https://bugzilla.novell.com/show_bug.cgi?id=210172 ihno@novell.com changed: What |Removed |Added ---------------------------------------------------------------------------- AssignedTo|ihno@novell.com |jsmeix@novell.com -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.

https://bugzilla.novell.com/show_bug.cgi?id=210172 jsmeix@novell.com changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |NEEDINFO Info Provider| |mark_1@gmx.de ------- Comment #2 from jsmeix@novell.com 2007-01-17 01:40 MST ------- See comment #1. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.

https://bugzilla.novell.com/show_bug.cgi?id=210172 jsmeix@novell.com changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |NEEDINFO Info Provider| |mark_1@gmx.de ------- Comment #4 from jsmeix@novell.com 2007-01-22 02:16 MST ------- In attachment #114057 you will see (near the end) various USB related problems: --------------------------------------------------------------------------- [plustek] Available and supported devices: [plustek] Device: >libusb:001:010< - 0x04a9x0x220d .. [sanei_usb] sanei_usb_open: trying to open device `libusb:001:010' .. [sanei_usb] sanei_usb_open: opened usb device `libusb:001:010' (*dn=0) . . . A lot of succesful IO happens but then it starts to break down: . . . [sanei_usb] sanei_usb_read_bulk: trying to read 1 bytes [sanei_usb] sanei_usb_read_bulk: read returned EOF [plustek] UIO error .. USB error: error submitting URB: No such device [sanei_usb] sanei_usb_write_bulk: write failed: No such device .. USB error: could not clear/halt ep 3: No such device [plustek] Scanner not ready!!! .. FAIL Error: Error during device I/O .. USB error: failed to open /dev/bus/usb/001/010: No such file or directory --------------------------------------------------------------------------- First it works to do some IO via the USB device 001:010 but then it breaks down and the device has gone away. My Canon Lide 30 worked and works well with the plustek driver for all Suse Linux versions since 9.0 and for openSUSE 10.2. Therefore I assume it is no problem in the driver but a lower-level USB related problem. Did you use the older sane package on openSUSE 10.2 and then it works o.k. for you or do you mean that you used the older sane package on an older Suse Linux version? -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.

https://bugzilla.novell.com/show_bug.cgi?id=210172 jsmeix@novell.com changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEEDINFO |ASSIGNED Info Provider|mark_1@gmx.de | ------- Comment #6 from jsmeix@novell.com 2007-01-23 02:48 MST ------- So it simply hangs up somehow for you. In your debug log there are zillions of same messages like ------------------------------------------------------------------------- [sanei_usb] sanei_usb_read_bulk: trying to read 1 bytes [sanei_usb] 0000: 02 ................ [sanei_usb] sanei_usb_read_bulk: wanted 1 bytes, got 1 bytes [sanei_usb] sanei_usb_write_bulk: trying to write 4 bytes [sanei_usb] 0000: 01 07 00 01 ................ [sanei_usb] sanei_usb_write_bulk: wanted 4 bytes, wrote 4 bytes ------------------------------------------------------------------------- before finally the "read returned EOF" message which seems to be the moment when you unplugged it. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is.

