Description of problem: $ cat /proc/scsi/scsi Attached devices: Vendor: AGFA Model: SnapScan Rev: 1.40 Type: Scanner ANSI SCSI revision: 02 $ sane-find-scanner # sane-find-scanner will now attempt to detect your scanner. If the # result is different from what you expected, first make sure your # scanner is powered up and properly connected to your computer. # No SCSI scanners found. If you expected something different, make sure that # you have loaded a kernel SCSI driver for your SCSI adapter. # Also you need support for SCSI Generic (sg) in your operating system. # If using Linux, try "modprobe sg". problem fixed if issuing, as root: # modprobe sg $ sane-find-scanner # sane-find-scanner will now attempt to detect your scanner. If the # result is different from what you expected, first make sure your # scanner is powered up and properly connected to your computer. found SCSI scanner "AGFA SnapScan 1.40" at /dev/sg3 # Your SCSI scanner was detected. It may or may not be supported by SANE. Try # scanimage -L and read the backend's manpage. And scanner is found and usable in gimp.
May be a duplicate of Bug 3698 But as this bug is related to SCSI rather than USB, maybe opening a new bug is correct.
CC: (none) => olivier_le_royTarget Milestone: --- => Mageia 3
CC: (none) => thierry.vignaudAssignee: bugsquad => mageia
Is it really a missing udev rule? Isn't it more that the sg module is simply not loaded automatically on boot?
On a second computer, with "Mageia 2" rather than "cauldron", the sg driver is not loaded at boot automatically. # lsmod | grep sg returns nothing after boot. sane used to recognize my SCSI scanner on my first computer with Mageia 2, without issuing manually "modprobe sg".
*** Bug 10377 has been marked as a duplicate of this bug. ***
CC: (none) => wrw105
I've got a second install of mga2 on a system with mga3. lsmod is showing sg as loaded at startup in mga2. It is not loaded at startup in mga3.
Is this still valid in Mageia 4 & 5?
Keywords: (none) => NEEDINFO
It is still valid in Mageia 4 & 5. I use to add the scanner manually, but I can't figure out if it is actually a missing udev rule, because there was no such rule in Mageia 1. For tests purposes, the machine with the SCSI scanner on which I can reproduce, I will only have it in 2 weeks time. Therefore I won't be able to test a fix before 2 weeks time.
Whiteboard: (none) => MGA4TOO MGA5TOOKeywords: NEEDINFO => (none)Target Milestone: Mageia 3 => ---
*** Bug 16306 has been marked as a duplicate of this bug. ***
CC: (none) => yann.cantin
See https://bugzilla.novell.com/show_bug.cgi?id=761109 Actual openSUSE fix is adding /usr/lib/modules-load.d/sg.conf in systemd package to autoload the sg module at boot time : systemd.spec, %install : # create %{_libexecdir}/modules-load.d mkdir -p %{buildroot}%{_libexecdir}/modules-load.d cat << EOF > %{buildroot}%{_libexecdir}/modules-load.d/sg.conf # load sg module at boot time sg EOF
Created attachment 6880 [details] systemd.spec (mga5) : add /usr/lib/modules-load.d/sg.conf
Hmm, looks like a bad way to fix it. The module should be loaded automatically when the h/w is detected. Using module-load.d/ is really a nasty hack... :(
Problem is : sg is not related to any hardware, so no kernel autoloading. Actual fix is adding sg to /etc/modprobe.preload or any other boot-time module loading conf file, same difference... Another fix (used before by opensuse) could be to add that to udev/rules.d/80-drivers.rules : SUBSYSTEM=="scsi", ENV{DEVTYPE}=="scsi_device", TEST!="[module/sg]", IMPORT{builtin}="kmod load sg" Pro : inserted when needed Con : false positive (but doesn't hurt), maybe a race condition until the first loading complete.
Hmm, I wonder then if it's something that should just be compiled into the kernel (i.e. not as a model)? I mean if the fix is to always load it, then why not just keep it loaded... Thomas, what do you think? What would you prefer? Bake it in or have a userspace tool always load the module?
CC: (none) => tmb
CCing new sane maintainer.
CC: (none) => lists.jjorge
This bug has slept for years, and SCSI has mostly disappeared... closing.
Resolution: (none) => WONTFIXStatus: NEW => RESOLVED