https://bugzilla.novell.com/show_bug.cgi?id=235059
Summary: /etc/resmgr.conf.d/99-avrdude_parport.conf is out of date Product: openSUSE 10.3 Version: unspecified Platform: i686 OS/Version: SuSE Other Status: NEW Severity: Normal Priority: P5 - None Component: Development AssignedTo: sndirsch@novell.com ReportedBy: rschiele@gmail.com QAContact: qa@suse.de
/etc/resmgr.conf.d/99-avrdude_parport.conf is out of date and only produces log messages like:
Jan 14 20:28:55 sighup resmgr[13786]: /etc/resmgr.conf.d/99-avrdude_parport.conf:1: command 'add' no longer supported Jan 14 20:28:55 sighup resmgr[13786]: /etc/resmgr.conf.d/99-avrdude_parport.conf:2: command 'add' no longer supported Jan 14 20:28:55 sighup resmgr[13786]: /etc/resmgr.conf.d/99-avrdude_parport.conf:3: command 'add' no longer supported Jan 14 20:28:55 sighup resmgr[13786]: /etc/resmgr.conf.d/99-avrdude_parport.conf:4: command 'add' no longer supported Jan 14 20:28:55 sighup resmgr[13786]: /etc/resmgr.conf.d/99-avrdude_parport.conf:5: command 'add' no longer supported Jan 14 20:28:55 sighup resmgr[13786]: /etc/resmgr.conf.d/99-avrdude_parport.conf:6: command 'add' no longer supported Jan 14 20:28:55 sighup resmgr[13786]: /etc/resmgr.conf.d/99-avrdude_parport.conf:7: command 'add' no longer supported Jan 14 20:28:55 sighup resmgr[13786]: /etc/resmgr.conf.d/99-avrdude_parport.conf:8: command 'add' no longer supported
https://bugzilla.novell.com/show_bug.cgi?id=235059
sndirsch@novell.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |jw@novell.com, lnussel@novell.com Status|NEW |NEEDINFO Info Provider| |lnussel@novell.com
------- Comment #1 from sndirsch@novell.com 2007-01-14 20:41 MST ------- /etc/resmgr.conf.d/99-avrdude_parport.conf: add /dev/parport0 desktop add /dev/parport1 desktop add /dev/parport2 desktop add /dev/parport3 desktop add /dev/parport4 desktop add /dev/parport5 desktop add /dev/parport6 desktop add /dev/parport7 desktop
Ludwig, "add" has been replaced with what?
BTW, /etc/resmgr.conf.d/99-uisp_parport.conf is identical. Actually 99-avrdude_parport.conf has been a copy of 99-uisp_parport.conf from the beginning.
https://bugzilla.novell.com/show_bug.cgi?id=235059
lnussel@novell.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|NEEDINFO |NEW Info Provider|lnussel@novell.com |
------- Comment #2 from lnussel@novell.com 2007-01-15 02:27 MST ------- There is no replacement. resmgr no longer knows about devices, hal has them instead. Granting access to all parallelports just because some random package is installed is broken anyways, please remove the resmgr config file.
https://bugzilla.novell.com/show_bug.cgi?id=235059
sndirsch@novell.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC|jw@novell.com |sndirsch@novell.com AssignedTo|sndirsch@novell.com |jw@novell.com
------- Comment #3 from sndirsch@novell.com 2007-01-15 02:45 MST -------
[...], please remove the resmgr config file.
done. Jürgen, your turn (uisp package).
https://bugzilla.novell.com/show_bug.cgi?id=235059
jw@novell.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |NEEDINFO Info Provider| |lnussel@novell.com
------- Comment #4 from jw@novell.com 2007-01-15 08:41 MST ------- this degrades us to manually chmodding the /dev/parport? Done. Ludwig, are there any examples how to use hal instead?
https://bugzilla.novell.com/show_bug.cgi?id=235059
lnussel@novell.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|NEEDINFO |NEW Info Provider|lnussel@novell.com |
------- Comment #5 from lnussel@novell.com 2007-01-15 08:53 MST ------- /usr/share/doc/packages/hal-resmgr/README
- hal needs to know about /dev/parport* first - I would advise against using some generic 'parport' class where anyone gets access to all parallel ports. Introduce a separate one for the specific purpose of granting access to the controller instead.
