With the latest hal and dbus ivman apparently no longer is working properly. It also seems according to the ivman website that the software is no longer maintained. There is a replacement based on ivman called halevt, that is actively being developed. So far in my early tests it seems to work great, autoplugging my devices without any configuration modifications. Boolstuff is a library that halevt depends on. Reproducible: Always Steps to Reproduce: 1. emerge dev-libs/boolstuff 2. emerge sys-apps/halevt 3.
Created attachment 181190 [details] new ebuild for dev-libs/boolstuff depandancy for halevt
Created attachment 181192 [details] new ebuild for sys-apps/halevt halevt is a newer replacement for ivman
~amd64 can be added to the keywords...
Works great for me, replaced pretty advanced ivman/pmount config
Great! I was thinking that ivman neared the end of it's shelf life, even though it worked pretty decently. Aaanyway: Works great on amd64, should be keyworded. Needs an initscript, attached one. Either create the halevt user and groups or use halevt user and plugdev group (probably better). I run it as root user and plugdev group, so I don't need to use sudo everywhere.
Created attachment 186147 [details] init script for halevt
Doesn't work for me, I think, because my fstab uses usbdev devices, rather than /dev/sdfx devices. /dev/disk/by-id/usb-USB_Flash_Disk_90740C0025A6-0:0-part1 /mnt/usb/LyallsKey vfat defaults,noauto,user,gid=usbd,umask=0,umask=007,quiet 0 0 As such, whilst it successfully executes halevt: Running: halevt-mount -u /org/freedesktop/Hal/devices/volume_uuid_6E48_CF86 -o sync -m 002 The result is nothing, for me. However, if I use mount /mnt/usb/LyallsKey things work fine. Suggestions?
Ok, I have had a closer look. I am trying to use halevt 0.1.4 on my Gentoo x86 system with udev installed I am having difficulty configuring auto mounting. I do not use /dev/sdxx identifiers in my /etc/fstab, for example. Code: /dev/sdc1 /mnt/win_os ntfs-3g defaults,gid=ntfs,umask=0,umask=007,nls=utf8,silent 0 0 This is painful, particularly if you have USB disks, which are assigned different letters, depending upon when they are plugged in/powered up or if I change the order of disks in my BIOS. I have tried using disk ids, which has worked for quite a while, for me Code: /dev/disk/by-id/ata-WDC_WD10EAVS-00D7B1_WD-WCAU48133357-part1 /mnt/win_os ntfs-3g defaults,gid=ntfs,umask=0,umask=007,nls=utf8,silent 0 0 But, disk-id's don't work with halevt (at least not that I can fathom) as HAL has no attribute which matches this. So, since HAL has UUID available as a property, I figured I would simply setup my /etc/fstab as follows Code: /dev/disk/by-uuid/4C489837489821AE /mnt/win_os ntfs-3g defaults,gid=ntfs,umask=0,umask=007,nls=utf8,silent 0 0 where I obtain the UUID of each partition using the 'vol_id' command. Then, make my /etc/halevt/haletv.xml as follows... Code: ... <halevt:OnInit exec="halevt-mount -u $hal.udi$ -d /dev/disk/by-uuid/$hal.volume.uuid$"/> ... <halevt:Insertion exec="halevt-mount -u $hal.udi$ -d /dev/disk/by-uuid/$hal.volume.uuid$ -m 002"/> ... <halevt:Removal exec="halevt-umount -u $hal.udi$ -d /dev/disk/by-uuid/$hal.volume.uuid$; halevt-umount -s"/> The Problem. After brief inspection of the source code, I don't think this is working with halevt-mount, as it seems to search and match the device against the device listed in HAL, which will be /dev/sdc1. What I think we should do is take the supplied device and dereference it, if it is a symbolic link, until we end up at a real device file, then we do our search through HAL. I am investigating whether this is actually the case and whether the author of halevt accepts patches or has some form of source code repository.
(In reply to comment #6) > Created an attachment (id=186147) [edit] > init script for halevt > We probably don't need to specify a configuration file with -c, since *.xml files in /etc/halevt would be used anyway?
I will add this package on sunrise overlay at first and later on portage tree Thank you
Forget about sunrise, I will add this directly to portage when I am done testing it :)
Created attachment 192729 [details] New ebuild for boolstuff Improved ebuild :)
+*boolstuff-0.1.12 (30 May 2009) + + 30 May 2009; Markos Chandras <hwoarang@gentoo.org> + +boolstuff-0.1.12.ebuild, +metadata.xml: + Initial commit of boolstuff library. Partially fixes bug 257933. Thanks to + Mike Gualtieri <mike.gualtieri@gmail.com> for the initial ebuild. Giorgos + Kargiotakis <kargig@void.gr> will be proxy maintainer. + + +*halevt-0.1.4 (30 May 2009) + + 30 May 2009; Markos Chandras <hwoarang@gentoo.org> +halevt-0.1.4.ebuild, + +files/halevt, +metadata.xml: + Initial commit of halevt application. Fixes bug 257933. Thanks to + Mike Gualtieri <mike.gualtieri@gmail.com> for the initial ebuild. Giorgos + Kargiotakis <kargig@void.gr> will be proxy maintainer. + On tree now. Thank you all for your contribution :)
I had some trouble with mounting my devices (usb flash - vfat - and usb hdd - ntfs) with utf8-support. So, I rewrited halevt configs with pmount. <?xml version="1.0" encoding="UTF-8"?> <!DOCTYPE halevt:Configuration [ <!ENTITY MOUNTABLE ... > <!ENTITY MOUNT_COMMAND "pmount -t $hal.volume.fstype$ $hal.block.device$ '$hal.info.product$'"> ]> <halevt:Configuration version="0.1" xmlns:halevt="http://www.environnement.ens.fr/perso/dumas/halevt.html"> <halevt:Device match="&MOUNTABLE;"> <halevt:Insertion exec="&MOUNT_COMMAND;"/> </halevt:Device> ... <halevt:Device match="&MOUNTABLE;"> <halevt:OnInit exec="&MOUNT_COMMAND;"/> </halevt:Device>
