Subject: hal: two desktop icons for the same crypto volume
Package: hal
Version: 0.5.9.1-4
Severity: normal

Hello,

As discussed on IRC, I have upgraded to lenny from etch and still the
"two icons for the same crypto volume" is present.

Bellow is the log of the discussion Sjoerd and myself had before the
release of Etch about this issue.



mar 22 12:58:07 *       eddyp thinks he is seeing a regression regarding 
#414417...
sjoerd, are you around?
mar 22 12:59:47 <eddyp> sjoerd: I have *two* icons on my desktop for an
encrypted partition I have... with 0.5.8.1-9 for hal, libhal1 and 
libhal-storage1
mar 22 13:00:05 <eddyp> sjoerd: I didn't see this bahaviour with 0.5.8.1-6.1
mar 22 13:00:39 <eddyp> sjoerd: should I try adding g-v-m 1.5.15-2 to the mix?
mar 22 13:01:21 <eddyp> sjoerd: this is an encrypted partition on the hard disk
mar 22 13:02:03 <sjoerd>        It's a buglet in the hal/gnome-vfs combination
mar 22 13:02:27 <sjoerd>        because we use pmount, the encrypted volume 
isn't
mounted where hal/gnome-vfs expects it.. And you get it twice
mar 22 13:02:39 <sjoerd>        If you use gnome-mount and the g-v-m from exp, 
it's
``fixed''
mar 22 13:03:10 <eddyp> sjoerd: yes, but this is a fixed drive, so it shouldn't
be on my desktop at all
mar 22 13:03:38 <sjoerd>        Yes and no.. If we remove all fixed drives from 
your
desktop, we'll get bugs about that :)
mar 22 13:03:46 <eddyp> sjoerd: also, this would mean that installs on crypto
would have icons on the desktop by default, wouldn't it?
mar 22 13:03:59 <sjoerd>        eddyp: where is it mounted ?
mar 22 13:04:05 <eddyp> sjoerd: /crypto
mar 22 13:04:08 <sjoerd>        Right
mar 22 13:04:19 <sjoerd>        No, gnome-vfs2 has a list of things not to put 
on your
desktop
mar 22 13:04:30 <sjoerd>        Which includes things like / and /usr
mar 22 13:04:34 <eddyp> and my configuration says "no fixed partitions should
have icons on desktop"
mar 22 13:04:49 <sjoerd>        so a system that has encrypted / won't show up 
on the
desktop
mar 22 13:05:18 <eddyp> sjoerd: shouldn't the removable bit be the one that is
asked instead?
mar 22 13:05:59 <eddyp> sjoerd: it would be more acceptable is there was only
one icon ..
mar 22 13:06:12 <eddyp> although still buggy
mar 22 13:06:22 <sjoerd>        I'm not sure what the encrypted volume bit is 
set too
mar 22 13:06:38 <sjoerd>        Oh i agree, that hal && gnome-vfs should be 
smarter
about what to show
mar 22 13:06:53 <sjoerd>        Some issues have apparently been fixed in newer
gnome-vfs..
mar 22 13:07:03 <sjoerd>        But hal is still somewhat stupid when it comes 
to
encrypted/lvm/raid etc
mar 22 13:07:16 <eddyp> sjoerd: but those won't be allowed in etch, would they?
vorlon ?
mar 22 13:07:31 <sjoerd>        Gnome 2.18 gnome-vfs, no not really
mar 22 13:07:57 <sjoerd>        Imho, it looks silly but it's not harmfull
mar 22 13:08:19 <eddyp> sjoerd: couldn't the patch be reworked soemhow to revert
part of the behaviour so that the icon don't appear
mar 22 13:08:20 <eddyp> ?
mar 22 13:09:11 <sjoerd>        Then we have to introduce a new patch into 
gnome-vfs
properly
mar 22 13:09:31 <eddyp> doesn't hal have the information about the device, if is
a fixed disk or a removable one?
mar 22 13:09:32 <sjoerd>        And i don't think this is enough reason to 
allow a
patched gnome-vfs into etch
mar 22 13:09:59 <sjoerd>        For an encrypted one, depends.. Only if it's a 
luks
volume with one parent
mar 22 13:10:14 <eddyp> sjoerd: this is the case here
mar 22 13:10:16 <sjoerd>        That is, you can retrieve it's unencrypted 
parent and
work from there
mar 22 13:11:27 <sjoerd>        But again, i don't really see this going into 
etch at
this point
mar 22 13:11:41 <eddyp> still, thinking about the parent... do you think there
could be a *sane* scenario when soembody has both one removable and one fixed
parent?
mar 22 13:12:15 <sjoerd>        Hal doesn't know how to handle multiple 
parents, so
your stuck there anyway
mar 22 13:13:33 <eddyp> just one icon is annyoing, having two is stupid ... 
arrrrgh!
mar 22 13:14:08 <sjoerd>        But yes, i agree with you that we should fix 
this.. But
again, imho it's not of a severity that would allow an updated version to go
into etch
mar 22 13:14:39 <eddyp> sjoerd: one word "regression"
mar 22 13:15:25 <eddyp> probably if a clear and short fix for this is provided
fast... the RMs will agree ....
mar 22 13:15:36 *       eddyp looks at vorlon
mar 22 13:15:57 <sjoerd>        If that's the case, i've got no problems with 
finding a
quick fix
mar 22 13:16:16 <sjoerd>        but i want to know before hand that it's not 
just a
waste of time
mar 22 13:16:41 <sjoerd>        Note, that it's just a fix for having two 
icons.. Not
removing them at all
mar 22 13:16:42 <eddyp> sjoerd: meaning that you'd work on it if RMs agree?
mar 22 13:17:20 <sjoerd>        If the RM's agree that it's enough of an issue 
that a
small patch would be allowed into etch then yes..
mar 22 13:17:29 <eddyp> sjoerd: vorlon: for the record, 414417 never hit me
mar 22 13:21:14 <eddyp> sjoerd: one more thing 0.5.8.1-7 - does that fix the
issues for removable storage that can't be unmounted?
mar 22 13:22:06 <eddyp> AFAIK, this was seen in Ubuntu, too
mar 22 13:22:22 <sjoerd>        It fixes the issue where hal freezes after you 
pull out
a mounted removable device
mar 22 13:23:02 <sjoerd>        And the fact that local devices aren't shown in
nautilus (but now too many are shown :()
mar 22 13:23:18 *       karora reads "hal freezes over when you pull out a 
mounted
removable device" !
mar 22 13:25:31 <sjoerd>        When gnome-vfs first got it's hal backend i 
argued with
upstream that we really should only show stuff on the desktop which is mounted
in /media... Then we wouldn't have these issues, but obviously they didn't agre