https://bugzilla.novell.com/show_bug.cgi?id=235059
jw@novell.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |bwalle@novell.com
------- Comment #6 from jw@novell.com 2007-01-15 12:33 MST ------- CCed Bernhard, tfla-01 has the same problem.
https://bugzilla.novell.com/show_bug.cgi?id=235059
------- Comment #7 from rschiele@gmail.com 2007-01-15 12:56 MST ------- Ludwig, I agree with you that allowing access to all parport devices by default is not a very good idea but
1. there should be a documented way (example config file) how a sysadmin can setup this _in an easy way_ if he likes to have it that way,
2. I can't see how you want to give access to the controller device directly --- contrary to USB devices from your example file these devices cannot be easily identified by probing for them, and
3. the documentation situation of all that hal/resmgr/udev is extremely bad, the way to "do things the right way" changes quite frequently in recent times without having appropriate migration documentation available --- for instance documentation for simple questions like "how can one grant access to a specific device in a persistent way" is just not available to the average sysadmin. --- Note that I don't claim it does not exist at all but it is (a) not in a place where the average user will expect to find it and (b) it's complexity level is not appropriate to answer these simple questions in an appropriate way. Actually I know a huge number of sysadmins that work around this hal/resmgr/udev stuff instead of making use of it and thus this is a regression for them compared to the old static device files.
https://bugzilla.novell.com/show_bug.cgi?id=235059
------- Comment #8 from lnussel@novell.com 2007-01-16 01:58 MST ------- Well, I don't like some changes either but I can't do much about it. The scanner example is an example for example. Instead of vendor and product id you in theory could match any device via info.udi which is supposed to be persistent. In practice it doesn't seem to work as I just found via trial&error.
If you have any *specific* docu enhancements just tell me. I usually know where to look at so I have a hard time figuring out what's missing and where.
https://bugzilla.novell.com/show_bug.cgi?id=235059
------- Comment #9 from rschiele@gmail.com 2007-01-16 02:25 MST ------- It's not that I am against the changes. --- Not at all.
The problem with the documentation in the hal/resmgr/udev packages are mainly that (if existent) they describe and give examples for the shiny new features but basically don't explain how to do the things that were possible before in an easy way.
The _specific_ and most annoying problem for most people I actually see in exactly the question Stefan had initially here in this thread: How does one change permissions for a specific device (named just by the device name and no additional magic) in a persistent way. A short simple step-by-step instruction would do as long as this is maintained over future changes.
The current situation is more like giving a fancy modern VCR set to some old people not used to technology. When they ask how to record the current TV show on channel 2 you answer that it is not required to just record the current TV show because they could do programming their VCR over the internet instead of just telling them how to record the current TV show.
In this thread you did exactly this: Instead of just answering Stefan's and Juergen's questions on how to change permissions for /dev/parportX you gave the advice not to do so. While giving this advice is fine it would be even better if you could just answer the question as well. The file you mentioned (/usr/share/doc/packages/hal-resmgr/README) does not really provide this information.
Maybe actually the problem exactly _is_ that the people maintaining these packages typically know where to find all the information and thus don't see the problem any more.
https://bugzilla.novell.com/show_bug.cgi?id=235059
------- Comment #10 from bwalle@novell.com 2007-01-16 02:41 MST ------- As far as I understand, there's no way to do this without modifying HAL.
