hippy
Posts: 5787
Joined: Fri Sep 09, 2011 10:34 pm
Location: UK

Re: Buster bug report thread

Sat Jun 29, 2019 1:20 am

"sudo apt-get update" failure

Fresh Buster install, Raspbian Full Desktop without recommended software, Pi 3B (non-plus). Wired IPv4 network, WiFi disabled.

For some reason "sudo apt-get update" appears to have decided to use, what I am guessing was, an IPv6 address, even though my router doesn't support that. Hung for a while on the (2a00:1098:0:80:1000:75:0:3) before reporting that as 'failed to connect' to an IPv4 address.

I have never seen apt-get do that before on Stretch, or since installing Buster a few days ago. A second apt-get update worked as expected.

Code: Select all

Linux Pi3B 4.19.50-v7+ #896 SMP Thu Jun 20 16:11:44 BST 2019 armv7l GNU/Linux

Code: Select all

[email protected]:~ $ sudo apt-get update
Get:1 http://archive.raspberrypi.org/debian buster InRelease [25.1 kB]
Get:2 http://archive.raspberrypi.org/debian buster/main armhf Packages [201 kB]
Get:3 http://raspbian.raspberrypi.org/raspbian buster InRelease [15.0 kB]    
Ign:4 http://raspbian.raspberrypi.org/raspbian buster/main armhf Packages    
Ign:5 http://raspbian.raspberrypi.org/raspbian buster/contrib armhf Packages
Ign:4 http://raspbian.raspberrypi.org/raspbian buster/main armhf Packages
Ign:5 http://raspbian.raspberrypi.org/raspbian buster/contrib armhf Packages
Err:4 http://raspbian.raspberrypi.org/raspbian buster/main armhf Packages
  Cannot initiate the connection to raspbian.raspberrypi.org:80 (2a00:1098:0:80:1000:75:0:3). - connect (101: Network is unreachable) Could not connect to raspbian.raspberrypi.org:80 (93.93.128.193), connection timed out [IP: 93.93.128.193 80]
Err:5 http://raspbian.raspberrypi.org/raspbian buster/contrib armhf Packages
  Cannot initiate the connection to raspbian.raspberrypi.org:80 (2a00:1098:0:80:1000:75:0:3). - connect (101: Network is unreachable) [IP: 93.93.128.193 80]
Fetched 241 kB in 55s (4,419 B/s)
Reading package lists... Done
W: Failed to fetch http://raspbian.raspberrypi.org/raspbian/dists/buster/main/binary-armhf/Packages  Cannot initiate the connection to raspbian.raspberrypi.org:80 (2a00:1098:0:80:1000:75:0:3). - connect (101: Network is unreachable) Could not connect to raspbian.raspberrypi.org:80 (93.93.128.193), connection timed out [IP: 93.93.128.193 80]
W: Failed to fetch http://raspbian.raspberrypi.org/raspbian/dists/buster/contrib/binary-armhf/Packages  Cannot initiate the connection to raspbian.raspberrypi.org:80 (2a00:1098:0:80:1000:75:0:3). - connect (101: Network is unreachable) [IP: 93.93.128.193 80]
W: Some index files failed to download. They have been ignored, or old ones used instead.

Code: Select all

[email protected]:~ $ ifconfig
eth0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet 192.168.0.130  netmask 255.255.255.0  broadcast 192.168.0.255
        inet6 fe80::6abe:74e4:ba9b:f713  prefixlen 64  scopeid 0x20<link>
        ether b8:27:eb:8b:d2:9f  txqueuelen 1000  (Ethernet)
        RX packets 87376  bytes 40717693 (38.8 MiB)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 51641  bytes 6575254 (6.2 MiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

lo: flags=73<UP,LOOPBACK,RUNNING>  mtu 65536
        inet 127.0.0.1  netmask 255.0.0.0
        inet6 ::1  prefixlen 128  scopeid 0x10<host>
        loop  txqueuelen 1000  (Local Loopback)
        RX packets 1228  bytes 73680 (71.9 KiB)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 1228  bytes 73680 (71.9 KiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

plugwash
Forum Moderator
Forum Moderator
Posts: 3439
Joined: Wed Dec 28, 2011 11:45 pm

Re: Buster bug report thread

Sat Jun 29, 2019 2:39 am

It's normal for it to try both IPv4 and IPv6, it seems some of the error handling paths can only report one error even though multiple IPs were tried.

klricks
Posts: 6547
Joined: Sat Jan 12, 2013 3:01 am
Location: Grants Pass, OR, USA
Contact: Website

Re: Buster bug report thread

Sat Jun 29, 2019 3:01 am

Composite output does not work on 3B+. I know composite is now disabled by default but there does not seem to be any way to enable it on the 3B+.

see: https://www.raspberrypi.org/forums/view ... 6&t=243855
Unless specified otherwise my response is based on the latest and fully updated Raspbian Buster w/ Desktop OS.

