Bluetooth applet (OBEX send/browse) does not work since Natty (after bluetoothd 4.84 exactly)

An attempt to send a file using "bluetooth-applet - menu - <devicename> - Send files..." opens an error dialog after about 25 seconds (see attachment).

Also internet via bluetooth using my mobile phone doesn't work.

Additional info:
When I try to send a file to my (Nokia Symbian) phone, its bluetooth symbol changes: it is set inside brackets "( )". And if I send a file from the phone to Ubuntu 11.04 while running /usr/share/doc/bluez/examples/monitor-bluetooth then the following line is printed to the terminal:

{Device.PropertyChanged} [/org/bluez/692/hci0/dev_00_1C_9A_5C_23_87] Connected = 1

So both devices take note of each other, but any further communication seems to fali. To me it looks like a policy issue. Adding my $USER to the group "bluetooth" or editing files in /etc/dbus.1/* did not help yet.

This is the only severe problem with Natty. Please help.

To add some info, with a 3.1.0 kernel, I have BT working on an old thinkpad with karmic and my 3 android phones. I can browse with obex no problems.
With the same 3.1.0 kernel on my now upgraded oneiric laptop (where things used to work with lucid and maverick), I am able to pair with my phones, I have a custom app that's able to talk bluetooth once with my phone, but after that, I see:

connect(4, {sa_family=AF_BLUETOOTH, sa_data="*\177\2037Z<\31\0\0\0\0\0\0\0"}, 10) = -1 ECONNREFUSED
(Connection refused)

And again, phone and laptop are paired, unpairing/repairing works but only helps my app working once, after which it fails (and it used to work just fine with older ubuntus, and still fine with karmic), and browsing with obex fails reliably (while it works with karmic too).
browse files with bluetooth-applet still gives me: Failed to mount OBEX volume: Connection refused for my 3 devices that are otherwise paired fine.

Same error logs with oneiric:
(bluetooth-applet:24586): Bluetooth-DEBUG: killswitch 6 is KILLSWITCH_STATE_UNBLOCKED
(bluetooth-applet:24586): Bluetooth-DEBUG: killswitches state KILLSWITCH_STATE_UNBLOCKED
Failed to mount OBEX volume: Connection refused

Nov 19 07:42:32 gandalfthegrey dbus[2780]: [system] Rejected send message, 4 matched rules; type="method_return", sender=":1.106" (uid=0 pid=17845 comm="/usr/sbin/bluetoothd ") interface="(unset)" member="(unset)" error name="(unset)" requested_reply="0" destination=":1.170" (uid=500 pid=24586 comm="bluetooth-applet -d ")

hcidump shows 'Connection refused - security block'