https://bugzilla.novell.com/show_bug.cgi?id=235059
------- Comment #11 from lnussel@novell.com 2007-01-16 02:53 MST ------- Yes, for parport* hal needs to be patched so resmgr can do anything in the first place. Robert is right however. His request for matching a device name is probably so simple I didn't even think about it. Here's an example for a fictional device that is known to hal:
# mkdir /etc/hal/fdi/policy/20user # cat <<EOF > /etc/hal/fdi/policy/20user/50-missilelauncher.fdi <?xml version="1.0" encoding="ISO-8859-1"?> <deviceinfo version="0.2"> <device> <match key="linux.device_file" string="/dev/missilelauncher"> <merge key="resmgr.class" type="string">wmd</merge> </match> </device> </deviceinfo> EOF # cat <<EOF > /etc/resmgr.conf.d/50-wmd.conf ## define a new class 'wmd' class wmd
## grant access to class 'wmd' when access to 'desktop' is granted ## (ie everyone on a local console) class desktop includes wmd
## maybe you want to restrict access to 'wmd' to members of a group instead #allow wmd group=generals EOF # rcresmgr restart # rchal restart
https://bugzilla.novell.com/show_bug.cgi?id=235059
------- Comment #12 from rschiele@gmail.com 2007-01-16 03:34 MST ------- Yes! Exactly this is the information that people would expect to find in a README file in either hal, hal-resmgr, or resmgr.
And maybe packages like avrdude should reference this file in their documentation because these are the packages where this information is actually needed.
https://bugzilla.novell.com/show_bug.cgi?id=235059
------- Comment #13 from sndirsch@novell.com 2007-01-16 07:53 MST ------- It is known to hal after doing this one?
# mkdir /etc/hal/fdi/policy/20user # cat <<EOF > /etc/hal/fdi/policy/20user/50-missilelauncher.fdi <?xml version="1.0" encoding="ISO-8859-1"?> <deviceinfo version="0.2"> <device> <match key="linux.device_file" string="/dev/missilelauncher"> <merge key="resmgr.class" type="string">wmd</merge> </match> </device> </deviceinfo> EOF
Or is ther more required to make it known to hal?
https://bugzilla.novell.com/show_bug.cgi?id=235059
------- Comment #14 from jw@novell.com 2007-01-16 08:10 MST ------- Created an attachment (id=113138) --> (https://bugzilla.novell.com/attachment.cgi?id=113138&action=view) hal patch to make parport devices knwon.
Ludwig answered that on research. Thanks! I'll copy here: Attached PoC patch makes all available devices known to hal with at least basic information.
For 10.2 you could document a workaround like this:
# hal-device -a /org/freedesktop/Hal/devices/computer_ppdev << EOF linux.device_file = '/dev/parport0' (string) linux.subsystem = 'ppdev' (string) info.product = 'ppdev device /dev/parport0' (string) linux.hotplug_type = 2 (0x2) (int) linux.sysfs_path = '/sys/devices/virtual/ppdev/parport0' (string) info.parent = '/org/freedesktop/Hal/devices/computer' (string) EOF
https://bugzilla.novell.com/show_bug.cgi?id=235059
jw@novell.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |asvetter@cip.physik.uni-wuerzburg.de
------- Comment #15 from jw@novell.com 2007-01-19 09:48 MST ------- *** Bug 129340 has been marked as a duplicate of this bug. ***
https://bugzilla.novell.com/show_bug.cgi?id=235059
sndirsch@novell.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |dkukawka@novell.com
------- Comment #16 from sndirsch@novell.com 2007-02-13 10:43 MST ------- So how should we proceed here? In case we'll need to patch hal. Who will take care of this? Jürgen, Ludwig, Danny?
https://bugzilla.novell.com/show_bug.cgi?id=235059
sndirsch@novell.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |eich@novell.com
------- Comment #17 from sndirsch@novell.com 2007-05-12 04:28 MST ------- Egbert, JFYI. Since Matthias or me is in Cc of this bugreport or the reported itself, it might be interesting for you as well.
https://bugzilla.novell.com/show_bug.cgi?id=235059
sndirsch@novell.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |mhopf@novell.com
------- Comment #18 from sndirsch@novell.com 2007-05-12 04:42 MST ------- JFYI, Matthias. This is a bugreport, which is assigned to Egbert/me or with Egbert/me in CC or reported by Egbert/me.
https://bugzilla.novell.com/show_bug.cgi?id=235059#c19
--- Comment #19 from Stefan Dirsch sndirsch@novell.com 2007-11-25 13:15:01 MST --- Can't we work on this for 10.3?