User avatar
PeterO
Posts: 4942
Joined: Sun Jul 22, 2012 4:14 pm

Re: Buster bug report thread

Sat Jun 29, 2019 5:47 am

I've seen some odd "tearing" artefacts on openGLES. They are most visible with a wireframe model that is moving sideways across the window.
I tried to grab a frame with GIMP, but the buffer contents it grabs do not show the problem (or I failed to grab it when it happened, or grabbing the frame stopped it happening) anyway.....

This seems to be tied to the screen update process because eglSwapInterval(eglDisplay,0); stops it happening (but produces unpredictable and varying frame rates).

I'll try and grab a photo of the problem.

PeterO
Discoverer of the PI2 XENON DEATH FLASH!
Interests: C,Python,PIC,Electronics,Ham Radio (G0DZB),1960s British Computers.
"The primary requirement (as we've always seen in your examples) is that the code is readable. " Dougie Lawson

User avatar
jojopi
Posts: 3079
Joined: Tue Oct 11, 2011 8:38 pm

Re: Buster bug report thread

Sat Jun 29, 2019 9:17 am

The Debian package policykit-1 provides a file /etc/polkit-1/localauthority.conf.d/51-debian-sudo.conf that says:

Code: Select all

[Configuration]
AdminIdentities=unix-group:sudo
In the corresponding Raspbian package, this file is missing. The omission means that admin users cannot reboot or install packages, or are required to provide the root password when they should either be trusted or allowed to authenticate with their own credentials. It is a particular problem when no root password is set.

It seems that RPF/T images have long ago worked around this by adding unix-user:pi to 60-desktop-policy.conf. However, that is really an awful hack and would never have been necessary without the upstream Raspbian issue. It may also be a reason why people who create or rename users find they do not work as well as "pi".

Looking at the source package, the problem appears to be that there is logic in "rules" to check whether "dpkg-vendor --is" Debian or Ubuntu, and the "Raspbian" system falls down a crack.

User avatar
DougieLawson
Posts: 35808
Joined: Sun Jun 16, 2013 11:19 pm
Location: Basingstoke, UK
Contact: Website Twitter

Re: Buster bug report thread

Sat Jun 29, 2019 10:23 am

This isn't really a bug, more an annoyance (that's been there since Jessie or maybe Wheezy).

Can we get Buster to include vim (full fat version) installed by default, because vim-tiny is borderline unusable?
Note: Having anything humorous in your signature is completely banned on this forum. Wear a tin-foil hat and you'll get a ban.

Any DMs sent on Twitter will be answered next month.

This is a doctor free zone.

User avatar
DougieLawson
Posts: 35808
Joined: Sun Jun 16, 2013 11:19 pm
Location: Basingstoke, UK
Contact: Website Twitter

Re: Buster bug report thread

Sat Jun 29, 2019 11:15 am

I know you're recommending going with LibreELEC for Kodi (as it's still very broken in Buster).

But TVheadend is also broken: see https://www.raspberrypi.org/forums/view ... 6&t=243920
Note: Having anything humorous in your signature is completely banned on this forum. Wear a tin-foil hat and you'll get a ban.

Any DMs sent on Twitter will be answered next month.

This is a doctor free zone.

globiboulga666
Posts: 20
Joined: Sun Jun 17, 2018 6:38 pm

Re: Buster bug report thread

Sat Jun 29, 2019 1:02 pm

Less than one day after having updated a perfectly stable 24/24 running Pi2 to Raspbian 10, few minutes ago, it instantly crashed after typing "apt update && apt upgrade" (at the moment of pressing enter). Since it was through putty and I had no reaction anymore, I waited a little bit (but nothing came back until power reboot).

No one can be 100% sure that this is related to the new Raspbian, so just in case others people are getting the same strange behavior/instability after upgrading.

