https://bugzilla.novell.com/show_bug.cgi?id=414468 User AmigaPhil@ping.be added comment https://bugzilla.novell.com/show_bug.cgi?id=414468#c1 --- Comment #1 from Philippe Duchenne 2008-08-05 09:17:49 MDT --- Tried to send a CC to the authors of the pcsc-acr38 driver RPM (David Corcoran and Edward Cheung), but their email were rejected from the CC: field of the Bugzilla form. -- 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=414468 Cyril Hrubis changed: What |Removed |Added ---------------------------------------------------------------------------- AssignedTo|bnc-team-screening@forge.provo.novell.com |sbrabec@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.

https://bugzilla.novell.com/show_bug.cgi?id=414468 User AmigaPhil@ping.be added comment https://bugzilla.novell.com/show_bug.cgi?id=414468#c4 --- Comment #4 from Philippe Duchenne 2008-08-09 14:39:24 MDT --- Searching for clues here and there, I noticed in bug #413282 (about the pcsc-cyberjack driver) that the IFD Handler loaded is version 3.0, whereas with this pcsc-acr38 driver, the IFD Handler version is 2.0. Could it be that the ACR38 is using an old, outdated version of the IFD handler ? To do some more tests, I have edited /usr/lib/readers/ACR38UDriver.bundle/Contents/Info.plist to add: <key>ifdLogLevel</key> <string>0x000F</string> and restarted again the pcscd daemon in the foreground: Linux1:/home/amigaphil # pcscd --debug -a -f 00000000 pcscdaemon.c:275:main() pcscd set to foreground with debug send to stderr 00017571 pcscdaemon.c:493:main() pcsc-lite 1.4.100 daemon ready. 00156066 hotplug_libhal.c:304:get_driver() Looking a driver for VID: 0x072F, PID: 0x9000 00000104 hotplug_libhal.c:341:HPAddDevice() Adding USB device: usb_device_72f_9000_noserial_if0 01002235 readerfactory.c:1130:RFInitializeReader() Attempting startup of ACS ACR38U 00 00 using /usr/lib/readers/ACR38UDriver.bundle/Contents/Linux/ACR38UDriver.so 00001203 readerfactory.c:963:RFBindFunctions() Loading IFD Handler 2.0 00016756 hotplug_libhal.c:304:get_driver() Looking a driver for VID: 0x1D6B, PID: 0x0001 and insert a eid card: 49428655 eventhandler.c:437:EHStatusHandlerThread() Card inserted into ACS ACR38U 00 00 00000109 eventhandler.c:451:EHStatusHandlerThread() Error powering up card. (several attempt to re-insert the eid card, with always the same result.) with a bank card: 16689114 eventhandler.c:437:EHStatusHandlerThread() Card inserted into ACS ACR38U 00 00 00000089 Card ATR: 3B 67 00 00 00 00 00 00 00 90 00 -- 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=414468 User AmigaPhil@ping.be added comment https://bugzilla.novell.com/show_bug.cgi?id=414468#c6 --- Comment #6 from Philippe Duchenne 2008-08-09 19:21:15 MDT --- Here are some more info about my ACR38 reader: Model: ACR38U-CFC-ZET Serial number: 3010884261 Firmware: 1.00 (I'm not sure about this, I don't know how to check. Even the Diag tools from ACS does not report this on Windows XP) from hwinfo: 77: udi = '/org/freedesktop/Hal/devices/usb_device_72f_9000_noserial_if0' info.subsystem = 'usb' info.product = 'USB Interface' info.udi = '/org/freedesktop/Hal/devices/usb_device_72f_9000_noserial_if0' usb.vendor = 'Advanced Card Systems, Ltd' linux.hotplug_type = 2 (0x2) usb.bus_number = 1 (0x1) usb.product = 'USB Interface' linux.subsystem = 'usb' usb.linux.sysfs_path = '/sys/devices/pci0000:00/0000:00:07.2/usb1/1-2/1-2:1.0' usb.device_revision_bcd = 256 (0x100) usb.configuration_value = 1 (0x1) usb.max_power = 100 (0x64) info.category = 'smart_card_reader' usb.num_configurations = 1 (0x1) usb.num_ports = 0 (0x0) usb.num_interfaces = 1 (0x1) usb.linux.device_number = 2 (0x2) usb.device_class = 0 (0x0) usb.device_subclass = 0 (0x0) usb.speed = 12.0000 usb.device_protocol = 0 (0x0) usb.version = 1.10000 usb.vendor_id = 1839 (0x72f) usb.is_self_powered = false usb.product_id = 36864 (0x9000) usb.can_wake_up = false usb.interface.number = 0 (0x0) usb.interface.class = 0 (0x0) usb.interface.subclass = 0 (0x0) linux.sysfs_path = '/sys/devices/pci0000:00/0000:00:07.2/usb1/1-2/1-2:1.0' usb.interface.protocol = 0 (0x0) info.parent = '/org/freedesktop/Hal/devices/usb_device_72f_9000_noserial' info.addons = { 'hald-addon-openct' } 78: udi = '/org/freedesktop/Hal/devices/usb_device_72f_9000_noserial' info.subsystem = 'usb_device' usb_device.vendor = 'Advanced Card Systems, Ltd' usb_device.bus_number = 1 (0x1) linux.subsystem = 'usb' usb_device.product = 'ACR38 AC1038-based Smart Card Reader' info.product = 'ACR38 AC1038-based Smart Card Reader' linux.device_file = '/dev/bus/usb/001/002' usb_device.linux.sysfs_path = '/sys/devices/pci0000:00/0000:00:07.2/usb1/1-2' usb_device.device_revision_bcd = 256 (0x100) info.udi = '/org/freedesktop/Hal/devices/usb_device_72f_9000_noserial' usb_device.configuration_value = 1 (0x1) usb_device.max_power = 100 (0x64) usb_device.num_configurations = 1 (0x1) usb_device.num_ports = 0 (0x0) usb_device.num_interfaces = 1 (0x1) usb_device.linux.device_number = 2 (0x2) usb_device.device_class = 0 (0x0) linux.sysfs_path = '/sys/devices/pci0000:00/0000:00:07.2/usb1/1-2' usb_device.device_subclass = 0 (0x0) usb_device.speed = 12.0000 info.parent = '/org/freedesktop/Hal/devices/usb_device_1d6b_1_0000_00_07_2' usb_device.device_protocol = 0 (0x0) info.vendor = 'Advanced Card Systems, Ltd' usb_device.version = 1.10000 usb_device.vendor_id = 1839 (0x72f) usb_device.is_self_powered = false info.linux.driver = 'usb' usb_device.product_id = 36864 (0x9000) usb_device.can_wake_up = false linux.hotplug_type = 2 (0x2) -- 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=414468 User AmigaPhil@ping.be added comment https://bugzilla.novell.com/show_bug.cgi?id=414468#c8 --- Comment #8 from Philippe Duchenne 2008-08-11 12:46:01 MDT --- I got a reply from ACS' technical help desk today: "Thanks for your email. I am sorry that we do not support the modified version of our Linux driver. Please use our Linux driver from our web site http://www.acs.com.hk/drivers-manual.php?driver=ACR38 " :-( (Still stuck we are.) What is puzzling is that some people are able to use eID with the ACR38 reader on Linux from the software sources they have compiled (even on Suse 9.3). So maybe the bug is not in the ACR38 driver, but from the RPM build (???) -- 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=414468 User AmigaPhil@ping.be added comment https://bugzilla.novell.com/show_bug.cgi?id=414468#c10 --- Comment #10 from Philippe Duchenne 2008-08-11 17:59:55 MDT --- Here is another interesting thing I found on the web (about Debian distributions): [quote] I did just try another model of ACR38U reader and ... it works. So my ACR38U-CFC-BLI (S/N 4010301125) dont works with my Debian Lenny distro (but well with MSWindows). While an ACR38U-SPC-FDT (S/N 101021446) works well with my Debian Lenny distro. What is the difference between an ACR38U-CFC-BLI and an ACR38U-SPC-FDT readers ? [/quote] So, depending of the EXACT model, the ACR38 driver for Linux can work / almost / or not ???

From mine ("ACR38U-CFC-ZET"), I guess the "CFC" part means that the model is compliant to the European standards, and that "ZET" means this model was distributed (and modified ???) by the Zetes company (?).