> ACL data: handle 12 flags 0x02 dlen 84
    L2CAP(d): cid 0x0040 len 80 [psm 1]
        SDP SSA Rsp: tid 0x1 len 0x4b
          count 72
          record #0
              aid 0x0000 (SrvRecHndl)
                 uint 0x10007
              aid 0x0001 (SrvClassIDList)
                 < uuid-128 00001106-0000-1000-8000-00805f9b34fb >
              aid 0x0004 (ProtocolDescList)
                 < < uuid-16 0x0100 (L2CAP) > <
                 uuid-16 0x0003 (RFCOMM) uint 0x1e > >
              aid 0x0005 (BrwGrpList)
                 < uuid-16 0x1002 (PubBrwsGrp) >
              aid 0x0100 (SrvName)
                 str "OBEX FTP"
          cont 00
< HCI Command: Authentication Requested (0x01|0x0011) plen 2
    handle 12
> HCI Event: Command Status (0x0f) plen 4
    Authentication Requested (0x01|0x0011) status 0x00 ncmd 1
> HCI Event: Link Key Request (0x17) plen 6
    bdaddr 00:23:D4:B8:96:46
< HCI Command: Link Key Request Reply (0x01|0x000b) plen 22
    bdaddr 00:23:D4:B8:96:46 key 2887E8B72B004EE9188C6EC889335A77
> HCI Event: Command Complete (0x0e) plen 10
    Link Key Request Reply (0x01|0x000b) ncmd 1
    status 0x00 bdaddr 00:23:D4:B8:96:46
> HCI Event: Auth Complete (0x06) plen 3
    status 0x00 handle 12
< ACL data: handle 12 flags 0x00 dlen 12
    L2CAP(s): Connect req: psm 3 scid 0x0041
> HCI Event: Number of Completed Packets (0x13) plen 5
    handle 12 packets 2
> ACL data: handle 12 flags 0x02 dlen 16
    L2CAP(s): Connect rsp: dcid 0x0000 scid 0x0041 result 3 status 0
      Connection refused - security block
< ACL data: handle 12 flags 0x00 dlen 12
    L2CAP(s): Disconn req: dcid 0x0040 scid 0x0040
> ACL data: handle 12 flags 0x02 dlen 12
    L2CAP(s): Disconn rsp: dcid 0x0040 scid 0x0040
> HCI Event: Number of Completed Packets (0x13) plen 5
    handle 12 packets 1
< HCI Command: Disconnect (0x01|0x0006) plen 3
    handle 12 reason 0x13
    Reason: Remote User Terminated Connection
> HCI Event: Command Status (0x0f) plen 4
    Disconnect (0x01|0x0006) status 0x00 ncmd 1
> HCI Event: Disconn Complete (0x05) plen 4
    status 0x00 handle 12 reason 0x16
    Reason: Connection Terminated by Local Host

@Marc: thanks for confirming!
It's still an issue for me in Natty too. It did not help adding me to group 'bluetooth' or tweaking '/etc/dbus-1/system.d/bluetooth.conf'. If I start "obex-data-server -nd" manually and try to browse my successfully paired phone then I see:

obex-data-server 0.4.6
Using Session bus
Parsed[0]: ftp <-- BROWSING MY NOKIA STARTS HERE
FTP uuid selected, first checking for Nokia OBEX PC Suite Services uuid
Connected SDP session
SDP search process
SDP search completed
Service Name: Nokia OBEX PC Suite Services
getting RFCOMM channel
Connect in progress <-- HERE IT PAUSES FOR 10 secs
**Warning: DBUS NAME REMOVED: :1.131
**Warning: Finalizing session
attempting to close session
session closed
Removing listened DBUS name :1.131 (object: /org/openobex/session0)
Removed from listened DBUS names list
^Cme was killed <-- HERE I PRESSED CTRL+C AFTER MINUTES
Disposing manager
Manager disposed at once
quitting main loop
Finalizing manager

I have similar problems with DUN for months, see bug #822311
However, it works after 5-10 attempts (I've written a script for it).
Should I add bluez to affected projects?
Are there any non-Nokia devices affected?

I was able to mount it using
$ mount -t fuse "obexfs#-b00:11:22:33:44:55" /media/phone
as mentioned on http://ru.gentoo-wiki.com/wiki/%D0%A0%D0%B0%D0%B1%D0%BE%D1%82%D0%B0_%D1%81_%D0%BC%D0%BE%D0%B1%D0%B8%D0%BB%D1%8C%D0%BD%D1%8B%D0%BC%D0%B8_%D1%82%D0%B5%D0%BB%D0%B5%D1%84%D0%BE%D0%BD%D0%B0%D0%BC%D0%B8_%D1%87%D0%B5%D1%80%D0%B5%D0%B7_bluetooth#.D0.9C.D0.BE.D0.BD.D1.82.D0.B8.D1.80.D0.BE.D0.B2.D0.B0.D0.BD.D0.B8.D0.B5_.D1.82.D0.B5.D0.BB.D0.B5.D1.84.D0.BE.D0.BD.D0.B0

@Загорія: This bug causes bluetooth file transfers to NEVER work since Natty. I think project bluez is affected here because the dbus error says that messages from "/usr/sbin/bluetoothd" are rejected, and with "dpkg -S /usr/sbin/bluetoothd" I found that bluetoothd comes from package bluez.

