Bug 19140

Assigning to Bluez maintainer

CC: (none) => marja11
Assignee: bugsquad => shlomif

Running xfce4, all attempts to disable it have failed. Example:
  systemctl mask dbus-org.bluez.service

$ locate org.bluez.service
/etc/systemd/system/dbus-org.bluez.service
/usr/share/dbus-1/system-services/org.bluez.service

$ ll /etc/systemd/system/dbus-org.bluez.service
lrwxrwxrwx 1 root root 9 Oct 19 21:48 /etc/systemd/system/dbus-org.bluez.service -> /dev/null

$ cat /usr/share/dbus-1/system-services/org.bluez.service
[D-BUS Service]
Name=org.bluez
Exec=/bin/false
User=root
SystemdService=dbus-org.bluez.service

Source RPM: bluez? or some missing deps => bluez-5.42-1.mga6.src.rpm

Fixed in Cauldron!

CC: (none) => geiger.david68210

So fixed for mga6 too!

Assigning to QA,

Advisory:
========================

systemd obexd service cannot be correctly run because of a missing/wrong service name:

dbus[842]: [system] Activation via systemd failed for unit 'dbus-org.bluez.service': Unit dbus-org.bluez.service not found.

So this update fixes this issue.

========================

Packages in 6/core/updates_testing:
========================

bluez-5.45-2.3.mga6.x86_64.rpm
bluez-cups-5.45-2.3.mga6.x86_64.rpm
bluez-hid2hci-5.45-2.3.mga6.x86_64.rpm
lib64bluez3-5.45-2.3.mga6.x86_64.rpm
lib64bluez-devel-5.45-2.3.mga6.x86_64.rpm

bluez-5.45-2.3.mga6.i586.rpm
bluez-cups-5.45-2.3.mga6.i586.rpm
bluez-hid2hci-5.45-2.3.mga6.i586.rpm
libbluez3-5.45-2.3.mga6.i586.rpm
libbluez-devel-5.45-2.3.mga6.i586.rpm


Source RPM: 
========================
bluez-5.45-2.3.mga6.src.rpm

Assignee: shlomif => qa-bugs

bluez-5.45-2.2.mga6

# 



Sep 08 09:24:06 vega dbus[13329]: [system] Activating via systemd: service name='org.bluez' unit='dbus-org.bluez.service'
Sep 08 09:24:06 vega dbus[13329]: [system] Activation via systemd failed for unit 'dbus-org.bluez.service': Refusing activation, D-Bus is shutting down.
Jan 23 20:14:51 vega dbus[12795]: [system] Activating via systemd: service name='org.bluez' unit='dbus-org.bluez.service'
Mar 27 23:35:47 vega dbus[8540]: [system] Activating via systemd: service name='org.bluez' unit='dbus-org.bluez.service'


There were a couple of reboots there but it looks like the problem went away at some point, possibly after the last update.

Ran the current update:
- bluez-5.45-2.3.mga6.x86_64
- bluez-cups-5.45-2.3.mga6.x86_64
- bluez-hid2hci-5.45-2.3.mga6.x86_64
- lib64bluez-devel-5.45-2.3.mga6.x86_64
- lib64bluez3-5.45-2.3.mga6.x86_64

Askpass kicked in for rfkill setting state.

# systemctl start dbus-org.bluez.service
[root@vega lcl]# systemctl status dbus-org.bluez.service
● bluetooth.service - Bluetooth service
   Loaded: loaded (/usr/lib/systemd/system/bluetooth.service; enabled; vendor p
   Active: active (running) since Sun 2019-04-14 17:00:09 BST; 8min ago
     Docs: man:bluetoothd(8)
 Main PID: 21427 (bluetoothd)
   Status: "Running"
   CGroup: /system.slice/bluetooth.service
           └─21427 /usr/libexec/bluetooth/bluetoothd

Apr 14 17:01:48 vega bluetoothd[21427]: Authentication attempt without agent
Apr 14 17:01:48 vega bluetoothd[21427]: Access denied: org.bluez.Error.Rejected

bluetoothctl fails to find a default controller.

# systemctl mask dbus-org.bluez.service
Failed to mask unit: File /etc/systemd/system/dbus-org.bluez.service already exists and is a symlink to /usr/lib/systemd/system/bluetooth.service.

# systemctl disable dbus-org.bluez.service
Removed /etc/systemd/system/bluetooth.target.wants/bluetooth.service.
Removed /etc/systemd/system/dbus-org.bluez.service.
# systemctl enable dbus-org.bluez.service
Failed to enable unit: No such file or directory
# locate bluez.service
/data/system/etc/systemd/system/dbus-org.bluez.service
/data/system/system/systemd/system/dbus-org.bluez.service
/data/system/usr/share/dbus-1/system-services/org.bluez.service
/etc/systemd/system/dbus-org.bluez.service
/usr/share/dbus-1/system-services/org.bluez.service

$ blueman-report
Terminating blueman-adapters
Terminating blueman-applet
Terminating blueman-manager
Describe your next action (keep empty if done): 

blueman panel applet is now unresponsive.
blueman-assistant times out, so there is nothing there.
In system services (drakconf) bluetooth is registered as running and does not respond to stop and the gui freezes - killed it via window manager.