Code: Select all

Jun 29 07:14:50 Pi2-Cam-Crypte kernel: [   58.769583] IPv6: ADDRCONF(NETDEV_CHANGE): wlan0: link becomes ready
Jun 29 14:28:28 Pi2-Cam-Crypte kernel: [26041.173305] Alignment trap: not handling instruction e1943f9f at [<808350a0>]
Jun 29 14:28:28 Pi2-Cam-Crypte kernel: [26041.173322] Unhandled fault: alignment exception (0x001) at 0xb681029a
Jun 29 14:28:28 Pi2-Cam-Crypte kernel: [26041.179875] pgd = f8c1bd42
Jun 29 14:28:28 Pi2-Cam-Crypte kernel: [26041.182585] [b681029a] *pgd=3681141e(bad)
Jun 29 14:28:28 Pi2-Cam-Crypte kernel: [26041.186607] Internal error: : 1 [#1] SMP ARM
Jun 29 14:28:28 Pi2-Cam-Crypte kernel: [26041.190876] Modules linked in: fuse appletalk psnap llc ax25 r8188eu(C) lib80211 sha256_generic cfg80211 rfkill bcm2835_codec(C) bcm2835_v4l2(C) v4l2_mem2mem bcm2835_mmal_vchiq(C) snd_bcm2835(C) v4l2_common videobuf2_dma_contig videobuf2_vmalloc videobuf2_memops snd_pcm videobuf2_v4l2 videobuf2_common raspberrypi_hwmon videodev snd_timer hwmon snd media uio_pdrv_genirq uio fixed i2c_dev ip_tables x_tables ipv6
Jun 29 14:28:28 Pi2-Cam-Crypte kernel: [26041.227311] CPU: 2 PID: 17403 Comm: (time-dir) Tainted: G         C        4.19.50-v7+ #896
Jun 29 14:28:28 Pi2-Cam-Crypte kernel: [26041.235660] Hardware name: BCM2835
Jun 29 14:28:28 Pi2-Cam-Crypte kernel: [26041.239073] PC is at _raw_spin_lock_irqsave+0x2c/0x68
Jun 29 14:28:28 Pi2-Cam-Crypte kernel: [26041.244123] LR is at   (null)
Jun 29 14:28:28 Pi2-Cam-Crypte kernel: [26041.247086] pc : [<808350a4>]    lr : [<00000000>]    psr: 40000193
Jun 29 14:28:28 Pi2-Cam-Crypte kernel: [26041.253349] sp : 9bc63d38  ip : 9bc62000  fp : 9bc63d4c
Jun 29 14:28:28 Pi2-Cam-Crypte kernel: [26041.258570] r10: 9bc63e34  r9 : 80d04d48  r8 : 00000000
Jun 29 14:28:28 Pi2-Cam-Crypte kernel: [26041.263792] r7 : b174ecc0  r6 : b76a8fc0  r5 : 20000113  r4 : b681029a
Jun 29 14:28:28 Pi2-Cam-Crypte kernel: [26041.270316] r3 : 1000b762  r2 : 9bc63cd8  r1 : 00000000  r0 : 8083509c
Jun 29 14:28:28 Pi2-Cam-Crypte kernel: [26041.276842] Flags: nZcv  IRQs off  FIQs on  Mode SVC_32  ISA ARM  Segment user
Jun 29 14:28:28 Pi2-Cam-Crypte kernel: [26041.284062] Control: 10c5387d  Table: 1e30806a  DAC: 00000055
Jun 29 14:28:28 Pi2-Cam-Crypte kernel: [26041.289810] Process (time-dir) (pid: 17403, stack limit = 0x4cca733e)
Jun 29 14:28:28 Pi2-Cam-Crypte kernel: [26041.296249] Stack: (0x9bc63d38 to 0x9bc64000)
Jun 29 14:28:28 Pi2-Cam-Crypte kernel: [26041.300604] 3d20:                                                       b6810036 b681029a
Jun 29 14:28:28 Pi2-Cam-Crypte kernel: [26041.308783] 3d40: 9bc63d6c 9bc63d50 8029d4ac 80835084 9e32e900 b76a8fc0 b76a8fc0 2ec7059f
Jun 29 14:28:28 Pi2-Cam-Crypte kernel: [26041.316962] 3d60: 9bc63d84 9bc63d70 8027ea00 8029d478 9bc63e34 b76a8fc0 9bc63dbc 9bc63d88
Jun 29 14:28:28 Pi2-Cam-Crypte kernel: [26041.325139] 3d80: 80274ab4 8027e9f0 00000000 b67e4d10 a0000013 00000001 00000000 b67e329c
Jun 29 14:28:28 Pi2-Cam-Crypte kernel: [26041.333318] 3da0: b76a8fc0 00000036 80d04d48 9bc63e34 9bc63e1c 9bc63dc0 80236700 802749dc
Jun 29 14:28:28 Pi2-Cam-Crypte kernel: [26041.341496] 3dc0: 0012f85b 00000000 802ce810 b63cd000 0000002e b67e4d0c 8027511c 0000002e
Jun 29 14:28:28 Pi2-Cam-Crypte kernel: [26041.349675] 3de0: 00000040 a0000113 b67e31d0 684c0781 9bc63e1c 00000027 00000036 80d04d48
Jun 29 14:28:28 Pi2-Cam-Crypte kernel: [26041.357853] 3e00: 9bc63e34 00000040 00510000 00000054 9bc63e9c 9bc63e20 80275818 80236344
Jun 29 14:28:28 Pi2-Cam-Crypte kernel: [26041.366032] 3e20: 00000000 00000000 80d04d48 00000041 80d05688 b174ecc0 00000054 006200ca
Jun 29 14:28:28 Pi2-Cam-Crypte kernel: [26041.374210] 3e40: 00000027 00517000 9e308010 9e308010 00000000 00000000 00000000 00000000
Jun 29 14:28:28 Pi2-Cam-Crypte kernel: [26041.382389] 3e60: 9bcaa45c b73fd7fc 00000000 684c0781 9bc63e9c 9bc63fb0 9e32e900 00000017
Jun 29 14:28:28 Pi2-Cam-Crypte kernel: [26041.390568] 3e80: 00510398 b5e24a00 b5e24a3c 00000054 9bc63ef4 9bc63ea0 808358e8 80274de0
Jun 29 14:28:28 Pi2-Cam-Crypte kernel: [26041.398746] 3ea0: b63c4015 802a6a64 00000000 b67665d8 b69e4540 00000101 00000002 0000044e
Jun 29 14:28:28 Pi2-Cam-Crypte

