*** WoC is now known as Guest1262 | 00:36 | |
x2s | and again, bt is dead. This time I just deactivated and activated it. Restarting the bluetooth service doesn't help. | 11:21 |
---|---|---|
x2s | Will provide logs, soon. | 11:21 |
x2s | mal: http://paste.debian.net/hidden/86029cbb/ | 11:58 |
x2s | This time I've got to restart the phone to get bt working again, it seems. | 11:58 |
x2s | But since I don't expect any calls today anymore it can stay this way. | 11:59 |
x2s | So if you need me to test things, you can ask me the next 10 hours to do so. | 11:59 |
x2s | Maybe even 18 hours | 11:59 |
mal | ok | 12:04 |
elro | x2s: your issue is probably caused by busybox killall. I reported it in sailfishos-porters channel but probably nobody at this channel use xperia x. Try to installing psmisc-tools, this should help. Busybox alternatives: diff, bash, pidof caused few issues in past already | 16:21 |
x2s | elro: how does busybox killall cause this? | 16:25 |
elro | simply bluetooth-rfkill-event use --wait argument which is not supported by busybox killall: https://github.com/mer-hybris/bluetooth-rfkill-event/blob/master/bluetooth-rfkill-event/bluetooth_rfkill_event.c#L903. In your logs you have "killing hciattach failed". Maybe there is also another issue but this one is most obvious | 16:27 |
x2s | that would mean, if there is hciattach running and I kill it restarting bt should work. | 16:29 |
x2s | Let me try. | 16:29 |
elro | or simply install psmisc-tools | 16:30 |
x2s | hciattach is not running. But that doesn't mean it did run. | 16:31 |
x2s | The main issue seems to be "src/rfkill.c:rfkill_init() Failed to open RFKILL control device" | 16:32 |
elro | did you reboot? | 16:36 |
x2s | no, I've promised to keep this state for at least another five hours | 16:38 |
elro | I thought you can reproduce it on every BT enable/disable/enable after reboot if not then this is probably something else besides busybox incompatibility | 16:43 |
elro | this "src/rfkill.c:rfkill_init() Failed" is after first bluetooth-rfkill-event failure so I wouldn't worry about it for now | 16:49 |
x2s | This is not that easily reproducible. It happens with my car stereo, but last time I connected it did not. Then I just disabled and enabled it and were in that state. | 16:55 |
x2s | Oh, last time it did happen with my car stereo, but a restart of the bluetooth service fixed it. | 16:56 |
x2s | But before that I had the same problem and a restart did not fix it (just like now) | 16:56 |
x2s | so it's not quiet obvious to me how to trigger it. And no, I wont walk to my car and take a drive (because that was a good way to reproduce the problem lately, but not always) to test it. | 16:57 |
x2s | I will be waiting until tomorrow, then reboot, and then I can retry reproducing it by restarting bt | 16:58 |
mal | elro: if such an issue is known why hasn't anyone from community made a PR to that bluetooth-rfkill-event | 17:18 |
attah | I seem to recall there was a way to write custom/plain firejail rules, but i can't find how to have sailjail use them... does anyone know? | 17:20 |
elro | mal: I do not have device with such an issue and I do not want to use busybox version of these tools so I will not make PR. I just notice this while debugging some issue for other device. I do not know whether it's know, I just posted it at sailfishos-porters once and pinged you about it but never got any feedback. | 17:21 |
mal | elro: hmm, I must have missed it then | 17:22 |
mal | elro: the issue is that the busybox version is installed by default on devices | 17:25 |
elro | yes, for ports Requires: psmisc-tools is enough but for official devices proper fix is needed | 17:26 |
elro | Be aware I do not know whether this is x2s's main issue but for sure worth to try | 17:26 |
mal | x2s: so hciattach is not running at the moment? | 17:26 |
mal | x2s: do you see /dev/ttyHS0 device in the file system? can you should output of "rfkill" | 17:29 |
attah | Still can't find it in the docs, but apparently it is /etc/sailjail/permissions/<desktop file name>.profile | 18:03 |
x2s | mal: http://paste.debian.net/hidden/c8eafe2b/ no hciattach, devices are there. | 18:33 |
mal | x2s: you can see the failing hciattach command in logs, what happens if you run it manually, use devel-su if needed | 18:38 |
