Puppy Linux Discussion Forum

cut from xerrs.log re obs-studio, may be of use

/usr/bin/jwm
libpng warning: Interlace handling should be turned on when using png_read_image
/root/Startup/pa-applet_tray: waited 1 seconds for pulseaudio daemon.
/root/Startup/pulseaudio: waited 2 seconds for bluetoothd
/root/Startup/pulseaudio: starting pulseaudio daemon
/root/Startup/pulseaudio: executing start-pulseaudio-x11
Fontconfig warning: Directory/file mtime in the future. New fonts may not be detected.
QStandardPaths: XDG_RUNTIME_DIR not set, defaulting to '/tmp/runtime-root'
Attempted path: share/obs/obs-studio/locale/en-US.ini
Attempted path: /usr/share/obs/obs-studio/locale/en-US.ini
Attempted path: share/obs/obs-studio/locale.ini
Attempted path: /usr/share/obs/obs-studio/locale.ini
Attempted path: share/obs/obs-studio/themes/Dark.qss
Attempted path: /usr/share/obs/obs-studio/themes/Dark.qss
info: CPU Name: Intel(R) Core(TM) i5-3470S CPU @ 2.90GHz
info: CPU Speed: 3392.133MHz
info: Physical Cores: 4, Logical Cores: 4
info: Physical Memory: 3831MB Total, 47MB Free
info: Kernel Version: Linux 5.10.90
info: Distribution: Missing /etc/os-release !
info: Window System: X11.0, Vendor: The X.Org Foundation, Version: 1.20.8
info: Portable mode: false
QMetaObject::connectSlotsByName: No matching signal for on_advAudioProps_clicked()
QMetaObject::connectSlotsByName: No matching signal for on_advAudioProps_destroyed()
QMetaObject::connectSlotsByName: No matching signal for on_actionGridMode_triggered()
QMetaObject::connectSlotsByName: No matching signal for on_program_customContextMenuRequested(QPoint)
QMetaObject::connectSlotsByName: No matching signal for on_transitionRemove_clicked()
Attempted path: share/obs/obs-studio/themes/Dark/no_sources.svg
Attempted path: /usr/share/obs/obs-studio/themes/Dark/no_sources.svg
info: OBS 0.0.1 (linux)
info: ---------------------------------
info: ---------------------------------
info: audio settings reset:
samples per sec: 48000
speakers: 2
info: ---------------------------------
info: Initializing OpenGL...
info: Loading up OpenGL on adapter Intel Open Source Technology Center Mesa DRI Intel(R) HD Graphics 2500 (IVB GT1)
info: OpenGL loaded successfully, version 4.2 (Core Profile) Mesa 20.0.2, shading language 4.20
info: ---------------------------------
info: video settings reset:
base resolution: 1919x1079
output resolution: 1276x718
downscale filter: Bicubic
fps: 30/1
format: NV12
YUV mode: 709/Partial
info: NV12 texture support not available
info: Audio monitoring device:
name: Default
id: default
info: ---------------------------------
warning: Failed to load 'en-US' text for module: 'decklink-captions.so'
warning: Failed to load 'en-US' text for module: 'decklink-ouput-ui.so'
libDeckLinkAPI.so: cannot open shared object file: No such file or directory
warning: A DeckLink iterator could not be created. The DeckLink drivers may not be installed
info: No blackmagic support
warning: v4l2loopback not installed, virtual camera disabled
info: FFMPEG VAAPI supported
info: ---------------------------------
info: Loaded Modules:
info: text-freetype2.so
info: rtmp-services.so
info: obs-x264.so
info: obs-transitions.so
info: obs-outputs.so
info: obs-filters.so
info: obs-ffmpeg.so
info: linux-v4l2.so
info: linux-pulseaudio.so
info: linux-jack.so
info: linux-decklink.so
info: linux-capture.so
info: linux-alsa.so
info: image-source.so
info: frontend-tools.so
info: decklink-ouput-ui.so
info: decklink-captions.so
info: ---------------------------------
info: ==== Startup complete ===============================================

Same issue that of retiredt00 reported in his earlier post while booting from my internal drive (SSD NVME). I have been reporting this booting issue last several releases. However, nobody provided any work around for my issue.

This is a problem that gets reported regularly. I have received an email from mike:

When run dunfell on usb, it always ask for password before boot on any further, this is great as expected; thanks!
however, if dunfell installed on harddrive, the first re-boot, it did not even ask for password (even though I did setup passwod at the beginning). Without asking for password, it just went on to boot on, left me wondering where is the security?
is this normal? or bug?

My reply to mike:

mike,
Yeah, you will have to uninstall EasyOS from the hard drive and start
again. Well, no, actually, just delete everything under
/mnt/${WKG_DEV}/${WKG_DIR}
...there is also .session hidden folder, that must also be deleted.