I own a Nokia, but I can lend an Android in a few days to test if the phone OS matters. It might be related to the service channel number - we will see.

Thank you for the hint with obexfs, but it does not work either. It just times out after a bit more than a minute and syslog says "hci0 ACL tx timeout".

Oliver: Never? I've got transfers (to phone) working after 3 or 4 retries.

This is not a nokia problem, bluetooth connections to my android phones (all 3 of them) stopped working after maverick for me.
My older laptop with an older ubuntu still talks fine to those phones.
Oneiric can pair to them, but after that, nothing.

@Marc: thanks, then I skip the test with Android. You seem to see the very same bug as I do. What bluetooth adapter is in your failing laptop? Mine has:

lsusb | grep -i blue
Bus 003 Device 005: ID 03f0:011d Hewlett-Packard Integrated Bluetooth Module

Bus 004 Device 012: ID 0a5c:2145 Broadcom Corp. Bluetooth with Enhanced Data Rate II

Сергій, thanks for the hint, obexftp still fails for me while, but obexfs actually works.
gandalfthegrey:~# obexftp -b 3c:5a:37:83:7f:2a -l
Browsing 3c:5a:37:83:7f:2a ...
Connecting...failed: connect
Tried to connect for 2ms
The user may have rejected the transfer: Connection refused
gandalfthegrey:~# mount -t fuse "obexfs#-b3C:5A:37:83:7F:2A" /mnt/mnt
gandalfthegrey:~#

bluetooth-applet browse files still does not work (it used to).

Ok, now I have something very interesting: I have a custom app that talks over bluetooth and it wasn't working.
It was getting:
> ACL data: handle 12 flags 0x02 dlen 16
    L2CAP(s): Connect rsp: dcid 0x0000 scid 0x0041 result 3 status 0 Connection refused - security block

Now, if I use mount -t fuse obexfs first, I can browse my files and my
other bluetooth app also starts working magically.
It seems that fuse manages to do auth better than other apps like obexftp
and my custom app, and once that auth has been done, other things just happen to work.

As soon as I unmount fuse/obexfs, then things stop working agagin.
Clearly, it does something better than the other apps and once it has authed with my phone, other things start working.

< HCI Command: Create Connection (0x01|0x0005) plen 13
    bdaddr 3C:5A:37:83:7F:2A ptype 0xcc18 rswitch 0x01 clkoffset 0x0000
    Packet type: DM1 DM3 DM5 DH1 DH3 DH5
> HCI Event: Command Status (0x0f) plen 4
    Create Connection (0x01|0x0005) status 0x00 ncmd 1
> HCI Event: Connect Complete (0x03) plen 11
    status 0x00 handle 12 bdaddr 3C:5A:37:83:7F:2A type ACL encrypt 0x00
< HCI Command: Read Remote Supported Features (0x01|0x001b) plen 2
    handle 12
> HCI Event: Command Status (0x0f) plen 4
    Read Remote Supported Features (0x01|0x001b) status 0x00 ncmd 1
> HCI Event: Read Remote Supported Features (0x0b) plen 11
    status 0x00 handle 12
    Features: 0xbf 0xfe 0x8f 0xfe 0x9b 0xff 0x79 0x83
< HCI Command: Read Remote Extended Features (0x01|0x001c) plen 3
    handle 12 page 1
> HCI Event: Command Status (0x0f) plen 4
    Read Remote Extended Features (0x01|0x001c) status 0x00 ncmd 1
> HCI Event: Max Slots Change (0x1b) plen 3
    handle 12 slots 5
> HCI Event: Read Remote Extended Features (0x23) plen 13
    status 0x00 handle 12 page 1 max 1
    Features: 0x01 0x00 0x00 0x00 0x00 0x00 0x00 0x00
< HCI Command: Remote Name Request (0x01|0x0019) plen 10
    bdaddr 3C:5A:37:83:7F:2A mode 2 clkoffset 0x0000
< ACL data: handle 12 flags 0x00 dlen 10
    L2CAP(s): Info req: type 2
> HCI Event: Command Status (0x0f) plen 4
    Remote Name Request (0x01|0x0019) status 0x00 ncmd 1
> ACL data: handle 12 flags 0x02 dlen 16
    L2CAP(s): Info rsp: type 2 result 0
      Extended feature mask 0x00b8
< ACL data: handle 12 flags 0x00 dlen 10
    L2CAP(s): Info req: type 3
> ACL data: handle 12 flags 0x02 dlen 20
    L2CAP(s): Info rsp: type 3 result 0
      Fixed channel list
< ACL data: handle 12 flags 0x00 dlen 12
    L2CAP(s): Connect req: psm 1 scid 0x0040
> ACL data: handle 12 flags 0x02 dlen 16
    L2CAP(s): Connect rsp: dcid 0x0040 scid 0x0040 result 1 status 0
      Connection pending - No futher information available
> ACL data: handle 12 flags 0x02 dlen 10
    L2CAP(s): I...