https://bugzilla.novell.com/show_bug.cgi?id=235059#c20
--- Comment #20 from Bernhard Walle bwalle@novell.com 2007-11-26 03:07:13 MST --- Really 10.3 or 11.0?
https://bugzilla.novell.com/show_bug.cgi?id=235059#c21
--- Comment #21 from Stefan Dirsch sndirsch@novell.com 2007-11-26 03:10:14 MST --- Sorry, I mean 11.0. :-)
https://bugzilla.novell.com/show_bug.cgi?id=235059#c22
--- Comment #22 from Stefan Dirsch sndirsch@novell.com 2007-11-27 10:21:53 MST --- Danny mentioned in Bug #231171, that he won't accept the patch of comment #14. So what can we do here now?
https://bugzilla.novell.com/show_bug.cgi?id=235059#c23
Danny Kukawka dkukawka@novell.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |NEEDINFO Info Provider| |jw@novell.com
--- Comment #23 from Danny Kukawka dkukawka@novell.com 2007-11-27 10:33:19 MST --- (In reply to comment #19 from Stefan Dirsch)
Can't we work on this for 10.3?
Yes, we can work on this, but not by adding simply every device in sysfs to HAL. We should add only what make sense and is usefull, and not every sysfs entry which in the most cases provide abosultely no info.
about the ppdev device: I have no machine with such a device. Please provide more info about the device (grep . /sys/devices/virtual/ppdev/* -R, HAL log (start hal with additional parameter --verbose=yes --use-syslog in the rc file))
https://bugzilla.novell.com/show_bug.cgi?id=235059#c24
--- Comment #24 from Ludwig Nussel lnussel@novell.com 2007-11-28 01:38:48 MST --- As long as a device has a device node that information is already useful. I admit that the dozens of virtual console devices should probably be excluded though.
https://bugzilla.novell.com/show_bug.cgi?id=235059
User jw@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=235059#c25
--- Comment #25 from Juergen Weigert jw@novell.com 2007-12-03 13:33:08 MST --- How would the patch look like, if 'the dozens of virtual console devices were excluded'? I am asking this silly question, As I cannot see where they are included.
Furthermore, I cannot see a way how hal could guess to which parport uisp hardware (or anything else, which is write-only) is attached.
https://bugzilla.novell.com/show_bug.cgi?id=235059
Stefan Dirsch sndirsch@novell.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Info Provider|jw@novell.com |dkukawka@novell.com
https://bugzilla.novell.com/show_bug.cgi?id=235059
User lnussel@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=235059#c26
Ludwig Nussel lnussel@novell.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Attachment #113138|0 |1 is obsolete| | Status|NEEDINFO |NEW Info Provider|dkukawka@novell.com |
--- Comment #26 from Ludwig Nussel lnussel@novell.com 2007-12-04 01:53:45 MST --- Created an attachment (id=185738) --> (https://bugzilla.novell.com/attachment.cgi?id=185738) updated patch
The patchs adds a generic handler for unknown devices so all tty devices are included implicitly. This updated patch adds an ugly exception.
hal cannot guess what's connected behind a parallel port. You'd have to provide an fdi file that matches e.g. for /dev/parport0 and then adds the necessary information.
https://bugzilla.novell.com/show_bug.cgi?id=235059
User sndirsch@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=235059#c27
Stefan Dirsch sndirsch@novell.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC|sndirsch@novell.com |jw@novell.com AssignedTo|jw@novell.com |sndirsch@novell.com Severity|Normal |Enhancement Status|NEW |ASSIGNED
--- Comment #27 from Stefan Dirsch sndirsch@novell.com 2008-01-30 10:33:16 MST --- Unfortunately Jürgen is no longer interested into this. He gave up after having seen no improvement at all. I'm tracking this now as enhancement.
https://bugzilla.novell.com/show_bug.cgi?id=235059
User jw@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=235059#c28
--- Comment #28 from Juergen Weigert jw@novell.com 2008-02-04 08:30:56 MST --- I had my dose of udev hacking this weekend. I just found out, all my rules were ignored due to bnc#358391 .
