Bugzilla – Bug 116954
Installer fails on SATA hard-drive
Last modified: 2006-09-12 13:11:41 UTC
When trying to install RC1 (or any of the beta series) on an HP DX5151 system (Athlon 64, ATI chipset) the installer hangs near the beginning when it hits "Searching for info file". The CD drive is IDE. The system still responds to Caps-lock on/off, alt-cntrl-del etc, but never progresses from there. Fedora Core 4, Ubuntu "Breezy Badger" and Centos 4.1 all install perfectly on the same system. I have attached the logs from a Centos 4.1 install on the same system, plus lspci output etc. I can't make out if this is related to this bug also : https://bugzilla.novell.com/show_bug.cgi?id=116141
Created attachment 49899 [details] Output from boot logs, lspci etc on a Centos 4.1 install on the same system
Sorry - typo, the system is an HP dx5150, not a dx5151.
kernel issue
I think this is related to another similar bug. Could you please check the other virtual consoles for oops messages?
There are no oops messages - as I said, the kernel seems to still be running fine - just the installer freezes looking for the file. Output on the other tty's is : insmod /modules/libata.ko insmod /modules/sata_sil.ko ... ata1: SATA mac UDMA/100 ....... ata1: dev 0 ATA, max UDMA/100, 78165360 sectors : lba48 No errors at all. I can switch to the open console and type commands quite happily.
Just wondering if there is any update on this? I have 300 machines that I have to either install SuSE 10 on or RHEL ... if I can't test SuSE then it's a RHEL switchover for us...
You could check where the installer hangs. Do echo 1 > /proc/sys/kernel/sysrq echo t > /proc/sysrq-trigger and then dmesg and check the backtraces of the yast2 processes and their children (if all fails use a digital camera to photograph them, otherwise serial console)
Hi, At the point the system fails there isn't a workable shell which I can use to echo things into /proc/ - dmesg isn't available either. Billy.
Hi, I've now tried SuSE 10 final and the system doesn't even get as far as it did before. If I go into text-mode for the installer then the last entries I see now are : [ For "Safe settings" & "ACPI Disabled" ] PCI : Probing PCI hardware PCI : Probing PCI hardware (bus 00) [ For regular install ] ACPI : Assume root bridge [\_SB_.PCI0] ACPI : Assume root bridge [\_SB_.PCI0] bus is 0 Billy.
Sorry - forgot to change the status. Updated.
Hi, I experienced he same problem trying to install Suse Linux 9.3 on a 32 bit PC. Installing onto an EIDE drive had worked fine. When I replaced the EIDE with a SATA drive the install faled as described. I have now moved to Suse 10.0. The installer runs until a reboot is required (end of CD 1). The reboot fails. Following messages appear during boot sequence SCSI device sda : drive cache : write back sda : sda1 sda2 sda3 (sda5 sda6) Attached SCSI disk sda at SCSI 0, channel 0, id 0, lun 0 (Loads more normal boot messages) Waiting for device /dev/sda5 to appear (Note - swap partition) ... resume device /dev/sda5 not found (ignoring) Loading reiserfs Waiting for device /dev/sda6 to appear (Note - root partition) ... not found -- exiting to /bin/sh ata1: command 0x25, timeout, stat 0x50, host_stat 0x4 (Note - above line repeats every few seconds). When I boot from a Gentoo CD (Linux Rescue Disk) I can mount /dev/sda6 without any problems. The filesystem is populated with all the CD 1 packages. Any help appreciated. Thanks - Steve
Does pci=noacpi help (especially for comment #9)?
Closing due to lack of info.