Bugzilla – Bug 114741
Grub error 22
Last modified: 2005-09-19 10:38:41 UTC
I was unable to successfully install the Grub boot loader into my mbr. I tried many times without success. I also tried to do this on both Kanotix and Libranet 3.0, both which use grub, and the same error occurred. My AMD64 system is brand new. I used Libranet and installed Lilo-config, and was able to load Lilo with no problem. If you need I could send you the grub.conf file. My Hardware ********************************************************************** Operating System System Model Windows XP Professional, Service Pack 1, v.1050 MICRO-STAR INTERNATIONAL CO., LTD MS-7125 1.0 Processor a Main Circuit Board 2233 megahertz AMD Pentium 4 128 kilobyte primary memory cache 512 kilobyte secondary memory cache Board: MICRO-STAR INTERNATIONAL CO., LTD MS-7125 1.0 Bus Clock: 201 megahertz BIOS: Phoenix Technologies, LTD 6.00 PG 01/12/2005 Drives Memory Modules 809.93 Gigabytes Usable Hard Drive Capacity 446.91 Gigabytes Hard Drive Free Space HL-DT-ST DVD-ROM GDR8163B [CD-ROM drive] HL-DT-ST DVDRAM GSA-4163B [CD-ROM drive] 3.5" format removeable media [Floppy drive] ST3160023A (137.44 GB) [Hard drive] -- drive 0 ST3200822A 3.01 (200.05 GB) [Hard drive] -- drive 4 ST3200822A (137.44 GB) [Hard drive] -- drive 2 ST3250823A (137.44 GB) [Hard drive] -- drive 3 ST360021A (60.02 GB) [Hard drive] -- drive 1 1024 Megabytes Installed Memory 4096 Megabyte Module Size - 2 Installed 2 Memory Sockets are Empty Local Drive Volumes c: (on drive 0) 160.00 GB 139.91 GB free g: (on drive 2) 200.00 GB 58.89 GB free i: (on drive 3) 249.94 GB 126.55 GB free j: (on drive 4) 200.00 GB 121.56 GB free Network Drives Controllers Printers NVIDIA nForce4 ADMA Controller NVIDIA nForce4 ADMA Controller NVIDIA nForce4 Parallel ATA Controller Primary IDE Channel [Controller] Primary IDE Channel [Controller] Primary IDE Channel [Controller] Secondary IDE Channel [Controller] Secondary IDE Channel [Controller] Secondary IDE Channel [Controller] Amyuni Document Converter 2.10 on LPT1: Display MSI MS-StarForce GeForce 6600 GT (NVIDIA GeForce 6600 GT) [Display adapter] Bus Adapters Multimedia None detected Realtek AC'97 Audio Communications Other Devices MAC Bridge Miniport MAC Bridge Miniport - Packet Scheduler Miniport Marvell Yukon 88E8053 PCI-E Gigabit Ethernet Controller NVIDIA nForce Networking Controller - Packet Scheduler Miniport Network Card MAC Address: 82:72:93:0C:BA:6E VIA OHCI Compliant IEEE 1394 Host Controller American Power Conversion USB UPS [Joystick] USB Human Interface Device [Joystick] NVIDIA Network Bus Enumerator PCI Class USB 2.0 Enhanced Host Controller Standard OpenHCD USB Host Controller USB Root Hub Cheers Peter
Please, attach YaST logs. http://www.opensuse.org/index.php/Bug_Reporting_FAQ#YaST Thanks
The logs should be sifficient (they also contain grub.conf). Torsten, any idea about error 22 (duplicate of any known bug)?
Nope, Jiri, can't recall anything like this. Have a close look at the partition table, once Peter provides it. I assume there's something fishy about it, if it's not really broken somehow. We should also keep in mind the letters "GPT", just in case.
Created attachment 48673 [details] YaST2 log files This yast log file is from a fresh install. The grub error 22 appeared once again.
Thanks for adding the info - please next time use the button below ("This comment provides the needed information.") so that it's not anymore in state NEEDINFO and the engineer can start working on it...
Hmm, the GRUB installation looks fine. Torsten, any idea whjat could be wrong? grub> setup --stage2=/boot/grub/stage2 (hd0) (hd1,7) Checking if "/boot/grub/stage1" exists... yes Checking if "/boot/grub/stage2" exists... yes Checking if "/boot/grub/reiserfs_stage1_5" exists... yes Running "embed /boot/grub/reiserfs_stage1_5 (hd0)"... 18 sectors are embedded. succeeded Running "install --stage2=/boot/grub/stage2 /boot/grub/stage1 d (hd0) (hd0) 1+18 p (hd1,7)/boot/grub/stage2 /boot/grub/menu.lst"... succeeded Done.
I guess /dev/hdb had invalid C/H/S settings when the partitions were created. Why is the boot partition not primary? BTW, giving the whole first disk to a wannabe-OS that doesn't know of any other systems is asking for trouble IMNSHO, especially when there's so much disk space in the system there should be better ways. Those yast logs are pretty unreadable; is there a way to get "fdisk -lu" output somehow? And device.map? It's probably in there somewhere, but totally unreadable outside yast maintainers :-( I would check that /dev/hdb corresponds to BIOS device 0x81, if this was intended, and further that /sys/firmware/edd/int13_dev80/legacy_* match the fdisk / parted settings. BTW, the lilo.conf that made it boot could be of interest, but not as much as any of the above.
Don't understand the 1st paragraph... The detected device map was (according to the log) $["/dev/fd0":"(fd0)", "/dev/hda":"(hd0)", "/dev/hdb":"(hd1)", "/dev/hde":"(hd2)", "/dev/hdf":"(hd3)", "/dev/hdg":"(hd4)"]. This seems to be correct (provided it is intended that BIOS boots /dev/hda). Torsten, can I grab for you any other information from the log? Changing needinfo to reporter to provide the rest...
Right. output from `fdisk -l` and `fdisk -lu' would be nice, alternatively dumps of the 2 MBRs. As well as the contents of /sys/firmware/edd/int13_dev8?/legacy_*
No info provided after 2 weeks -- assuming fixed.