I checked again: obexfs does not work on my laptop with Natty.

But I have news: I removed the built-in Bluetooth adapter from my Compaq nx8220 (it's sticker says "Broadcom BCM92035NMD", in short a "BCM2035"). Then I plugged in an USB Bluetooth dongle from Cambridge Silicon Radio and guess what? Filetransfer, headset service and everything works in Natty!!

This means the Broadcom driver could cause this bug. So I attached the original Bluetooth adapter, and replaced all *.ko in /lib/modules/2.6.38-13-generic/kernel/drivers/bluetooth/ by those from Maverick - except btusb.ko, because of symbol conflicts. But this shows the same results as with an unmodified Natty. If the the driver is at fault, then it's btusb.ko.

Marc wrote in comment #5 that Kernel 3.1.0 with Karmic does works, but with Oneiric it does not. Next I will try Natty with other Kernels like 2.6.35 (probably next year...)

Oliver, first I do compile my own kernels, just to clear that up (not that it should make a difference).
Second, I'm pretty sure it's not a kernel issue since the same kernel does work with karmic and not with oneiric.
Then, since the fuse mount managed to use obex to talk to my phone, clearly the kernel bits work (especially since my non working app started working too at that time), it is userspace that broke in some way.

Marc, you're definitely right: userspace is involved here. To me it seems that newer kernels provide new functions, and newer apps were migrated to use these, but the Broadcom driver was not updated correctly. I think if this bug is solved simply by changing hardware (as I wrote above) then the kernel must be involved too.

My affected bluetooth is 0b05:b700

Googling your vendor:product and also "Broadcom Corp. Bluetooth with Enhanced Data Rate II" I found that you might have a BCM2045 chip. This chip is also in some Asus EeePCs, so it really should work with Ubuntu!

Yes, it's the EeePC 901. And it worked since my first install of 8.04 release.
But it is 'certified' only for 10.04: http://www.ubuntu.com/certification/catalog/component/usb:B700:0B05-BLUETOOTH

Sa problem for me on Ubuntu 11.10

auth.log

Dec 18 23:29:39 port-asus dbus[802]: [system] Rejected send message, 2 matched rules; type="method_return", sender=":1.10" (uid=0 pid=990 comm="/usr/sbin/bluetoothd ") interface="(unset)" member="(unset)" error name="(unset)" requested_reply="0" destination=":1.49" (uid=1000 pid=1588 comm="bluetooth-applet ")

Dongle:

Bus 002 Device 003: ID 0a12:0001 Cambridge Silicon Radio, Ltd Bluetooth Dongle (HCI mode)

Really 0a12:0001?? That is exactly the ID of my dongle that DOES work with my Natty! (see my comment #17). So I was wrong: although the hardware or the kernel/driver might still be involved, the major cause must be somewhere else. Please try the workaround of comment #10.

Another idea: I have reinstalled Natty from scratch (reformat disk, ect.). Is anybody affected who has upgraded from Maverick to Natty without reinstalling?