MrWhiter
Posts: 5
Joined: Sun Jun 11, 2017 5:14 pm

CEC not working in kodi

Sat Jun 29, 2019 6:08 pm

Just upgraded an pi3 to buster and switched to the VC4 V3D driver.

Although Kodi lists the CEC adapter, it doesn't respond to the remote control.
This did work just before the update, with the closed source driver on stretch.
Video plays fine, so the driver does do it's main task.

An excerpt from .kodi/temp/kodi.log:

Code: Select all

16:25:17.744 T:1832153104  NOTICE: Running on Raspbian GNU/Linux 10 (buster), kernel: Linux ARM 32-bit version 4.19.42-v7+

16:25:22.677 T:1832153104  NOTICE: GL_VENDOR = Broadcom
16:25:22.677 T:1832153104  NOTICE: GL_RENDERER = VC4 V3D 2.1
16:25:22.677 T:1832153104  NOTICE: GL_VERSION = OpenGL ES 2.0 Mesa 18.3.6
16:25:22.677 T:1832153104  NOTICE: GL_SHADING_LANGUAGE_VERSION = OpenGL ES GLSL ES 1.0.16

16:25:25.314 T:1398229648  NOTICE: Register - new cec device registered on cec->RPI: CEC Adapter (2708:1001)

Sir FICO
Posts: 29
Joined: Sat Nov 07, 2015 11:34 pm

Re: Buster bug report thread

Sat Jun 29, 2019 10:25 pm

netatalk for connection to apple mac computers fails. It can be installed with 'sudo apt-get install netatalk'. The mac 'finder' will show the rpi computer under the Network and will allow a 'connect as' login. However after the login the finder just says 'not connected'. I did not have any problems with stretch so I assume a buster issue.

globiboulga666
Posts: 20
Joined: Sun Jun 17, 2018 6:38 pm

Re: Buster bug report thread

Sat Jun 29, 2019 10:34 pm

Another Pi2, another SD card, read/write access frozen with green led on, while calling Terminal with ctrl+alt+t
There is definitely some stability issue with Pi2, I guess with the Kernel used in this new Raspbian 10

anfractuosity
Posts: 25
Joined: Wed Jun 21, 2017 7:23 am

Re: Buster bug report thread

Sun Jun 30, 2019 10:26 am

I recently installed Buster on a Pi zero, the footage from a Pi noir camera now seems to have dramatically changed to:

Image

From this in Stretch:

Image

As far as I can tell, it doesn't seem related to the version of Raspivid, as I tried the same commit version as in Stretch, and still got purple results.