mal | also check dmesg after running the command manually | 18:39 |
x2s | mal: what is the type I should use for hciattach? | 18:40 |
mal | x2s: check the logs you provided me | 18:42 |
mal | the latter log you provided shows the whole command | 18:42 |
mal | look for "executing hciattach" | 18:43 |
x2s | ah | 18:43 |
x2s | Can't get port settings: No such device | 18:43 |
x2s | Can't initialize device: No such device | 18:43 |
mal | can you install "strace" and the prepend strace to the command and send me the whole output | 18:44 |
x2s | please remind me how to install tools from the shell. | 18:44 |
x2s | pacman? | 18:45 |
mal | pkcon install strace | 18:45 |
x2s | many thanks | 18:45 |
x2s | mal: http://paste.debian.net/hidden/563c870c/ | 18:47 |
mal | anything in dmesg when you run that? | 18:48 |
x2s | [531256.948481] bluesleep may not be used | 18:48 |
x2s | that's it. | 18:48 |
mal | what do you see in output of "rfkill" | 18:49 |
x2s | 0 bluetooth bcm43xx Bluetooth blocked unblocked | 18:52 |
x2s | unblocking and running hciattach gave no error. | 18:53 |
mal | hmm | 18:54 |
mal | what do you mean, now hciattach worked? | 18:54 |
x2s | as I said, exit code 0 | 18:54 |
x2s | shall I try restarting the bluetooth service? | 18:55 |
mal | how does dmesg look like, any issues there? | 18:55 |
x2s | [531632.225102] Bluetooth: pK-error | 18:56 |
x2s | but let me show you more | 18:56 |
x2s | maybe there's something I don't see. | 18:56 |
x2s | http://paste.debian.net/hidden/5896ac18/ | 18:56 |
mal | you still got the bluesleep message? | 18:57 |
x2s | that was from earlier | 18:58 |
x2s | not much happend since then. | 18:58 |
x2s | but you are right, it showed up when I ran hciattach | 18:58 |
mal | you can try restarting bluetooth service | 18:59 |
x2s | (If you consider it helpful I could do some screen sharing over jitsi) | 19:02 |
mal | x2s: try toggling the device of and on again using rfkill | 19:04 |
mal | *off | 19:05 |
mal | and check what dmesg shows | 19:05 |
x2s | [532339.354805] Bluetooth: pK-error | 19:05 |
x2s | [532375.138503] Bluetooth: hsuart_power: Turning UART clock off failed (-1) | 19:05 |
x2s | unblocking does not create any dmesg message | 19:06 |
mal | now try hciattach | 19:07 |
x2s | but this might be interesting: http://paste.debian.net/hidden/1cddeb4a/ | 19:07 |
x2s | hciattach is still running. | 19:07 |
attah | How do i make an arbitrary non-build-artifact file available in my shadowbuild directory? | 19:15 |
attah | google is giving nothing :/ | 19:15 |
x2s | bluetooth runs somehow, but it doesn't show up in the ui | 19:17 |
mal | x2s: no at all or the toggle is in wrong mode? | 19:18 |
x2s | not at all | 19:19 |
x2s | well, not as symbol | 19:19 |
x2s | I can try to activate it, but nothing happens | 19:19 |
mal | I mean does the toggle exist in settings app | 19:19 |
x2s | well, rfkill shows bluetooth blocked afterwards. | 19:19 |
mal | ok | 19:20 |
mal | maybe it got out of sync somehow | 19:20 |
x2s | or the ui blocks the device automatically when nothing happens... | 19:21 |
mal | x2s: is there some way you know you can make bluetooth start failing? | 19:21 |
x2s | mal: it usually happens when I drive some time with it connected to my car stereo. When I switch the car off it wont reconnect afterwards, disabling it over the ui makes it not able to reenable it (like it is now) | 19:24 |
x2s | Today I made it happen by just disabling/enabling bt. I will try tomorrow after a reboot if I can make it fail that way. | 19:25 |
x2s | (and my car stereo seems to be something special. BT had problems with it before. Many.) | 19:26 |
mal | ok, maybe we can try to reproduce it by enabling and disabling bluetooth many times | 19:31 |
mal | x2s: one thing you could try, install bluez5-tools and run bluetoothctl and then in the terminal of that app try running "scan on" | 19:34 |
mal | do that after you made hciattach work again | 19:34 |
mal | try doing the rfkill toggling and the hciattach | 19:34 |
x2s | mal: Controller on not available | 21:36 |
x2s | hcitool dev doesn | 21:51 |
x2s | hcitool dev doesn't show any device at all. Even though hciattach is running. | 21:51 |
mal | ok | 23:00 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!