Doing a new install of 12.2 KDE x64. USB NTFS and ext2 hard drives are mounting at /media/volumename. But USB flash drives (fat32) are being mounted at /run/media/$USER/volumename.
Two questions please: What has changed since 11.3 that moves the mount points? What configuration do I modify to make the devices mount at /media (I’m still running desktops with 11.3 and SLED) without building fstab entries?
And things were never automaticaly mounted at /media, but always at a directory (created for the mount) inside /media (like your example /media/volumename). Logical, else it would be (more or less) impossible two have more then one device mounted at the same time.
I just connected an USB stick to a 12.2 system and opened it with the file manager (Dolphin in KDE) and it is now mounted at /media/343A-2B00. As expected. It is a VFAT file system. I do not have a FAT32 at hand.
I wonder if you are mixing openSUS and SLED behaviour. They could be different. As this is the openSUSE forums, most of us here can not check that.
The above is not a definite answer, but maybe it helps.
> Hm, it seems that the very existance of /run is rather new: ‘What’s
> this /run directory doing on my system and where does it come from?’
> (http://lists.fedoraproject.org/pipermail/devel/2011-March/150031.html)
> I do have a /run /on my 12.2, but there is no /run/media.
>
> And things were never automaticaly mounted at /media, but always at a
> directory (created for the mount) inside /media (like your example
> /media/volumename). Logical, else it would be (more or less) impossible
> two have more then one device mounted at the same time.
Apparently the intention is that media itself be a symlink to somewhere
under /run, which is a dynamic filesystem, in memory. I guess that the
intended changes have not been fully applied and then you get some
external media mounted at /run/something, and others at /media/something.
–
Cheers / Saludos,
Carlos E. R.
(from 11.4 x86_64 “Celadon” (Minas Tirith))
It looks as if it is creating confusion elsewhere: https://bbs.archlinux.org/viewtopic.php?id=140134
They blame Gnome, but the OP uses KDE. My guess is that it is a system feature (udev and udisks, and friends).
Strange that until now nobody reported problems here.
I guess partly becaue usage of spontanious connected devices is done by the end-user mostly using one of the applications offered by tthe Devices applet. This will hide the mount point, when you walk around using e.g. Dolphin, you do not need to know the mount point at all.
At least they seem to try to make a solution to the multi-user problem, by now using the “username” somewhere in the path (and hopefulle creating the correct access going with it). Thus more users can connect devices for their own usage and the devices still having the same volume name.
But it is of course still a problem when a decive is connected and the device notifier on all the users deskjtops pop-up. The first user opening it is then the owner
No, that is modern behavior of systemd + udisks2 (may be + systemd-logind) + whatever user agent is used by DE. Intention is that drives mounted “on user request” are mounted under /run/media/${user_who_requested_mount}/…
On 2012-11-23 13:26, hcvv wrote:
> But it is of course still a problem when a decive is connected and the
> device notifier on all the users deskjtops pop-up. The first user
> opening it is then the owner
In theory it is the active one in the machine.
–
Cheers / Saludos,
Carlos E. R.
(from 11.4 x86_64 “Celadon” (Minas Tirith))
Can you define “active one” (with respect to the theory you mention)
To illustrate this I will try to repeat a story that was posted and reposted here (although some time ago).
M. starts the system amd logs in. After a while he locks his screen and walks away.
His wife W. wants to use the system, sees hat it is locked and uses the appropiate button on the lock window to get a second GUI screen with a KDM login. She logs in with her username. After a while she walks away levaing her session open.
Daughter D enters having an USB stick she wants to look on. She finds the open session of her mother and being a good girl, she uses the Change user button from the Kmenu and logs in on third logical graphical screen with her username. She connect the USB stick.
Now on behalf of which “active user” is this device mounted by the system?
Now on behalf of which “active user” is this device mounted by the system?
Can you define “active one” (with respect to the theory you mention)
Among all sessions on a given seat exactly one has property “Active”. And if user belongs to this session, drive is mounted on his/her behalf.
How sessions manager defines active session is another matter. It is normally assumed that this is the session that currently has control over display/keyboard. On Linux console login session is normally associated with VT, so you switch session when you switch VT.
Is it clear if they are active or not?
Yes, it should be session that you currently see on your display.
And how can they manage that?
Ctrl-Alt-Fn
Change User in DE menu
systemd-loginctl activate (assuming enough privileges)
On 2012-11-23 15:56, hcvv wrote:
>
> robin_listas;2505961 Wrote:
> Can you define “active one” (with respect to the theory you mention)
>
> To illustrate this I will try to repeat a story that was posted and
> reposted here (although some time ago).
> M. starts the system amd logs in. After a while he locks his screen and
> walks away.
> His wife W. wants to use the system, sees hat it is locked and uses the
> appropiate button on the lock window to get a second GUI screen with a
> KDM login. She logs in with her username. After a while she walks away
> levaing her session open.
> Daughter D enters having an USB stick she wants to look on. She finds
> the open session of her mother and being a good girl, she uses the
> Change user button from the Kmenu and logs in on third logical graphical
> screen with her username. She connect the USB stick.
>
> Now on behalf of which “active user” is this device mounted by the
> system?
The daughter, that’s the only local active session
–
Cheers / Saludos,
Carlos E. R.
(from 11.4 x86_64 “Celadon” (Minas Tirith))
It would also mean that the situation I painted above (with M., W. and D.) is now a defined situation where D. wins (and that is logical and acceptable IMHO).
Be assured that it wasn’t the case when this situation was actual (in the days of HAL). D. could not access the stick. she did not even get a device notify pop-up (that was on another logical screen, probably M.'s).
I guess that remote logins also will never be owner of such an USB device.
Well, thinking over the whole thread, that is one thing I do not undersatnd: sometimes you see old functionality and sometimes new, all within the same system. I couldn’t reproduce it with 12.2, but now even you seem to have problems reproducing it. At the same time all the links we have above point to the fact that something is going on in this area.