I'm wondering if something has changed in the firmware relating to the camera?

(I get the same results from picamera and raspivid)

User avatar
PeterO
Posts: 4942
Joined: Sun Jul 22, 2012 4:14 pm

Re: Buster bug report thread

Sun Jun 30, 2019 11:45 am

I'm started updating some of my openGLES code and I'm seeing some artifacts which seem related to frame buffer swapping.
The picture below was extracted from a 480 frame/second video taken on a phone. The artifacts are visible by eye as well,
RedGreen2.png
RedGreen2.png (29.76 KiB) Viewed 2166 times
Alternate frames of the wire frame model are being drawn with red and green lines. Screen swapping is set to "eglSwapInterval(eglDisplay,1);"

Now the camera is not synchronized to the monitor frame rate, so the change from red to green could happen at any point on the frame.
But the picture is showing green to red to green to red transitions. Remember that red and green pixels are from consecutive frames.

I tried to capture a frame directly with the GIMP, but the data it gets from the frame buffer does not show this problem.

Note also that the effect happens at the same vertical position on the monitor regardless of how the window is positioned.

PeterO
Discoverer of the PI2 XENON DEATH FLASH!
Interests: C,Python,PIC,Electronics,Ham Radio (G0DZB),1960s British Computers.
"The primary requirement (as we've always seen in your examples) is that the code is readable. " Dougie Lawson

User avatar
rin67630
Posts: 461
Joined: Fri Mar 04, 2016 10:15 am

Re: Buster bug report thread HDMI mode 1024*600

Sun Jun 30, 2019 2:57 pm

Just freshly installed Buster on a Raspi 3a+ that ran previously Stretch.
I am running quasi headless over VNC in a fast LAN connection, but have a 1024*600 Px monitor attached over HDMI to verify.
To get that resolution I have entered the following in /boot/config.txt:

Code: Select all

hdmi_force_hotplug=1
hdmi_cvt=1024 600 60 1 0 0 0
hdmi_group=2
hdmi_mode=87 
since I have done that, the local HDMI screen output is perfect and fast, but the VNC video is catastrophically slow. Copy/paste isn't possible any more.

With the same settings in Stretch everything was OK.

Without the above mentioned parameters in /boot/config.txt e.g. a stndard resolution, everything is OK again.

To mention: my /boot/config.tx contains also these (unrelated) settings

Code: Select all

[all]
#dtoverlay=vc4-fkms-v3d
enable_uart=0

# Swap serial ports to Pi2 state
# Swap Serial ports and Bluetooth
dtoverlay=pi3-miniuart-bt
enable_uart=1

Gadgetguy
Posts: 99
Joined: Fri Aug 15, 2014 2:55 am

Re: Buster bug report thread

Sun Jun 30, 2019 9:44 pm

I guess this post may not qualify as an official bug report but is more my somewhat subjective but distinct impressions of problems I experienced with Raspbian Buster. I loaded a fresh Buster image onto a 32gb Samsung Evo plus sd for my Raspberry Pi 3B. (It is not my fastest sd card - my sandisk extreme plus 32 gb seems to me to be more responsive on things like web browsing etc. but the samsung evo plus still has a pretty good reputation as a speedy card for the Raspberry Pi.)

As I do in stretch, I allocated 128gb ram to the gpu and used the legacy graphics driver. The comparison I make below is for this same card running stretch versus buster

My “impression” is that the Chromium browser version 72 in Stretch is on a 3B noticeably faster, more fluid and responsive and uses less cpu and ram and consequently runs at a lower cpu temperature than Chromium version 74 in Buster. I find in Buster I get frequent temperature warnings that I rarely get in stretch. Buster loads tabs more slowly switches between tabs with more difficulty, and starts to slow down, freeze or overheat with fewer tabs open. I am wondering if Raspbian Buster’s Chromium browser has been optimized for the new Raspberrry 4 at the expense of the experience on older Pi’s. Will stretch’s Chromium be updated periodically as in the past or will it be frozen.

I have done the Ookla internet speed test a number of times with both OS on this same card and I consistently get much higher internet speeds on the Stretch system. (This test may not be definitive because my internet speeds of late seem to fluctuate even operating on the same system)

I am wondering if anyone else has experienced these problems with older Raspberry Pis on Buster’s Chromium browser using the legacy driver? The Chromium experience on Buster with an older pi using the fkms driver is even more problematic with more frequent freezes and artifacts that stay on the screen untill reboot.