The filesystem in the hard drive partition must be ext4 and must have
its "encrypt" flag enabled. Do, that, and when you first bootup, Easy
will see that the partition has the encrypt flag enabled and will ask
for a password.

This is how you enable encryption, unmount the partition, then in a terminal, type this,
substitute whatever the partition is:

# tune2fs -O encrypt,^has_journal /dev/${WKG_DEV}

...that is a letter capital "O". This is also mentioned here:

https://easyos.org/install/how-to-insta ... drive.html

Extra notes:

You can check if a partition has encrypt enabled by doing this, on my PC looking at sda8:

# tune2fs -l /dev/sda8 | grep 'Filesystem features'
Filesystem features:      ext_attr resize_inode dir_index filetype extent 64bit flex_bg encrypt sparse_super large_file huge_file dir_nlink extra_isize metadata_csum

@ BarryK

Thank you
Would by handy of the BOOT_SPECS parameters could be passed as kernel line arguments, as it is done for DEV location, so the initrd exercise could be bypassed.

Another mater in my setup appears to be the kernel SATA support that hangs the boot process for ~100 sec as it can be seen in the log.

[    0.317837] fotg210_hcd: FOTG210 Host Controller (EHCI) Driver
[    0.318723] Warning! fotg210_hcd should always be loaded before uhci_hcd and ohci_hcd, not after
…
…
…
[    0.576212] ata1: SATA link down (SStatus 0 SControl 300)
[    0.891638] ata2: SATA link up 1.5 Gbps (SStatus 113 SControl 300)
[    1.225155] tsc: Refined TSC clocksource calibration: 2194.925 MHz
[    1.229227] clocksource: tsc: mask: 0xffffffffffffffff max_cycles: 0x1fa3789b378, max_idle_ns: 440795248825 ns
[    1.230567] clocksource: Switched to clocksource tsc
[    1.385914] input: ImExPS/2 Generic Explorer Mouse as /devices/platform/i8042/serio1/input/input2
[    6.105271] ata2.00: qc timeout (cmd 0xa1)
[    6.109438] ata2.00: failed to IDENTIFY (I/O error, err_mask=0x4)
[    6.426687] ata2: SATA link up 1.5 Gbps (SStatus 113 SControl 300)
[   16.558347] ata2.00: qc timeout (cmd 0xa1)
[   16.561918] ata2.00: failed to IDENTIFY (I/O error, err_mask=0x4)
[   16.875348] ata2: SATA link up 1.5 Gbps (SStatus 113 SControl 300)
[   47.491954] ata2.00: qc timeout (cmd 0xa1)
[   47.495436] ata2.00: failed to IDENTIFY (I/O error, err_mask=0x4)
[   47.812365] ata2: SATA link up 1.5 Gbps (SStatus 113 SControl 300)
[   48.124998] ata3: SATA link up 1.5 Gbps (SStatus 113 SControl 300)
[   53.251589] ata3.00: qc timeout (cmd 0xa1)
[   53.254159] ata3.00: failed to IDENTIFY (I/O error, err_mask=0x4)
[   53.569856] ata3: SATA link up 1.5 Gbps (SStatus 113 SControl 300)
[   63.705231] ata3.00: qc timeout (cmd 0xa1)
[   63.708698] ata3.00: failed to IDENTIFY (I/O error, err_mask=0x4)
[   64.022462] ata3: SATA link up 1.5 Gbps (SStatus 113 SControl 300)
[   95.278549] ata3.00: qc timeout (cmd 0xa1)
[   95.282332] ata3.00: failed to IDENTIFY (I/O error, err_mask=0x4)
[   95.598871] ata3: SATA link up 1.5 Gbps (SStatus 113 SControl 300)
[   95.915699] ata4: SATA link down (SStatus 0 SControl 300)
[   96.232837] ata5: SATA link down (SStatus 0 SControl 300)
[   96.549342] ata6: SATA link down (SStatus 0 SControl 300)

#and later if sr0 is a SATA, but not if an IDE 
[  107.976002] sr 7:0:1:0 [sr0] GET_EVENT and TUR disagree continuously, suppress GET_EVENT events 

A fast comparison with other kernel configs indicate that

CONFIG_SATA_ZPODD
CONFIG_SATA_DWC and
CONFIG_SATA_DWC_OLD_DMA

are not set.
would be nice if added in the next iteration

THX

retiredt00 wrote: Sun Jan 09, 2022 9:30 am

@ BarryK

Thank you
Would by handy of the BOOT_SPECS parameters could be passed as kernel line arguments, as it is done for DEV location, so the initrd exercise could be bypassed.