Downgraded the five packages and rebooted.
The journal reported
Apr 14 17:40:30 vega dbus[8466]: [system] Activation via systemd failed for unit 'dbus-org.bluez.service': Unit dbus-org.bluez.service not found.

This update has killed bluetooth on this machine and the subsequent downgrade failed to restore it.  Another dead kitten.

CC: (none) => tarazed25

Er, ignore those /data/.... references.  That was a backup of some kind I think, from long ago.

Reinstalled the update.
# updatedb
# locate org.bluez.service
/usr/share/dbus-1/system-services/org.bluez.service
# systemctl status bluez
Failed to dump process list, ignoring: Unit bluez.service not found.
● bluez.service
   Loaded: not-found (Reason: No such file or directory)
   Active: inactive (dead)
# systemctl status dbus-org.bluez.service
Failed to dump process list, ignoring: Unit dbus-org.bluez.service not found.
● dbus-org.bluez.service
   Loaded: not-found (Reason: No such file or directory)
   Active: inactive (dead)

And, just to confirm it:
# systemctl start dbus-org.bluez.service
Failed to start dbus-org.bluez.service: Unit dbus-org.bluez.service not found.

# ll /usr/share/dbus-1/system-services/org.bluez.service
-rw-r--r-- 1 root root 95 Mar 31 19:11 /usr/share/dbus-1/system-services/org.bluez.service

Adding an attachment for:
$ journalctl -xb | grep blue
in case it is of any interest.

Hardware matters:
$ lsusb
Bus 003 Device 002: ID 0a5c:21e8 Broadcom Corp. BCM20702A0 Bluetooth 4.0

$ rfkill list
1: hci0: Bluetooth
	Soft blocked: no
	Hard blocked: no

$ dmesg | grep blue
[    2.995901] bluetooth hci0: Direct firmware load for brcm/BCM20702A1-0a5c-21e8.hcd failed with error -2

# lsmod | grep btusb
btusb                  53248  0
btrtl                  16384  1 btusb
btbcm                  16384  1 btusb
btintel                16384  1 btusb
bluetooth             598016  5 btrtl,btintel,btbcm,btusb
usbcore               286720  10 xhci_hcd,ehci_pci,usbhid,em28xx_dvb,usb_storage,ehci_hcd,btusb,xhci_pci,uas,em28xx

So, where do we go from here?

Created attachment 10932 [details]
Extract from last boot journal relating to bluetooth activation

Mate desktop.
$ journalctl -xb | grep blue

Did check Cauldron on another machine and confirmed that obexd service comes into play.  Bluetooth working fine.

On the test machine (mga6) obex exists:
/usr/lib/systemd/user/dbus-org.bluez.obex.service

Restart.  Tried the update on another machine and bluetooth came up running.

I have no idea what obex is; it exists on this system but
# systemctl enable dbus-org.bluez.obex.service
Failed to enable unit: No such file or directory
$ ll /usr/lib/systemd/user/dbus-org.bluez.obex.service
lrwxrwxrwx 1 root root 45 Mar 31 19:12 /usr/lib/systemd/user/dbus-org.bluez.obex.service -> ../../../../usr/lib/systemd/user/obex.service

A search for obex includes these results:
/usr/lib/systemd/user/obex.service
/usr/libexec/obex-client
/usr/libexec/bluetooth/obexd
/usr/share/dbus-1/services/obex-client.service
/usr/share/dbus-1/services/org.bluez.obex.service
/usr/share/doc/obexd
/usr/share/doc/kernel-doc/ABI/testing/configfs-usb-gadget-obex
/usr/share/doc/obexd/AUTHORS
/usr/share/doc/obexd/README
/usr/share/doc/obexd/client-api.txt

Can the update be given an OK on the basis of bluetooth working normally?

Ah.  obex is to do with exchanging binary data between bluetooth devices, so it needs to work in mga6, but is it relevant to this bug?

(In reply to Len Lawrence from comment #10)
> Restart.  Tried the update on another machine and bluetooth came up running.

> Can the update be given an OK on the basis of bluetooth working normally?

I have no idea who feedback was directed at.

I have no bluetooth devices to test with. I only opened the bug because
I don't like my systems spitting out useless failures.

I attempted to do a clean mga6 xfce only install and found out xfce will not
come up. 

$ cat /etc/release
Mageia release 7 (Cauldron) for x86_64

# journalctl | grep bluez
#

# systemctl list-unit-files | grep blue
bluetooth.target           static  
 
The fact that Len Lawrence got it to work is outstanding and good enough for me to have him mark this bug resolved.

Thanks Bit Twister.  Setting it loose.

Keywords: feedback => (none)
Whiteboard: (none) => MGA6-64-OK

Then by all means, let's not keep it waiting. Validating. Advisory in Comment 4.

Keywords: 6sta2 => validated_update
CC: (none) => andrewsfarm, sysadmin-bugs

An update for this issue has been pushed to the Mageia Updates repository.

https://advisories.mageia.org/MGAA-2019-0030.html

Status: NEW => RESOLVED
Resolution: (none) => FIXED