On Stretch I enjoy using vlc player and also omxplayer ( omxplayergui) and they both work well on Buster with the legacy driver. I am wondering if Vlc on stretch will be kept up to date to reflect changes in youtube code etc? I guess if we use fkms on Buster we will lose osd timings and subtitles on omxplayer?


However on Stretch I also use mpv video player (compiled with mmal enabled in ffmpeg) with smplayer as a gui. So far I have not succeeded in getting it to work satisfactorily with hw acceleration in Buster and I don’t want to lose it.

On stretch mpv in smplayer plays just about any video format up to 1080p
( although some in software). By configuring smplayer to open mpv in mpv’s own (overlay) window and by toggling back and forth between full screen and a small defined window I can get gui control of mpv which opens up all kinds of options. Smplayer in addition to using youtube-dl has its own very fast internal youtube code that is kept up to date and by using the open with extension by G Lankow in the Chrome store I am able to speedily and efficiently browse youtube videos and watch them either full screen or in a small overlay window defined as a specific size and at a specific unobtrusive location on the screen so as to permit multi-tasking. To make a long story short I really do not want to lose a functional mpv player on Buster. Has anyone been able to get mpv working in Buster with an older pi?

On Buster using fkms graphics driver and the ffmpeg preinstalled in Buster and the mpv in the buster repository and configuring mpv to use the default gpu driver mpv will play a play a 1080p youtube video albeit at very high cpu usage which quickly results in overheating.

It also “seems” to me that Buster in general is somewhat more prone to overheating in an older Pi.
I do note that glxgears now seems to work fine with the legacy driver.

I look forward to getting a Raspberry Pi 4 at some point but I hope to keep my older Pis working just as well as they did in Stretch.

beta-tester
Posts: 1236
Joined: Fri Jan 04, 2013 1:57 pm
Location: de_DE

Re: Buster bug report thread

Mon Jul 01, 2019 12:17 pm

fd_ wrote:
Thu Jun 27, 2019 9:02 pm
... keyboard layout selection is completely garbled (some layout names are displayed in French, some in Dutch, some as black squares, really strange):
same glitches on a RRi1 rev 1.2 with 2019-06-20-raspbian-buster-lite.zip
and also after the very first login, a bunch of start messages are jamming the command line screen.

also the old glitch still exist: Raspbian: login cursor jumps to pos0 of line
{ I only give negative feedback }
RPi Model B (rev1, 256MB) & B (rev2, 512MB) & B+, RPi2B, RPi3B, RPi3B+, RPiZeroW, ...

pista24
Posts: 2
Joined: Sat Jun 29, 2019 9:47 am

Re: Buster bug report thread

Tue Jul 02, 2019 8:01 am

Buster won't boot headless with the Sense Hat attached

https://www.raspberrypi.org/forums/view ... 8&t=243949

User avatar
DougieLawson
Posts: 35808
Joined: Sun Jun 16, 2013 11:19 pm
Location: Basingstoke, UK
Contact: Website Twitter

Re: Buster bug report thread

Tue Jul 02, 2019 9:51 am

Another one that's not really a bug (more a requirement).

Can Raspbian include ntfs-3g by default?
That would stop 10s/100s of threads appearing on here asking why folks can't write to their NTFS drive.
Note: Having anything humorous in your signature is completely banned on this forum. Wear a tin-foil hat and you'll get a ban.

Any DMs sent on Twitter will be answered next month.

This is a doctor free zone.

tpyo kingg
Posts: 588
Joined: Mon Apr 09, 2018 5:26 pm
Location: N. Finland

Re: Buster bug report thread

Tue Jul 02, 2019 10:21 am

Mentioned in the other thread, clementine is missing from the buster repositories. It would be nice to have again.

Code: Select all

$ lsb_release -rd
Description:    Raspbian GNU/Linux 10 (buster)
Release:        10

$ apt-cache search clementine
gnome-shell-extension-mediaplayer - GNOME Shell extension to control media players

$ apt-cache search strawberry

I know elsewhere there is some talk of the fork, strawberry, but it is not in the repository either.

Ernst
Posts: 1215
Joined: Sat Feb 04, 2017 9:39 am
Location: Germany

Re: Buster bug report thread

Tue Jul 02, 2019 11:28 am

Problem with static IP addresses, bug or a feature ?