I installed warty or something and upgraded since then. I have never re-installed.
Mind you with ubuntu, you're usually going to end up with a better system if you wipe than if you upgrade, since they don't test upgrades that carefully and do not support 2 version jumps or partial upgrades.

I made a fresh Oneiric installation, no upgrade.

My firend have the same problem using a different dongle and his Oneiric is an upgrade from an old version.

For what it's worth, this bug seems to affect me too, running LinuxMint Debian squeeze, and trying to connect via bluetooth to a Nokia N900. If I run bluetooth-applet from the terminal, I get these messages:

quixote@lark-bunty:~$ bluetooth-applet
** Message: adding killswitch idx 0 state 1
** Message: adding killswitch idx 12 state 1
** Message: killswitch 0 is 1
** Message: killswitch 12 is 1
** Message: killswitches state 1
** (bluetooth-applet:7555): DEBUG: Unhandled UUID 00005005-0000-1000-8000-0002ee000001 (0x5005)
** (bluetooth-applet:7555): DEBUG: Unhandled UUID 00005601-0000-1000-8000-0002ee000001 (0x5601)
** Message: killswitch 0 is 1
** Message: killswitch 12 is 1
** Message: killswitches state 1
Failed to mount OBEX volume: Connection refused

I have news from the kernel mailing list linux-bluetooth! This issue might be caused by a known kernel bug. A patch already exists since August and has been committed to GIT today.

If anyone who is affected has the time and skills then please give it a try and give feedback! If you want to apply the patch yourself, its final v3 can be found here: (it did not change since then apart from line numbering)
http://thread.gmane.org/gmane.linux.bluez.kernel/15447

is that still an issue in the precise version?

Yes.
I just checked it by booting the iso of Precise alpha 2: pairing works, whether initiated by Ubuntu or by my phone. But data transmissions do not work at all to any of my Bluetooth devices.

Meanwhile I "solved" it in my Natty 11.04 by downgrading to bluetoothd 4.84! My Bluetooth headset, my keyboard and Wiimotes work perfectly with 4.84. Unfortunately another bug 543163 causes regular delays while browsing remote folders.

Downgrading can be done by installing the debian package bluez_4.84-0ubuntu1_i386.deb e.g. downloadable from https://launchpad.net/ubuntu/natty/i386/bluez/ or one can unpack this .deb manually and replace only the executable /usr/sbin/bluetoothd by that of 4.84.

Oliver,

I tried your work-around, but it did not work for me.

tau ~] sudo dpkg -i bluez_4.84-0ubuntu1_i386.deb
dpkg: warning: downgrading bluez from 4.96-0ubuntu4 to 4.84-0ubuntu1.

And it re-started bluetooth. So far so good.

The GUI remains completedly buggared as described before.

Something is different though. When I request to browse files
on one of my phones, an icon appears on the desktop. It offers me
to cancel the operation, and after a few seconds, dies:

Failed to mount OBEX volume: DBus error org.freedesktop.DBus.Error.NoReply: Did not receive a reply. Possible causes include: the remote application did not send a reply, the message bus security policy blocked the reply, the reply timeout expired, or the network connection was broken.

What a mess.

How can this still be broken after all this time, when
it was working flawlessly last year?

@Steve: I know what you mean! Bluetooth worked much better in 10.04. As we talked earlier, I really think you have multiple problems at once, since this bug here has nothing to do with GUI problems. You should split the mess in solvable parts and treat your GUI problems separately (e.g. search for "nomodeset").

The last idea I have: you could try to boot from a live CD, and if Bluetooth does not work immediately, then also try it with downgrading to 4.84. If you make it work with a live CD then it might be worth to reinstall your Ubuntu.

This bug is not a duplicate of bug 872044. (although its symptoms sound similar)

I tested the workaround that is mentioned there: it does not work for me. My bluetooth adapter is not even able to change the "sspmode": it answers "Can't set Simple Pairing mode on hci0 - Input/Output error (5)".