I am adding /etc/udev/rules.d/98-parport.rules containing the following line:
KERNEL=="parport*" GROUP="uucp", MODE="0660"
to make this work without reboot, run
rmmod ppdev; modprobe ppdev
https://bugzilla.novell.com/show_bug.cgi?id=235059
User lnussel@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=235059#c29
--- Comment #29 from Ludwig Nussel lnussel@novell.com 2008-02-04 08:51:34 MST --- Note that users are not supposed to be in group uucp. This group is used for by the uucp system only. Adding users to that group is a security problem as it opens race conditions in /var/lock. I'd recommend to only install the udev file as example, ie have the entry commented out and show a user instead of a group.
https://bugzilla.novell.com/show_bug.cgi?id=235059
User jw@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=235059#c30
--- Comment #30 from Juergen Weigert jw@novell.com 2008-02-04 09:19:09 MST --- I need to have multiple users accessing the device. My local homed username and my suse-nfs-homed username. would group lp be more suitable, or a 'dongle' group of our own choice?
If users need to join a group is an acceptable burden. Editing config files manually is not my understanding of user-experience.
https://bugzilla.novell.com/show_bug.cgi?id=235059
Stefan Dirsch sndirsch@novell.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|ASSIGNED |NEEDINFO Info Provider| |lnussel@novell.com
https://bugzilla.novell.com/show_bug.cgi?id=235059
User lnussel@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=235059#c31
Ludwig Nussel lnussel@novell.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|NEEDINFO |ASSIGNED Info Provider|lnussel@novell.com |
--- Comment #31 from Ludwig Nussel lnussel@novell.com 2008-02-06 01:12:43 MST --- members of the lp group will be able to mess with the printing system ie potentially get root access by modifying cups' config. If you want to go via group you have to add your own one. But then that's still retarded. The clean solution is via hal so ACLs can be used. Even then I'd in general vote against allowing users access to all parallel ports just because some package got installed (just like you don't start arbitrary deamons on package install). There could still be a printer connected to any of the available ports. So some sort of configuration would be needed always.
https://bugzilla.novell.com/show_bug.cgi?id=235059
User jw@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=235059#c32
--- Comment #32 from Juergen Weigert jw@novell.com 2008-02-06 04:43:40 MST --- <offtopic> Free resources should not be in jail.
IMHO, it should be the other way round. All unused devices should be world accessible, only those, where a connected hardware (printer/scanner/modem/...) is configured, should restrict access to the responsible driver. </offtopic>
HAL acls are fine with me, if you can make it work, without having the user to mess with config scripts.
https://bugzilla.novell.com/show_bug.cgi?id=235059
User jw@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=235059#c33
--- Comment #33 from Juergen Weigert jw@novell.com 2008-02-16 07:26:33 MST --- Group="dialout" is a requirement for kinternet. Nothing is wrong with that, afaik.
https://bugzilla.novell.com/show_bug.cgi?id=235059
Stefan Dirsch sndirsch@novell.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Summary|/etc/resmgr.conf.d/99-avrdude_parport.conf is |/dev/parport only accessible by superuser |out of date |
https://bugzilla.novell.com/show_bug.cgi?id=235059
User sndirsch@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=235059#c34
Stefan Dirsch sndirsch@novell.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |sndirsch@novell.com AssignedTo|sndirsch@novell.com |jw@novell.com Status|ASSIGNED |NEW
--- Comment #34 from Stefan Dirsch sndirsch@novell.com 2008-07-16 14:53:31 MDT --- Back to Jürgen. I no longer use a machine with parallel port.
https://bugzilla.novell.com/show_bug.cgi?id=235059
User jw@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=235059#c35
--- Comment #35 from Juergen Weigert jw@novell.com 2008-07-17 02:54:02 MDT --- Having seen Ludwigs explanation, I believe it should be (or become) possible to do with hal, what was done with resmgr before.