While I can afford to buy another card reader (which one would you recommend ? I would like to buy one with the best Linux support, and that can read/write to as much smart card types as possible.), I find it unfair that the ACR38 still is promoted this way (at least here in Belgium, even by the fedict (Belgium government)), with such a poor support for other machines than the Microsoft (who ""legally"" (!) tie Windows sales to PC (!)) Windows ones. Should I try to contact Zetes too ? (I feel a bit alone complaining about the ACR38 reader. Am I really the only one left who can't use the beID soft on Linux ? Have I been cheated by the dealer - PhotoHall, Charleroi - about this smart card reader ?) -- 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=414468 User AmigaPhil@ping.be added comment https://bugzilla.novell.com/show_bug.cgi?id=414468#c12 --- Comment #12 from Philippe Duchenne 2008-09-12 14:56:28 MDT --- A new version (3.5) of the Beid middleware is now available from the Fedict web site: http://eid.belgium.be/fr/Comment_installer_l_eID/Linux/index.jsp (there is even a package for OpenSuse 11.0 !). Quickly browsing the archive content, it does not seem there is a new driver for the ACR38 card reader :-( I haven't tried to installed it yet and see if the bug is gone. Can I safely install that version from the Fedict site without messing up with Yast's list of installed software ? -- 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=414468 User AmigaPhil@ping.be added comment https://bugzilla.novell.com/show_bug.cgi?id=414468#c14 Philippe Duchenne changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEEDINFO |NEW Info Provider|AmigaPhil@ping.be | --- Comment #14 from Philippe Duchenne 2008-09-15 11:43:15 MDT --- Unfortunately, I don't have any other reader to test. Simon Bris also has experimented some problems reading a BeID card with an ACR38; but after some tries, he finally got it to read the card at least once (I never had any success so far). See here: http://devzilla.novell.com/eID-belgium/show_bug.cgi?id=9#c3 I don't know how to identify the reader externally. Mine is a USB model ACR38U-CFC-ZET, that's pretty all I can see on it. There was no doc and no driver supplied with it. In comment #10, I quoted someone else who got his ACR38U (model ACR38U-SPC-FDT) to work with his Debian distribution, while another model (ACR38U-CFC-BLI) didn't work. (Note that the model I use also have a -CFC- in name. A clue ?) I've just tried to uninstall the pcsc-acr38 driver, and install the pcsc-ccid one for a test : # rcpcscd stop # pcscd --debug -a -f 00000000 pcscdaemon.c:275:main() pcscd set to foreground with debug send to stderr 00002106 pcscdaemon.c:493:main() pcsc-lite 1.4.100 daemon ready. 00391546 hotplug_libhal.c:304:get_driver() Looking a driver for VID: 0x072F, PID: 0x9000 00134742 hotplug_libhal.c:304:get_driver() Looking a driver for VID: 0x1D6B, PID: 0x0001 It seems the pcsc-ccid driver is not used at all with this card reader. So I'm going to remove the pcsc-ccid driver and re-install the pcsc-acr38 one. -- 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=414468 User sbrabec@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=414468#c16 Stanislav Brabec changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |NEEDINFO Info Provider| |AmigaPhil@ping.be --- Comment #16 from Stanislav Brabec 2008-09-16 05:48:38 MDT --- My model is ACR38-SPC-R and it is a CCID version (driver pcsc-ccid). I have no Belgian eID card to test it. So I guess, that ACR38U-CFC-BLI is a USB-to-serial version (driver pcsc-acr38). lsusb should confirm it, but I think, that this problem only affects USB-to-serial version. Could you try binary-only driver version 1.8.0 from ACS web page? (You may need to install 32-bit pcsc-lite to test it on 64-bit machine.) -- 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=414468 User sbrabec@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=414468#c18 Stanislav Brabec changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |NEEDINFO Info Provider| |AmigaPhil@ping.be --- Comment #18 from Stanislav Brabec 2008-09-17 04:00:48 MDT --- I found a formal error in the code, which may cause random behavior of Adm_GetAcrStats() call in IFDHSetCapabilities(). Fix has been placed to security:chipcard repository in the openSUSE Build service. Could you try this one? (You should see this as a last changelog entry: Mon Sep 15 15:42:14 CEST 2008 - sbrabec@suse.cz - Fixed implicit declaration. There is a small chance, that it could help. If not, debugging could be more complicated. -- 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=414468 User flop.m@usa.net added comment https://bugzilla.novell.com/show_bug.cgi?id=414468#c20 --- Comment #20 from Issa Gorissen 2008-12-16 09:36:47 MST --- I have almost the same problem technically but definitely the same problem functionally (can't read my belgian eid). My reader is a Gemplus GemPC400 (PCMCIA) reader. openSUSE 11.0. Using pcsc-lite+ccid+acr38. When I insert the eid, pcscd prints the card ATR. When I try to use the beid-tool from eID-belgium package, it cannot power up the card and print its ATR. Another tool from the Belgium gov, beidgui, cannot power up the card either and for this case, pcscd prints the card is absent or mute! -- 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=414468 User sbrabec@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=414468#c22 Stanislav Brabec changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |NEEDINFO Info Provider| |AmigaPhil@ping.be --- Comment #22 from Stanislav Brabec 2008-12-17 08:14:36 MST --- Can other testers confirm this harware issue? Additionally, if you will be able to confirm the wrong state of reader or pcscd, please let me or upstream know. Maybe this issue trigger a bug in software or hardware. -- 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=414468 User AmigaPhil@ping.be added comment https://bugzilla.novell.com/show_bug.cgi?id=414468#c24 Philippe Duchenne changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEEDINFO |NEW Info Provider|AmigaPhil@ping.be | --- Comment #24 from Philippe Duchenne 2008-12-17 14:20:48 MST --- (In reply to comment #22 from Stanislav Brabec)

Can other testers confirm this harware issue?

