot/map default=linux keytable=/boot/us.klt prompt nowarn timeout=100 message=/boot/message menu-scheme=wb:bw:wb:bw image=/boot/vmlinuz-2.4.19-16mdk label=linux root=/dev/hdb6 initrd=/boot/initrd.img append=devfs=mount hdc=ide-scsi read-only image=/boot/vmlinuz label=failsafe root=/dev/hdb6 initrd=/boot/initrd.img append=devfs=nomount hdc=ide-scsi failsafe read-only other=/dev/fd0 label=floppy unsafe image=/boot/vmlinuz-2.4.19-35mdk label=2419-35 root=/dev/hdb6 read-only optional vga=788 append= initrd=alt0/all.rdz ramdisk_size=32000 initrd=/boot/initrd-2.4.19-35mdk.img Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [expert] followup to boot failure message
On Sat, 2003-07-26 at 13:03, David E Fox wrote: I think I found a problem in /etc/lilo.conf Originally I had /dev/ram3 in the 2.4.19-35 image, and that didn't work. Also I noticed boot= set to /dev/hda, which is /var, and my boot is on /dev/hdb. However, I have a boot.0300 file. I'm going to switch it back to /dev/hda for now. I think my lilo is stored in /dev/hda for some reason, and I don't want to fiddle. The only thing I can think of to explain the 'finding module dependencies' issue is the presence of 'append devfs=nomount' vs 'yes' in the appended lilo.conf. One stanza says yes (mount), the other says no. But does that explain not being able to use the cdrecorder? And I was incorrect on the kernel issue. Apparently both are using linux 2.4.19-35mdk, since vmlinuz points there. I've set it up so that one will boot 2.4.19-16, the other 2.4.19-35. But I don't think that will make much difference. I'll try and reboot the 'linux' image now and report back. boot=/dev/hdb map=/boot/map default=linux keytable=/boot/us.klt prompt nowarn timeout=100 message=/boot/message menu-scheme=wb:bw:wb:bw image=/boot/vmlinuz-2.4.19-16mdk label=linux root=/dev/hdb6 initrd=/boot/initrd.img append=devfs=mount hdc=ide-scsi read-only image=/boot/vmlinuz label=failsafe root=/dev/hdb6 initrd=/boot/initrd.img append=devfs=nomount hdc=ide-scsi failsafe read-only other=/dev/fd0 label=floppy unsafe image=/boot/vmlinuz-2.4.19-35mdk label=2419-35 root=/dev/hdb6 read-only optional vga=788 append= initrd=alt0/all.rdz ramdisk_size=32000 initrd=/boot/initrd-2.4.19-35mdk.img David, Try this move /etc/lilo.conf /etc/lilo.conf.bkp then create a new lilo.conf this way boot=/dev/hdb map=/boot/map default=linux vga=788 keytable=/boot/us.klt prompt nowarn timeout=100 message=/boot/message menu-scheme=wb:bw:wb:bw image=/boot/vmlinuz-2.4.19-16mdk label=linux root=/dev/hdb6 initrd=/boot/initrd-2.4.19-16mdk.img append=devfs=mount hdc=ide-scsi read-only image=/boot/vmlinuz-2.4.19-16mdk label=failsafe root=/dev/hdb6 initrd=/boot/initrd-2.4.19-16mdk.img append=devfs=nomount hdc=ide-scsi failsafe read-only other=/dev/fd0 label=floppy unsafe image=/boot/vmlinuz-2.4.19-35mdk label=2419-35 root=/dev/hdb6 read-only append=devfs=nmount hdc=ide-scsi initrd=/boot/initrd-2.4.19-35mdk.img Then run lilo and try your reboot into one of the regular kernels. James Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [expert] followup to boot failure message
Try this move /etc/lilo.conf /etc/lilo.conf.bkp then create a new lilo.conf this way Hmm. I tried something along those lines and still had the same issue. 2419-35 boots in vga mode and tries to find the root partition, then panics. 'linux' starts up but can't get past the fixing module dependencies stage.. I still haven't been able to sort that out. 'failsafe' works; at least i can get up to telinit 3. About the cdrw - seems ok now, after a power cycle. The drive would just sit there with the drive light on when I inserted a blank cdr and not when I inserted a written CD. So now I'm burning yet another CD. k3b managed to choke and burn and that was the start of the trouble. I can't for the life of me -- at least not yet -- figure out the issue with respect to the fixing module dependencies step... :( it would be nice to know wtf is going on at that point. It was the reason why I lost /dev/hdb7, my more or less up to date cooker with all the goodies :). I'm gonna save off this message and compare it to the existing /etc/lilo.conf. David E. Fox Thanks for letting me [EMAIL PROTECTED]change magnetic patterns [EMAIL PROTECTED] on your hard disk. --- Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [expert] followup to boot failure message