I cannot work on that. If there are no takers, I propose to turn this into a feature request.
https://bugzilla.novell.com/show_bug.cgi?id=235059
User jw@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=235059#c36
Juergen Weigert jw@novell.com changed:
What |Removed |Added ---------------------------------------------------------------------------- AssignedTo|jw@novell.com |bnc-team-screening@forge.provo.novell.com
--- Comment #36 from Juergen Weigert jw@novell.com 2008-07-17 02:56:06 MDT --- Back to screening team for resource allocation.
https://bugzilla.novell.com/show_bug.cgi?id=235059
Robert Vojcik rvojcik@novell.com changed:
What |Removed |Added ---------------------------------------------------------------------------- AssignedTo|bnc-team-screening@forge.provo.novell.com |lnussel@novell.com
https://bugzilla.novell.com/show_bug.cgi?id=235059
User lnussel@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=235059#c37
Ludwig Nussel lnussel@novell.com changed:
What |Removed |Added ---------------------------------------------------------------------------- AssignedTo|lnussel@novell.com |dkukawka@novell.com
--- Comment #37 from Ludwig Nussel lnussel@novell.com 2008-08-18 03:10:25 MDT --- pingpong alert.
Danny, this is how it looks in sysfs AFAICS, could you add this to hal?
# l -R /sys/class/ppdev/ /sys/class/ppdev/: insgesamt 0 drwxr-xr-x 2 root root 0 18. Aug 11:05 ./ drwxr-xr-x 39 root root 0 18. Aug 11:05 ../ lrwxrwxrwx 1 root root 0 18. Aug 11:05 parport0 -> ./../devices/pnp0/00:0a/ppdev/parport0/ # l -R /sys/devices/pnp0/00:0a/ppdev /sys/devices/pnp0/00:0a/ppdev: insgesamt 0 drwxr-xr-x 3 root root 0 18. Aug 11:05 ./ drwxr-xr-x 4 root root 0 18. Aug 11:05 ../ drwxr-xr-x 3 root root 0 18. Aug 11:05 parport0/
/sys/devices/pnp0/00:0a/ppdev/parport0: insgesamt 0 drwxr-xr-x 3 root root 0 18. Aug 11:05 ./ drwxr-xr-x 3 root root 0 18. Aug 11:05 ../ -r--r--r-- 1 root root 4096 18. Aug 11:05 dev lrwxrwxrwx 1 root root 0 18. Aug 11:05 device -> ../../../00:0a/ drwxr-xr-x 2 root root 0 18. Aug 11:05 power/ lrwxrwxrwx 1 root root 0 18. Aug 11:05 subsystem -> ./../../../../class/ppdev/ -rw-r--r-- 1 root root 4096 18. Aug 11:05 uevent
/sys/devices/pnp0/00:0a/ppdev/parport0/power: insgesamt 0 drwxr-xr-x 2 root root 0 18. Aug 11:05 ./ drwxr-xr-x 3 root root 0 18. Aug 11:05 ../ -rw-r--r-- 1 root root 4096 18. Aug 11:05 wakeup
https://bugzilla.novell.com/show_bug.cgi?id=235059
Danny Kukawka dkukawka@novell.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Priority|P5 - None |P4 - Low
https://bugzilla.novell.com/show_bug.cgi?id=235059
User dkukawka@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=235059#c38
Danny Kukawka dkukawka@novell.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |NEEDINFO Info Provider| |lnussel@novell.com
--- Comment #38 from Danny Kukawka dkukawka@novell.com 2008-10-17 03:04:17 MDT --- Has anyone a machine with /dev/parport I can login to for testing?
https://bugzilla.novell.com/show_bug.cgi?id=235059
User lnussel@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=235059#c39
Ludwig Nussel lnussel@novell.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|NEEDINFO |NEW Info Provider|lnussel@novell.com |
--- Comment #39 from Ludwig Nussel lnussel@novell.com 2008-10-17 03:11:21 MDT --- Any machine with parallel port I suppose :) Mine has one too
https://bugzilla.novell.com/show_bug.cgi?id=235059
User dkukawka@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=235059#c43
Danny Kukawka dkukawka@novell.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Component|Development |Development Product|openSUSE 10.3 |openSUSE 11.1 Version|unspecified |Factory
--- Comment #43 from Danny Kukawka dkukawka@novell.com 2008-10-17 09:00:26 MDT --- Fixed in the hal package in this repo: http://download.opensuse.org/repositories/home:/dkukawka:/hal-testing/openSU...