Additionally, if you will be able to confirm the wrong state of reader or pcscd, please let me or upstream know. Maybe this issue trigger a bug in software or hardware.

Ok. I've just tried to reproduce Simon Brys' experiment (comment #21) here. 1) I plugged in the ACR38 reader and started the pcsc daemon in the foreground. # rcpcscd stop # pcscd --debug -a -f I tried about 20 times to insert the Eid card, but had not any single success ("eventhandler.c:465:EHStatusHandlerThread() Error powering up card."). 2) I stopped the daemon (Ctrl-C), and start it again. This time, the reader did no more detect a card when inserted. Here is the log of pcscd: 00000000 pcscdaemon.c:280:main() pcscd set to foreground with debug send to stderr 00002038 pcscdaemon.c:498:main() pcsc-lite 1.4.102 daemon ready. 00033556 hotplug_libhal.c:305:get_driver() Looking a driver for VID: 0x072F, PID: 0x9000 00000056 hotplug_libhal.c:342:HPAddDevice() Adding USB device: usb_device_72f_9000_noserial_if0 01002256 readerfactory.c:1082:RFInitializeReader() Attempting startup of ACS ACR38U 00 00 using /usr/lib/readers/ACR38UDriver.bundle/Contents/Linux/ACR38UDriver.so 00001118 readerfactory.c:915:RFBindFunctions() Loading IFD Handler 2.0 00000359 readerfactory.c:267:RFAddReader() Using the pcscd polling thread 50003952 ifdwrapper.c:485:IFDStatusICC() Card not transacted: 612 00000068 eventhandler.c:178:EHSpawnEventHandler() Initial Check Failed on ACS ACR38U 00 00 00000030 readerfactory.c:275:RFAddReader() ACS ACR38U init failed. 00000030 eventhandler.c:109:EHDestroyEventHandler() Thread never started (reader init failed?) 00000027 readerfactory.c:1133:RFUnInitializeReader() Attempting shutdown of ACS ACR38U 00 00. 00001842 readerfactory.c:994:RFUnloadReader() Unloading reader driver. 00000297 hotplug_libhal.c:395:HPAddDevice() Failed adding USB device: usb_device_72f_9000_noserial_if0 00173391 hotplug_libhal.c:305:get_driver() Looking a driver for VID: 0x1D6B, PID: 0x0001 3) Stopped the daemon again, unplugged the reader, replugged the reader, and restarted pcscd. Try to test the card as in 1), without success (card detected, but not powered up). 4) Unplugged and replugged the reader (without stopping the daemon), tested the card as in 1), and still failed. 5) Retried step 2) (stop/restart daemon) : failed with the same "Initial Check Failed on ACS ACR38U 00 00" error from pcscd. 6) Retried step 2) again, but waiting a bit longer before restarting the daemon. No changes (still failed as in 2) and 5) ). 7) Tried variations of 3) and 4), but still no good reading. Note: I haven't tried recently, but last time I tested the same ACR38 reader on a Windows XP laptop, the Eid card was properly read each time inserted. It seems that effectively, the ACR38 reader with CFC as a second part of its model number (like "ACR38U-CFC-ZET") - I think you can recognize it by its translucent blue plastic casing - is defective (bad slot design / alignment of the pins (?) ). HOWEVER, it also seems that there is something on the software side that contribute a workaround for this problem. Longer timeout delay (?) More retries (?) Better chip type detection (?) -- 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.