Furthermore this bug exists since Natty 11.04, whereas bug 872044 affects Gnome 3 (Oneiric 12.04 and later).

Thanks for pointing out this other bug. Although it's sad to see that Linux bluetooth support gets even worse.

This should now be fixed with a kernel patch. The 3.3 and newer contains the fix and for the stable
branches wait for those two commits to appear in the next release.

3.2-stable review patch. If anyone has any objections, please let me know.

------------------

From: Peter Hurley <email address hidden>

commit 18daf1644e634bae951a6e3d4d19d89170209762 upstream

Commit 330605423c fixed l2cap conn establishment for non-ssp remote
devices by not setting HCI_CONN_ENCRYPT_PEND every time conn security
is tested (which was always returning failure on any subsequent
security checks).

However, this broke l2cap conn establishment for ssp remote devices
when an ACL link was already established at SDP-level security. This
fix ensures that encryption must be pending whenever authentication
is also pending.

Signed-off-by: Peter Hurley <email address hidden>
Tested-by: Daniel Wagner <email address hidden>
Acked-by: Marcel Holtmann <email address hidden>
Signed-off-by: Johan Hedberg <email address hidden>

3.0-stable review patch. If anyone has any objections, please let me know.

------------------

From: Peter Hurley <email address hidden>

commit 18daf1644e634bae951a6e3d4d19d89170209762 upstream

Commit 330605423c fixed l2cap conn establishment for non-ssp remote
devices by not setting HCI_CONN_ENCRYPT_PEND every time conn security
is tested (which was always returning failure on any subsequent
security checks).

However, this broke l2cap conn establishment for ssp remote devices
when an ACL link was already established at SDP-level security. This
fix ensures that encryption must be pending whenever authentication
is also pending.

Signed-off-by: Peter Hurley <email address hidden>
Tested-by: Daniel Wagner <email address hidden>
Acked-by: Marcel Holtmann <email address hidden>
Signed-off-by: Johan Hedberg <email address hidden>

Indeed, that's fixed in a kernel patch in Oneiric and Precise; marking as a duplicate of bug 872044, since it is. The fact that the workaround doesn't work for some devices is normal given that all devices are different and some may not support changing the mode.

I checked if my bug is solved with 12.04 - unfortunately it is not. I tested with the daily iso from April 21th which has kernel 3.2.0-23-generic-pae. I have attached the output of hcidump, which looks pretty different than my initial dump on Natty. But the workaround (replacing /usr/sbin/bluetoothd 4.98 with 4.84) still works! (but 4.84 still has those sporadic 30sec delays)

I consider this a strong evidence that:
* This bug is not a duplicate of bug 872044: symptoms and workaround are different and this bug exists since Natty, 872044 since Oneiric (see comment 4 there). And the fix for 872044 is included in 3.2.0-19.30, but this bug is still exists with 3.2.0-23.
* This bug is not caused by a hardware flaw: it has a software-only workaround (see bug description). The same hardware works with Ubuntu Maverick and before, and even with WinXP.

@Mathieu: I could be wrong, so what makes you think that this could be a duplicate of 872044? And what exactly do you mean by "some may not support changing the mode"? What mode?

mode> was specific to sspmode; which was a suggested workaround. If that's not supported, fine. Some devices may not support this.

If you say it's different, fine, but right now I have no way to figure out what might be failing to work on your system. We definitely did fix the largest underlying issues with sending and browsing in Oneiric and Precise now, nothing excludes that some specific hardware can still have other issues.

Would you be able to report your bug to the developers mailing list for bluez (upstream)? See http://www.bluez.org/contact/

Not a gnome-bluetooth bug since downgrading just bluez reportedly fixes the issue.

I have the issue with natty and Nokia N9. I can see the device paired in bluetooth list but iam unable to browse n9 or send or receive files. Any help is appreciated.Thanks.

Does not work for me also.
Ubuntu 12.04 and N900.
Also unable to use phone's internet.