In dhcpcd.conf the identical static IP addresses has been configured for eth0 and wlan0,
wpa_supplicant.conf is available and valid.

Using Stretch with the ethernet cable attached the address is assigned to eth0 and wlan0 is not assigned. (good)
Using Stretch with the ethernet cable removed the address is assigned to wlan0 and eth0 is not assigned. (good)

Using Buster with the ethernet cable attached the address is assigned to both eth0 and wlan0 (bad)
Using Buster with the ethernet cable removed the address is assigned to wlan0 and eth0 is not assigned. (good)

With Buster and both eth0 and wlan0 connected the following can be seen in a packet trace:

Code: Select all

Giga-Byt_62:fc:75      Broadcast          ARP        42    Who has 192.168.0.235? Tell 192.168.0.5
192.168.0.235          Giga-Byt_62:fc:75  ARP        60   192.168.0.235 is at dc:a6:32:03:65:20
dc:a6:32:03:65:21      Giga-Byt_62:fc:75  ARP        60   192.168.0.235 is at dc:a6:32:03:65:21
Stretch on Pi3b:

/etc/dhcpcd.conf:

Code: Select all

interface eth0
static ip_address=192.168.0.240/24
static routers=192.168.0.254
static domain_name_servers=192.168.0.254

interface wlan0
static ip_address=192.168.0.240/24
static routers=192.168.0.254
static domain_name_servers=192.168.0.254

ifconfig (eth+wlan):

Code: Select all

eth0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet 192.168.0.240  netmask 255.255.255.0  broadcast 192.168.0.255
        ether b8:27:eb:d1:0c:99  txqueuelen 1000  (Ethernet)

wlan0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        ether b8:27:eb:84:59:cc  txqueuelen 1000  (Ethernet)
ifconfig (wlan):

Code: Select all

eth0: flags=4099<UP,BROADCAST,MULTICAST>  mtu 1500
        ether b8:27:eb:d1:0c:99  txqueuelen 1000  (Ethernet)

wlan0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet 192.168.0.240  netmask 255.255.255.0  broadcast 192.168.0.255
        ether b8:27:eb:84:59:cc  txqueuelen 1000  (Ethernet)
Buster on Pi4B (also Pi3B):

/etc/dhcpcd.conf:

Code: Select all

interface eth0
static ip_address=192.168.0.235/24
static routers=192.168.0.254
static domain_name_servers=192.168.0.254

interface wlan0
static ip_address=192.168.0.235/24
static routers=192.168.0.254
static domain_name_servers=192.168.0.254
ifconfig (eth0+wlan)

Code: Select all

eth0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet 192.168.0.235  netmask 255.255.255.0  broadcast 192.168.0.255
        ether dc:a6:32:03:65:20  txqueuelen 1000  (Ethernet)

wlan0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet 192.168.0.235  netmask 255.255.255.0  broadcast 192.168.0.255
        ether dc:a6:32:03:65:21  txqueuelen 1000  (Ethernet)
ifconfig (wlan)

Code: Select all

eth0: flags=4099<UP,BROADCAST,MULTICAST>  mtu 1500
        ether dc:a6:32:03:65:20  txqueuelen 1000  (Ethernet)

wlan0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet 192.168.0.235  netmask 255.255.255.0  broadcast 192.168.0.255
        ether dc:a6:32:03:65:21  txqueuelen 1000  (Ethernet)
The road to insanity is paved with static ip addresses

jamesh
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 23380
Joined: Sat Jul 30, 2011 7:41 pm

Re: Buster bug report thread

Tue Jul 02, 2019 1:13 pm

klricks wrote:
Sat Jun 29, 2019 3:01 am
Composite output does not work on 3B+. I know composite is now disabled by default but there does not seem to be any way to enable it on the 3B+.

see: https://www.raspberrypi.org/forums/view ... 6&t=243855
OK, have found the issue here. Will be sorted out in a future rpi-update.
Principal Software Engineer at Raspberry Pi (Trading) Ltd.
Contrary to popular belief, humorous signatures are allowed. Here's an example...
"My grief counseller just died, luckily, he was so good, I didn't care."

swahren
Posts: 102
Joined: Mon Sep 19, 2016 5:24 pm
Location: Germany

Re: Buster bug report thread

Tue Jul 02, 2019 4:56 pm

bluesign2k wrote:
Wed Jun 26, 2019 4:27 pm
I found myself needing to re-image an SD that I've been using on a Pi Zero-W yesterday and found that Buster was the current available image.
FWIW, I use the "lite" image and after imaging I put my standard wpa_supplicanf.conf file in boot along with and SSH file so I can do a headless setup and only supply power.

