From Bugzilla Helper: User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8) Gecko/20051111 Firefox/1.5 Description of problem: I'm trying to recover a dead FC4 install from an external USB2 drive. Anaconda, while detecting the USB device, fails to load the usb-stroage module, and there is no manual way of doing it, (Even in expert and or dd mode) More-ever, Anaconda sourly a tool that will enable user to manually select/start drivers during the rescue/install procedure. Version-Release number of selected component (if applicable): How reproducible: Always Steps to Reproduce: 1. Connect an external USB drive 2. Reboot using Fedora disk 1 3. Select "linux rescue" and/or "linux". Actual Results: External USB drive is not detected. usb-storage module not loaded and cannot be manually loaded. Expected Results: External USB drive should be detected during the "hardware-probe" stage. Additional info:
Thank you for the report. Moving this to devel since this isnt specific to FC4 as such.
*** Bug 174888 has been marked as a duplicate of this bug. ***
Opps... yet another duplicate: https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=174909 I had severe case of FF timing out. Guess it didn't :( Sorry...
*** Bug 174909 has been marked as a duplicate of this bug. ***
usb-storage is loaded, but we don't list USB storage devices with probe-harddrives as it gives identical results to using the installer (where we also don't support installing to USB hard drives)
OK. After looking into the modules.cgz file, I can see that usb-storage is indeed there. What I don't understand is, when "linux rescue" is invoked, why doesn't the external USB device get scanned? AFAIK (by lsmod) usb-storage isn't loaded (in rescue mode), and opening the initrd image, followed by the modules.cgz just to extract the usb-storage by hand, strikes me as odd... Why no enable manual loading of drivers (usb-storage in this case) during rescue mode? Cheers,
(Sorry for reopening the bug...)
The module should be loaded -- if you look in /tmp/anaconda.log and/or /tmp/syslog you should see any errors from it being loaded.
Created attachment 121865 [details] Anaconda log AFAICS usb-storage isn't loaded. (Generated by boot from DVD with "linux rescue" command line)
Created attachment 121866 [details] Linux rescue syslog (Generated by boot from x86-64 DVD with "linux rescue" command line)
Created attachment 121868 [details] Anaconda boot kernel log (Generated by boot from x86-64 DVD with "linux rescue" command line)
Hmm, it doesn't look like kudzu is finding anything which uses usb-storage and thus we're not loading it. What's the output of python -c 'import kudzu; print kudzu.probe(kudzu.CLASS_UNSPEC, kudzu.BUS_USB, kudzu.PROBE_ALL)'
OK. I cannot reboot my primary machine, so I switching to my secondary (i386) machine and used a 512MB disk-on-key. See attached files. I've copied the first batch of logs. Manually extracted and inserted the usb-storage.ko and sg.ko modules. Generated the second batch of logs.
Created attachment 121890 [details] Pre-manual insmod system log. (i386/USB disk-on-key)
Created attachment 121891 [details] Post-manual insmod system log. (i386/USB disk-on-key)
Created attachment 121892 [details] Pre-manual insmod dmesg. (i386/USB disk-on-key)
Created attachment 121893 [details] Post-manual insmod dmesg. (i386/USB disk-on-key)
Created attachment 121894 [details] Anaconda log. (i386/USB disk-on-key)
Created attachment 121895 [details] Pre-manual insmod python log. (i386/USB disk-on-key)
Created attachment 121896 [details] Post-manual insmod python log. (i386/USB disk-on-key) AFACS, Kudzu, even with usb-storage and sg, doesn't 'see' the USB disk on key.
This should be fixed now
Doesn't work. FC5 rescue (x86_64) Anaconda doesn't load the usb-storage module.
REOPENED status has been deprecated. ASSIGNED with keyword of Reopened is preferred.
Does this still happen with FC6?
No go. Anaconda doesn't detect the USB storage device; usb-storage isn't loaded. F7-test1-desktop
Happens to me too -- the problem is common between Anaconda and the installed system; on the latter, I'd have to modprobe usb_storage manually. Note that on FC6, both Anaconda and the installed system initializes usb_storage properly. Relevant lines from dmesg; it seems that libusual is indeed trying to load usb-storage, but could not find it (not in initrd?) -- right now I'm resorting to manually loading usb-storage usb 3-3.4: new high speed USB device using ehci_hcd and address 6 usb 3-3.4: configuration #1 chosen from 1 choice libusual: modprobe for usb-storage succeeded, but module is not present input: Western Digital External HDD as /class/input/input5 input: USB HID v1.11 Device [Western Digital External HDD] on usb-0000:00:13.2-3 .4 (Harald located the problem -- initrd does not contain usb_storage) https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=204396
I've hit the same problem with F8/x86_64 in rescue mode. I had to manually unpack the initrd.img and modules.cgz files in-order to unpack and load missing modules. It seems that if fail to connect the USB disk during the initialization, the usb-storage will be rmmod. Shouldn't usb-strage and sg loaded by default when rescue mode is selected? - Gilboa
Under rescue mode in Rawhide, booting with both a USB disk attached and booting with one attached after anaconda has run, it works as expected both times. The appropriate modules are loaded, the device nodes are created, and the filesystems are mountable. No need to do any initrd unpacking or module loading yourself.