Hello,

Confirming this bug on 12.04 LTS AMD64 using kernel 3.2.0-27-generic. Downgrading the BLUEZ package to Oneric's (bluez_4.96-0ubuntu3_amd64.deb) solves the Bluetooth issue with send/recieve OBEX push.

- CH

@CH: thanks for your comment, but if 4.96 works for you then you found a different issue! This report is about the bug that occurs with all versions since 4.85. Please compare all symptoms you see with those in this report. If you find differences then you should open a new report and describe all your details. Also check the similar bugs 872044 and 822275.

@Mathieu: the upstream bluez developers are already informed, on August 3 2011 by Nikolay Kachanov and on December 23 2011 by me. Daniel Wagner recommended a kernel patch of Peter Hurley. I did not try his patch yet. But I guess it does not help since I still see this issue with the latest kernel in Quantal 12.10 alpha.

My bluez discussion: http://www.spinics.net/lists/linux-bluetooth/msg20155.html

Nokia 5800 (Symbian) same problem as well...

I have problems with pairing phones using the Bluetooth manager in Precise. Solution I found is outlined at:
http://ubuntuforums.org/showthread.php?t=1976888

It entails using Blueman to manage Bluetooth devices rather than the default manager.

I have been struggling with this on an Asus Eee 1005HA. Both regular utilities and blueman pair, but browsing (obex ftp) fail. They also fail in Mint 14 (based on quantal) as well as Precise 12.04 (launched as live USB). Interestingly, everything seems to work OK with an Asus USB dongle on three other machines (I could not sort out how to use 2 bluetooth devices on the 1005HA). I suspect the dongle has a different chip. If it is useful, contact me (nashjc at uottawa.ca) offlist and tell me what info to get and also the commands to do so.

!0.04 Lucid, also run as live USB) DOES work. This supports the suggestion in other bug posts (e.g., one on fedora bugzilla
https://bugzilla.redhat.com/show_bug.cgi?id=747575 that the issue is somewhere in the kernel. However, it does also seem to be hardware-dependent.

JN

For me, the problem was fixed in Ubuntu 12.04.

@J C & kvaml: Sine I (as reporter of this bug) think it has to do with bluetoothd and certain hardware: could you please share your specs? You can print it by executing the following 2 commands in a Terminal:

  bluetoothd --version
  lsusb | grep -i blue

@John: I tried your solution in both Linux Mint 14 and Ubuntu 13.04 beta. But with blueman it does not work either with the bluetooth hardware in my HP Laptop. Both Distros use bluetoothd 4.101 and my workaround to replace it with 4.84 fails meanwhile too (bluetoothd 4.84 is not able to connect dbus anymore).

I also have problems with browsing files on the phone. I think that it last worked flawlessly for me on 11.04 and after some updates on 12.04 (I use Mint Maya 13 which is based on Ubuntu 12.04) I have Dell Latitude 6500 and a small usb dongle on another laptop. Both have problems but everything(mouse, serial port ...) except obex works just fine.
I doubt if this bug will be fixed since it's active for so many years, I personally lost hope and hold on to Mint Maya.

Hi I looked at it again, this time in Ubuntu 12.10.

It's still broken.

HOWEVER

If "auto-authorize" is set on the phone, IT WORKS!!
The phones file directories open (in not-unreasonable time)

Otherwise, you go through the process of typing in a confirmation number
in the phone, then typing the same number into a dialog in Linux,
and NOTHING.

SO the remaining problem has to do with the authorization step.

Fix this, and we'll be HALFWAY THERE!!!
Because... BT file browsing is a SIGSEV on 64-bit machines. Made a separate bug for that.

Thank you for reporting this bug to Ubuntu. Ubuntu 12.10 (quantal) reached end-of-life on May 16, 2014. And Ubuntu 11.04 (natty) reached end-of-life on October 28, 2012.

See this document for currently supported Ubuntu releases:
https://wiki.ubuntu.com/Releases

Please upgrade to the latest version and re-test.

[Expired for bluez (Ubuntu) because there has been no activity for 60 days.]