Arch Linux

Please read this before reporting a bug:
https://wiki.archlinux.org/title/Bug_reporting_guidelines

Do NOT report bugs when a package is just outdated, or it is in the AUR. Use the 'flag out of date' link on the package page, or the Mailing List.

REPEAT: Do NOT report bugs for outdated packages!
Tasklist

FS#50130 - [bluez] Bluetoothctl commands via pipe broken

Attached to Project: Arch Linux
Opened by Joseph Rice (Daviljoe193) - Friday, 22 July 2016, 21:18 GMT
Last edited by Andreas Radke (AndyRTR) - Sunday, 24 July 2016, 16:04 GMT
Task Type Bug Report
Category Upstream Bugs
Status Closed
Assigned To Andreas Radke (AndyRTR)
Architecture All
Severity Low
Priority Normal
Reported Version
Due in Version Undecided
Due Date Undecided
Percent Complete 100%
Votes 0
Private No

Details

Description:
When sending commands to Bluetoothctl via stdin from a pipe, it will print out 'Controller ??:??:??:??:??:?? not available' when trying to select a device, 'No default controller available' when trying to power the selected device, and 'Device ??:??:??:??:??:?? not available' when trying to connect or disconnect a device. This is most likely a bug because when Bluetoothctl is invoked normally, it can see all existing controllers and devices, and works just fine. I believe this started with the current version of bluez-utils.


Additional info:
* package version(s)
bluez-utils 5.41-1
* config and/or log files etc.
none

Steps to reproduce:
Just send commands to bluetoothctl via a pipe.
This task depends upon

Closed by  Andreas Radke (AndyRTR)
Sunday, 24 July 2016, 16:04 GMT
Reason for closing:  Fixed
Additional comments about closing:  5.41-2
Comment by Andreas Radke (AndyRTR) - Saturday, 23 July 2016, 06:22 GMT
Please downgrade the pkg to confirm the older version working. Couldn't this be some other update (shell, systemd,...)? Please check your pacman.log.
Comment by Joseph Rice (Daviljoe193) - Saturday, 23 July 2016, 06:34 GMT
After downgrading just bluez-utils from 5.41-1 to 5.40-1 and nothing else immediately fixes the problem, and upgrading back to 5.41-1 immediately reintroduces the problem. Here's the output of grep bluez < /var/log/pacman.log | grep '\-07'

[2016-07-14 21:39] [ALPM] upgraded bluez-qt (5.23.0-1 -> 5.24.0-1)
[2016-07-22 03:28] [ALPM] upgraded bluez (5.40-1 -> 5.41-1)
[2016-07-22 03:28] [ALPM] upgraded bluez-cups (5.40-1 -> 5.41-1)
[2016-07-22 03:28] [ALPM] upgraded bluez-hid2hci (5.40-1 -> 5.41-1)
[2016-07-22 03:28] [ALPM] upgraded bluez-libs (5.40-1 -> 5.41-1)
[2016-07-22 03:28] [ALPM] upgraded bluez-plugins (5.40-1 -> 5.41-1)
[2016-07-22 03:28] [ALPM] upgraded bluez-utils (5.40-1 -> 5.41-1)
[2016-07-23 02:26] [PACMAN] Running 'pacman -U bluez-utils-5.40-1-x86_64.pkg.tar.xz'
[2016-07-23 02:27] [PACMAN] Running 'pacman -U /var/cache/pacman/pkg/bluez-utils-5.40-1-x86_64.pkg.tar.xz'
[2016-07-23 02:27] [ALPM] downgraded bluez-utils (5.41-1 -> 5.40-1)
[2016-07-23 02:27] [PACMAN] Running 'pacman -U /var/cache/pacman/pkg/bluez-utils-5.41-1-x86_64.pkg.tar.xz'
[2016-07-23 02:27] [ALPM] upgraded bluez-utils (5.40-1 -> 5.41-1)

Pretty sure this is the fault of this particular version of bluetoothctl shipped with bluez-utils 5.41-1.
Comment by Andreas Radke (AndyRTR) - Saturday, 23 July 2016, 09:35 GMT
Then report this bug upstream and feel free to git bisect your issue.
Comment by Joseph Rice (Daviljoe193) - Saturday, 23 July 2016, 18:53 GMT
Not to be a complete help vampire, but this is the first time I've ever submitted a bug report. Not sure where else I would go to report this bug.
Comment by Andreas Radke (AndyRTR) - Sunday, 24 July 2016, 07:29 GMT
Report this to the bluez mailing list or file a bug to the upstream bluez project tracker (part of the kernel bugzilla).
Comment by Joseph Rice (Daviljoe193) - Sunday, 24 July 2016, 07:56 GMT
Done! Thanks for the guidance. :)
Hopefully something comes of this. https://bugzilla.kernel.org/show_bug.cgi?id=141731

Loading...