https://bugzilla.novell.com/show_bug.cgi?id=210172#c8 Johannes Meixner changed: What |Removed |Added ---------------------------------------------------------------------------- Status|ASSIGNED |RESOLVED Resolution| |FIXED --- Comment #8 from Johannes Meixner 2007-08-08 01:17:30 MST --- Very many thanks for comment #7! (I would have searched for ages until I may have found this reason.) And this is the matching explanation on the sane-devel mailing list: http://lists.alioth.debian.org/pipermail/sane-devel/2007-July/019680.html In openSUSE 10.3 beta1 the issue should be solved in the sane-backends package via a udev rule file /etc/udev/rules.d/55-libsane.rules which disables the USB autosuspend feature as described in the above mail on the sane-devel mailing list. Therefore I close the report as fixed (in openSUSE 10.3). For versions before openSUSE 10.3 beta1 you may create a udev rule file like (see "man udev" for details) but I didn't test it: ----------------------------------------------------------------------- ACTION!="add", GOTO="libsane_rules_end" # Canon CanoScan N670U / N676U / LiDE20 SYSFS{idVendor}=="04a9", SYSFS{idProduct}=="220d", MODE="0664", GROUP="lp", ENV{libsane_matched}="yes" # Canon CanoScan LiDE 35 / LiDE 40 / LiDE 50 SYSFS{idVendor}=="04a9", SYSFS{idProduct}=="2213", MODE="0664", GROUP="lp", ENV{libsane_matched}="yes" # The following rule will disable USB autosuspend for the device ENV{libsane_matched}=="yes", RUN+="/bin/sh -c 'test -e /sys/$env{DEVPATH}/power/level && echo on > /sys/$env{DEVPATH}/power/level'" LABEL="libsane_rules_end" ----------------------------------------------------------------------- Long lines are wrapped here at " GROUP" and " RUN" and " &&" in the real file all settings must be on one line. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=210172#c10 --- Comment #10 from Andreas Schallenberg 2007-08-12 18:36:37 MST --- I tried to build a new kernel on my 10.2 system without USB_SUSPEND. When making cloneconfig I noticed that the current kernel was already built without USB_SUSPEND. However, I can reproduce the bug with the current kernel, too. So there seems to be another reason for this behaviour here. (The current kernel here is: Linux noname 2.6.18.8-0.5-default #1 SMP Fri Jun 22 12:17:53 UTC 2007 i686 athlon i386 GNU/Linux It is a SUSE package.) -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=210172#c12 --- Comment #12 from Andreas Schallenberg 2007-08-26 15:04:29 MST --- Made some additional tests: My computer has a nforce-2 mainboard and an extra USB PCI card with an ALi chipset. The bug is reproducible with USB ports driven by each of the chips. The I upgraded the test partition to 10.3 Beta-2: Bug still there. Then I looked aside and tested the current alpha release of Kubuntu (Gutsy). The scanner works there. Maybe the Ubuntu/Debian guys know more about this topic? (The kernel is named 2.6.22-generic) -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=210172#c14 --- Comment #14 from Mark Blank 2007-08-27 02:50:38 MST --- to comment #13: "...sane 1.0.18-34 (suse 10.2) or 1.0.17-18 (suse 10.1)." means these are the packages that came with suse 10.2 and 10.1 respectively. All testing with the desktop system of comment #13 was done using suse 10.2. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=210172#c16 --- Comment #16 from Andreas Schallenberg 2007-09-16 03:17:39 MST --- Just tested SUSE 10.3 Beta 3, scanner doesn't work. Added my scanner to /usr/src/linux/drivers/usb/core/quirks.c, (04a9:2213 Canon, Inc. LiDE 50/LiDE 35), recompiled it: works! I'm going to add this information to the sane bug #85488. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=210172#c18 Johannes Meixner changed: What |Removed |Added ---------------------------------------------------------------------------- Severity|Normal |Major --- Comment #18 from Johannes Meixner 2007-09-18 07:34:59 MST --- /etc/udev/rules.d/55-libsane.rules cannot work because it does .. RUN+="/bin/sh -c 'test -e /sys/$env{DEVPATH}/power/level \ && echo on > /sys/$env{DEVPATH}/power/level'" where $env{DEVPATH} evaluates to something like /devices/pci0000:00/0000:00:06.0/0000:02:00.1/usb2/2-1 But on my openSUSE 10.3 beta2 system running the 2.6.22.3-7-default kernel, there is no file /sys/devices/pci0000:00/0000:00:06.0/0000:02:00.1/usb2/2-1/power/level so that the 'test -e /sys/$env{DEVPATH}/power/level' fails. Obviously the above is not the right method to disable USB autosuspend. I asked a kernel guru what the right way is and made this a MAJOR bug because many scanners will not work well if USB autosuspend is not disabled. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=210172#c19 Johannes Meixner changed: What |Removed |Added ---------------------------------------------------------------------------- Severity|Major |Minor Summary|disable USB autosuspend doesn't work |disable USB autosuspend may not work in any case --- Comment #19 from Johannes Meixner 2007-09-18 08:28:45 MST --- The kernel guru replied: ----------------------------------------------------------------------- This is the correct way, precisely because it does nothing if the file does not exist. These attributes exist only if the kernel supports autosuspend. On my system: oneukum@oenone:~> zcat /proc/config.gz | grep SUSP CONFIG_SOFTWARE_SUSPEND=y CONFIG_SUSPEND_SMP=y # CONFIG_USB_SUSPEND is not set oneukum@oenone:~> uname -r 2.6.22.3-7-default ----------------------------------------------------------------------- It is exactly the same on my openSUSE 10.3 beta2 x86_64 test system: caps:~ # zcat /proc/config.gz | grep SUSP CONFIG_SOFTWARE_SUSPEND=y CONFIG_SUSPEND_SMP=y # CONFIG_USB_SUSPEND is not set caps:~ # uname -r 2.6.22.3-7-default I.e. our default kernel does not do any USB autosuspend. Therefore USB autosuspend it cannot be disabled and therefore I never had this problem on my systems. I reduce the bug severity to MINOR. Did you use a self-compiled kernel? What results the above commands on your systems? -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=210172#c21 Johannes Meixner changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEEDINFO |ASSIGNED Info Provider|andreas.schallenberg@eurobild.com | Summary|disable USB autosuspend may not work in any case|sometimes USB scanners jam since sane-backends > | |1.0.15 --- Comment #21 from Johannes Meixner 2007-09-19 00:35:26 MST --- It seems we are back at the beginning: - It is no USB autosuspend problem. (If yes it would have been strange why it can be solved by using an older SANE version.) - It happens only occasionally. - It happens for different SANE backends (see comment #5). - Neither I nor the people of the SANE project can reproduce it nor have an idea what the real cause might be. Currently I am at my wits end but I will keep the bug open for further comments and/or ideas. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=210172#c23 Fibonacci Prower changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |fibonacci.prower@gmail.com --- Comment #23 from Fibonacci Prower 2007-10-09 03:27:25 MST --- I can confirm this happens on both Ubuntu 7.04 and Fedora Core 5, using a CanoScan LiDE 20. On Fedora Core, it didn't happen "only occasionally" - the slider froze on init with the lamps on and making a screaming noise 100% of the time whenever I tried to scan anything, whether I used scanimage or the GUIs. Unplugging the scanner was the only way to, at least, end the noise. I reported it to sane-devel, but my message was ignored. On Ubuntu the success rate is about 50% (i.e. it jams every other scan) using scanimage only, since XSane simply refuses to scan anything (though the jamming is absent when using the GUIs). I have noticed there are certain operations which appear to drastically reduce the success rate: obtaining a preview-quality scan (never works), scanning with a resolution less than 100dpi (rarely works), scanning a large area (where "large" is close to or greater than 150×250 mm; the larger it is, the smaller are the chances of success), and scanning many images in rapid succession (will quickly fail). Not a single case of jamming was observed in either scenario when using the 1.0.15 backends. Furthermore, using an old Knoppix LiveCD which includes those very backends on the same machine in which Fedora and Ubuntu were tested, the scanner works flawlessly. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=210172#c25 --- Comment #25 from Johannes Meixner 2007-10-16 01:50:00 MST --- Mark, very many thanks for your detailed analysis! -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=210172#c27 --- Comment #27 from Johannes Meixner 2007-10-25 03:52:20 MST --- I provide for testing sane-backends CVS snapshot dated 2007 Oct 24 for openSUSE 10.2 and openSUSE 10.3 and for the openSUSE development version openSUSE "factory" for 32-bit Intel compatible (i586) and 64-bit AMD (x86_64) via the openSUSE build service at http://software.opensuse.org/download/home:/jsmeix/ The packages are * only for testing * without any guarantee or warranty * without any support