-- System Information:
Debian Release: lenny/sid
  APT prefers testing
  APT policy: (900, 'testing'), (500, 'stable'), (50, 'unstable')
Architecture: amd64 (x86_64)

Kernel: Linux 2.6.21-2-amd64 (SMP w/2 CPU cores)
Locale: LANG=ro_RO.UTF-8, LC_CTYPE=ro_RO.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/bash

Versions of packages hal depends on:
ii  adduser                   3.104          add and remove users and groups
ii  dbus                      1.1.1-3        simple interprocess messaging syst
ii  hal-info                  20070618-1     Hardware Abstraction Layer - fdi f
ii  libc6                     2.6.1-1        GNU C Library: Shared libraries
ii  libdbus-1-3               1.1.1-3        simple interprocess messaging syst
ii  libdbus-glib-1-2          0.74-1         simple interprocess messaging syst
ii  libexpat1                 1.95.8-4       XML parsing C library - runtime li
ii  libgcc1                   1:4.2.1-4      GCC support library
ii  libglib2.0-0              2.14.0-2       The GLib library of C routines
ii  libhal-storage1           0.5.9.1-4      Hardware Abstraction Layer - share
ii  libhal1                   0.5.9.1-4      Hardware Abstraction Layer - share
ii  libsmbios1                0.13.6-1       Provide access to (SM)BIOS informa
ii  libstdc++6                4.2.1-4        The GNU Standard C++ Library v3
ii  libusb-0.1-4              2:0.1.12-7     userspace USB programming library
ii  libvolume-id0             0.114-2        libvolume_id shared library
ii  lsb-base                  3.1-24         Linux Standard Base 3.1 init scrip
ii  pciutils                  1:2.2.4~pre4-1 Linux PCI Utilities
ii  udev                      0.114-2        /dev/ and hotplug management daemo
ii  usbutils                  0.72-8         Linux USB utilities

Versions of packages hal recommends:
ii  eject                         2.1.5-5    ejects CDs and operates CD-Changer

-- no debconf information

Attachment: signature.asc
Description: OpenPGP digital signature

Reply via email to