I've used this method (and files) countless times and it works perfectly on Pi2, Pi3B+ and Pi Zero-W, however, this now seem to be broken with the current Buster image. Using an old Stretch image and the same file on the same hardware (Pi, SD, powersupply) and everything works fine.
Doing iwconfig confirms connection to the hotspot, but ifconfig occasionally shows an ipv6 address occasionally, but 9 times out of 10, no ip.

In trying to debug I found that dmesg is spammed is endless coppies of:

Code: Select all

[ 1373.354269] ------------[ cut here ]------------
[ 1373.354809] WARNING: CPU: 0 PID: 341 at drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c:5126 brcmf_cfg80211_set_pmk+0x64/0x84 [brcmfmac]
For the timebeing I've gone back to Stretch, but am happy to try re-imaging another card to try suggestions for a fix on Buster.
This is an issue in wpa_supplicant 2.8. Please look at http://lists.infradead.org/pipermail/ho ... 40288.html

swahren
Posts: 102
Joined: Mon Sep 19, 2016 5:24 pm
Location: Germany

Re: Buster bug report thread

Tue Jul 02, 2019 5:04 pm

hippy wrote:
Thu Jun 27, 2019 2:51 pm
Using gpiod / libgpiod, what is reported by the gpiod related commands is not the same as reported by the gpio utility. here the UART (14,15) has been enabled for userland use with enable_uart=1
This is a kernel / pinctrl issue. I cannot promise this will be fixed in mainline, but you can look at this https://github.com/lategoodbye/rpi-zero/issues/42 for more information.

hippy
Posts: 5787
Joined: Fri Sep 09, 2011 10:34 pm
Location: UK

Re: Buster bug report thread

Tue Jul 02, 2019 6:12 pm

Can't find the report but boot to CLI with auto-login was positioning the cursor to left of screen meaning over-typing command prompt on first command.

After an rpi-update that seems to be fixed but introduces 'random cases' of cursor jumping back to far left. Noticed mostly with command history cursor key movements.

I say 'random' because I haven't found a definite always reproducible case. As noted that's after rpi-update, using a composite display, so may just be me / my config. This seems to reproduce for me -

Code: Select all

uname -a
sudo reboot
<up arrow> <up arrow> <return>
<up arrow> <return>
Keep doing the <up arrow> <return> until it fails, usually on the <up arrow>.

Not the best photo I've ever produced ...
CURSOR.gif
CURSOR.gif (21.87 KiB) Viewed 1698 times
Added: Composite only it seems. Cannot reproduce using HDMI. Also on HDMI. May be time related (?!) I had rebooted, wandered off for a while, failed on first up arrow.

Extra added: Time related. Booted, cursor to the right of the prompt. Leave it. Some time later cursor has moved to far left of screen.

ShiftPlusOne
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 5879
Joined: Fri Jul 29, 2011 5:36 pm
Location: The unfashionable end of the western spiral arm of the Galaxy

Re: Buster bug report thread

Wed Jul 03, 2019 10:11 am

hippy wrote:
Tue Jul 02, 2019 6:12 pm
Can't find the report but boot to CLI with auto-login was positioning the cursor to left of screen meaning over-typing command prompt on first command.

After an rpi-update that seems to be fixed but introduces 'random cases' of cursor jumping back to far left. Noticed mostly with command history cursor key movements.

I say 'random' because I haven't found a definite always reproducible case. As noted that's after rpi-update, using a composite display, so may just be me / my config. This seems to reproduce for me -

Code: Select all

uname -a
sudo reboot
<up arrow> <up arrow> <return>
<up arrow> <return>
Keep doing the <up arrow> <return> until it fails, usually on the <up arrow>.

Not the best photo I've ever produced ...
CURSOR.gif

Added: Composite only it seems. Cannot reproduce using HDMI. Also on HDMI. May be time related (?!) I had rebooted, wandered off for a while, failed on first up arrow.

Extra added: Time related. Booted, cursor to the right of the prompt. Leave it. Some time later cursor has moved to far left of screen.
This has been a problem in Stretch as well. IIRC, the systemd's fsck service prints a new line regardless of whether there has been any output, but I may be wrong about that.

Return to “Raspbian”