That's what HAL now has: udi = '/org/freedesktop/Hal/devices/ppdev_parport0' info.capabilities = {'ppdev'} (string list) info.category = 'ppdev' (string) info.parent = '/org/freedesktop/Hal/devices/pnp_PNP0401' (string) info.product = 'Parallel Port Device' (string) info.subsystem = 'ppdev' (string) info.udi = '/org/freedesktop/Hal/devices/ppdev_parport0' (string) linux.device_file = '/dev/parport0' (string) linux.hotplug_type = 2 (0x2) (int) linux.subsystem = 'ppdev' (string) linux.sysfs_path = '/sys/devices/pnp0/00:07/ppdev/parport0' (string)
https://bugzilla.novell.com/show_bug.cgi?id=235059
User dkukawka@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=235059#c44
--- Comment #44 from Danny Kukawka dkukawka@novell.com 2008-10-17 09:04:15 MDT --- @Ludwig: please tel me in which ACL class the device should get added or what the settings for the device should be.
https://bugzilla.novell.com/show_bug.cgi?id=235059
User lnussel@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=235059#c45
Ludwig Nussel lnussel@novell.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|NEEDINFO |NEW Info Provider|lnussel@novell.com |
--- Comment #45 from Ludwig Nussel lnussel@novell.com 2008-10-21 00:39:33 MDT --- I'd use a separate class, something like org.freedesktop.hal.device-access.parport
Default no access (auth_admin_keep_always)
https://bugzilla.novell.com/show_bug.cgi?id=235059
User dkukawka@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=235059#c46
Danny Kukawka dkukawka@novell.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |NEEDINFO Info Provider| |lnussel@novell.com
--- Comment #46 from Danny Kukawka dkukawka@novell.com 2008-10-21 02:27:41 MDT --- For active and inactive?
https://bugzilla.novell.com/show_bug.cgi?id=235059
User lnussel@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=235059#c47
Ludwig Nussel lnussel@novell.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|NEEDINFO |NEW Info Provider|lnussel@novell.com |
--- Comment #47 from Ludwig Nussel lnussel@novell.com 2008-10-21 02:34:28 MDT --- Yes. The parallelports are normally used by the printer daemon. Normal users shouldn't be able to interfere by default.
https://bugzilla.novell.com/show_bug.cgi?id=235059
User bwalle@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=235059#c48
--- Comment #48 from Bernhard Walle bwalle@novell.com 2008-10-21 02:41:36 MDT --- To clarify that a bit in general:
/dev/lp is used for printing.
/dev/parport is used for raw access to the parallel port. Some low-level hardware programming needs that, e.g. devices to program microcontrollers, devices to access LCD displays (HD 4480), logic analysers like (http://tfla-01.berlios.de) or even some old scanners. libieee1284 can be used to access /dev/parport, and in most cases it is.
A normal user (in terms of John Doe, not in terms of a Unix account) does not need /dev/parport. Except scanners, but since only few parallel port scanners ever worked with Linux and they might be more than 5 years (at least) old now, that are very few.
https://bugzilla.novell.com/show_bug.cgi?id=235059
User dkukawka@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=235059#c49
Danny Kukawka dkukawka@novell.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution| |FIXED
--- Comment #49 from Danny Kukawka dkukawka@novell.com 2008-10-21 09:50:37 MDT --- fixed in git. Will be part of CODE11
https://bugzilla.novell.com/show_bug.cgi?id=235059
Danny Kukawka dkukawka@novell.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status Whiteboard| |marked-ForNext11.0YOU:HAL
https://bugzilla.novell.com/show_bug.cgi?id=235059
User jw@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=235059#c50
--- Comment #50 from Juergen Weigert jw@novell.com 2009-02-27 07:21:23 MST --- A possible alternative to 'adding the user to group uucp' or 'running everything under sudo' might now be:
polkit-auth --obtain org.freedesktop.hal.device-access.parport
-
bugzilla_noreply＠novell.com