As an extreme example, this means that if your complete computer center crashes because of these packages, it is only your problem. In particular note that this is a CVS snapshot which means that there could be whatever unstable or conflicting stuff in the code which may result arbitrary unexpected effects. Nevertheless, I am very interested in your feedback because I don't have a CanoScan LiDE 20 to test it on my own and my CanoScan LiDE 30 works and worked well all the time in particular also with this CVS snapshot. Of course I am also interested if there are other regressions with other scanners or other drivers but then please file separated bug reports and make it obvious which package, which package version, which hardware architecture and which openSUSE version you are talking about, e.g.: "Feedback regarding sane-backends-2007.10.24cvs-66.1.i586.rpm from http://software.opensuse.org/download/home:/jsmeix/ for openSUSE 10.3 used on 64-bit AMD hardware." Ideally provide also the "rpm -q --changelog sane-backends | head" output to make it unambiguous which exact package release you have. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=210172#c29 Johannes Meixner changed: What |Removed |Added ---------------------------------------------------------------------------- Status|ASSIGNED |NEEDINFO Info Provider| |andreas.schallenberg@eurobild.com --- Comment #29 from Johannes Meixner 2007-10-25 05:32:29 MST --- Regarding the CanoScan LiDE 50 / "genesys" driver: Andreas, does the sane-backends CVS snapshot (see comment #27) work for you? -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=210172 Johannes Meixner changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEEDINFO |ASSIGNED Info Provider|andreas.schallenberg@eurobild.com | -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=210172#c32 --- Comment #32 from Fibonacci Prower 2007-11-03 22:33:41 MST --- I can confirm that the CVS backends from 24th of October, 2007, work fine with my CanoScan LiDE 20. As a side note, they appear to work even *better* than the last known working backends - i.e. 1.0.15. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=210172#c34 --- Comment #34 from Heinz Altschul 2007-11-08 08:15:54 MST --- Regarding comment #33 changes in /etc/udev/rules.d/libsane.rules (line 617) #Epson PM-A900 | Epson Stylus Photo RX700 | Epson RX-700 J deleted the "#" sign at beginning of the line. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=210172#c36 --- Comment #36 from Heinz Altschul 2007-11-09 02:35:52 MST --- Regarding comment 35 You are right. J put the comment sign "#" in on the old place and the scanner still works fine. sorry! -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=210172#c37 --- Comment #37 from Johannes Meixner 2007-11-20 04:54:09 MST --- For the fun ;-) Right now my CanoScan LiDE 30 which worked well all the time (also with sane-backends-2007.10.24cvs) started to make ugly noise. I guess it is just the usual USB crap if it happens or not. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

https://bugzilla.novell.com/show_bug.cgi?id=210172 User jsmeix@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=210172#c39 --- Comment #39 from Johannes Meixner 2008-02-28 12:00:52 MST --- I provide for testing sane-backends version 1.0.19 for openSUSE 10.2 and openSUSE 10.3 and for the openSUSE development version openSUSE "factory" for 32-bit Intel compatible (i586) and 64-bit AMD (x86_64) via the openSUSE build service at http://software.opensuse.org/download/home:/jsmeix/ The packages are * only for testing * without any guarantee or warranty * without any support As an extreme example, this means that if your complete computer center crashes because of these packages, it is only your problem. Nevertheless, I am very interested in your feedback. In particular my CanoScan LiDE 30 works again well with sane-backends version 1.0.19 (see comment #37). Of course I am also interested if there are other issues with other scanners or other drivers but then please file separated bug reports and make it obvious which package, which package version, which hardware architecture and which openSUSE version you are talking about, e.g.: "Feedback regarding sane-backends-1.0.19-128.1.i586.rpm from http://software.opensuse.org/download/home:/jsmeix/ for openSUSE 10.3 used on 64-bit AMD hardware." Ideally provide also the "rpm -q --changelog sane-backends | head" output to make it unambiguous which exact package release you have. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.