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 David E Fox wanted us to know: I think I found a problem in /etc/lilo.conf Originally I had /dev/ram3 in the 2.4.19-35 image, and that didn't work. Also I noticed boot= set to /dev/hda, which is It sounds like you installed the kernel rpm in rescue mode. When you do that, it grabs the current kernel commandline and attempts to keep all the stuff it thinks it needs to. It seems that it grabbed the components used to boot into rescue mode instead of figuring out which settings it needs from the lilo.conf file. I can't blame anybody for that as that could be insanely difficult, but it probably ought to be tweaked a little to not use the ramdisk and initrd settings from rescue mode. I'm cc'ing this to gc and pixel for that. - -- Blue skies... ToddPublic key: http://www.mrball.net/todd.asc There's a medical term for that: freakishly dorky. --Gash Oh wait, is today cynical Friday? --NANOG Linux kernel 2.4.19-24mdk 4 users, load average: 0.00, 0.14, 0.18 -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.2 (GNU/Linux) Comment: http://www.mrball.net/todd.asc iD8DBQE/IxpgIBT1264ScBURAvnjAJ0WDK+wzkj7uU5BLKh40L6jLCmXFACdFrqs oWHNUwmP/Ry160cVet+3Ovk= =0mQm -END PGP SIGNATURE- Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [expert] followup to boot failure message
It sounds like you installed the kernel rpm in rescue mode. When you do that, it grabs the current kernel commandline and attempts to keep all That may explain what happened. You might know that originally I managed to hose (or I think I did) my /dev/hdb7 install which was cooker, when installing a new DVD/CDRW combo drive last weekend. I rebooted the system (which had been rebooted before, successfully) and the system hung completely at finding module dependencies (I think). I couldn't boot /dev/hdb7 so I installed mandrake 9.0 on /dev/hdb6. Then I had to install all the missing things (only had one CD of that) just to get to the point where I could burn CDs. I managed to download all three 9.1 ISOs and burn them. I started to reinstall 9.1, and the box locked up solidly just before the exit install step. So, back to square 1. I tried other things too, and that's maybe when I screwed up. ATM, /dev/hdb6 has a 9.0 that's been partially upgraded to cooker. /dev/hdb7 still exists, I just can't boot it, but it'll probably not work since /var has been rebuilt since then (/dev/hda1) so /var/lib/rpm and so forth are way out of sync with respect to what's on /dev/hdb7. So I might as well stick it out with what I have, at least until I have a stable enough cooker from which to burn from. (I mirror usually 2x a day with wget from ftp.sunet.se.) And everything else, I get via urpmi. (James has been helpful I'll admit -- thanks James)... Now ATM I have a working setup - the CD issue seems to have fixed itself, dunno why. But I can only boot in failsafe mode. Otherwise, I get either 1) same issue with respect to the hang at module dependencies or 2) can't find the root filesystem. Once failsafe boots, telinit 3 can bring up the rest of the system, and then I can startx. So - moral - don't reboot :). Blue skies... ToddPublic key: http://www.mrball.net/todd.asc There's a medical term for that: freakishly dorky. --Gash Oh wait, is today cynical Friday? --NANOG David E. Fox Thanks for letting me [EMAIL PROTECTED]change magnetic patterns [EMAIL PROTECTED] on your hard disk. --- Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [expert] followup to boot failure message
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 David E Fox wanted us to know: installing a new DVD/CDRW combo drive last weekend. I rebooted the system (which had been rebooted before, successfully) and the system hung completely at finding module dependencies (I think). My cooker box did the same thing a couple weeks ago. Only way to get a good bootup was failsafe mode. It gets better, I promise :) This probably better belongs in Cooker ML. - -- Blue skies... Todd | Get a bigger hammer! | schroom niet--if you consider doing| | http://www.mrball.net | it, stop cosidering and *do* it. | | http://faq.mrball.net | --Wouter Teepe | Linux kernel 2.4.19-24mdk 5 users, load average: 0.00, 0.01, 0.00 -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.2 (GNU/Linux) Comment: http://www.mrball.net/todd.asc iD8DBQE/IzzTIBT1264ScBURAs2bAKCCS9Nd633bJBWJPch64fPCtX1F5wCeJrii 21pII8jjA0tbN0+iGHsPQSs= =46NX -END PGP SIGNATURE- Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