Actually, you can. I never tested it though. The 'init' script in the initrd will recognize these kernel parameters, any of which will override the internal BOOT_SPECS file:

boot_dev or boot_uuid
boot_dir
wkg_dev or wkg_uuid
wkg_dir

Hi @BarryK
Two small problems for the moment. I have a window that opens when starting EasyOS (sound card assistant). Looks like my choice has not been made. For Chrome, after an update request (version 95 to version 97), I have two Chrome icons in the menu. The one identifying Google Chrome does not work.

Good work!!!

Airdale wrote: Sun Jan 09, 2022 2:48 am

BarryK wrote: Sun Jan 09, 2022 2:37 am

I'll checkout that cups add new printer problem.

I installed Avidemux from the package manager. This is a simple, but very nice, video editor. I discovered that it requires one extra dependency, 'fdk-aac' --this is also installable from the package manager.

Avidemux is a PET package, and when I created it, I left out that dep. from the list of deps. Have now fixed the PET, but not yet uploaded it.

Have done a bit of playing with LiVES, still very buggy, despite the author's recent fixes. I have just now reported a bug to his github site.

Alfonso sent an email, OBS Studio doesn't work for him, that incompatible-gpu error message at startup. Perhaps we need to go to their project site, complain that OBS won't work on so many computers. Will wait for more feedback from testers first. I have so far tested on 3 computers; 2 work, 1 fail.

Of note, OBS starts up without issues on my Dell Vostro laptop.

Concerning the CUPS problem, I think that I confirmed that it is a permissions problem of some type.

Since EasyOS now comes with HPLIP, I tried to install my printer with the command hp-setup.

That installed the printer without difficulty. However, it did not actually print.

When I opened CUPS it showed the printer had been installed. However, the following errors were shown.

capture17387.png

So I set permissions for both /usr/libexec/cups/backend and /usr/libexec/cups/filter to permissions 555 recursively.
Now CUPS works perfectly and I can add printers and print

It seems CUPS does not like everyone to have write access. By default it looks like the permissions for these folders were 777.

May be something to look into for future versions of Easy, though. I would think this problem would come up for others as well.

Thanks!

BarryK wrote: Sun Jan 09, 2022 11:00 am

retiredt00 wrote: Sun Jan 09, 2022 9:30 am

@ BarryK

Thank you
Would by handy of the BOOT_SPECS parameters could be passed as kernel line arguments, as it is done for DEV location, so the initrd exercise could be bypassed.

Actually, you can. I never tested it though. The 'init' script in the initrd will recognize these kernel parameters, any of which will override the internal BOOT_SPECS file:

boot_dev or boot_uuid
boot_dir
wkg_dev or wkg_uuid
wkg_dir

It works!
THX

Another issue while running in the VM that simulates connection as a wired one with automatic IP assignment. The menu network applet show no connection while IPinfo and browser say otherwise.
More important there is really no network connection when the browser, the terminal or dunfell is run in container.

retiredt00 wrote: Sun Jan 09, 2022 3:47 pm

BarryK wrote: Sun Jan 09, 2022 11:00 am

Actually, you can. I never tested it though. The 'init' script in the initrd will recognize these kernel parameters, any of which will override the internal BOOT_SPECS file:

boot_dev or boot_uuid
boot_dir
wkg_dev or wkg_uuid
wkg_dir

It works!
THX

retiredt00,
rEFInd? would you be kind enough to post your modified boot stanzas?

williwaw wrote: Sun Jan 09, 2022 5:42 pm

retiredt00 wrote: Sun Jan 09, 2022 3:47 pm

BarryK wrote: Sun Jan 09, 2022 11:00 am

Actually, you can. I never tested it though. The 'init' script in the initrd will recognize these kernel parameters, any of which will override the internal BOOT_SPECS file:

boot_dev or boot_uuid
boot_dir
wkg_dev or wkg_uuid
wkg_dir

It works!
THX

retiredt00,
rEFInd? would you be kind enough to post your modified boot stanzas?

I actually use grub4dos, in parallel with the other puppies.
So first you have to know the UUID of the partitions that have the boot files and the work dirs (use 'blkid' to identify them) as well as the folders the may be in.
In my case boot files are at the root of a small drive and work dir is in the folder “Easy” in another drive so the grub4dos entry looks like this

title Easy3.2 (sdb1)
  find --set-root uuid () 8e9b1091-616b-49c5-bdb2-5213766988dc
  kernel /vmlinuz  rw  vga=792 boot_uuid=8f9b1091-626b-49c5-bdc2-5219766988dc boot_dir=/ wkg_uuid=ad2e0c34-3466-4fba-9c56-217aab2cfb70 wkg_dir=/Easy
  initrd /initrd

In rEFInd all this should go in the options line.
Something like this I guess in my setup (where “test2” is the volume that has initrd and vmlinuz or use the GUID number in a gpt disk)

menuentry “Easy3.2” {
    volume test2
    loader /vmlinuz
    initrd /initrd
    options "rw  vga=792 boot_uuid=8f9b1091-626b-49c5-bdc2-5219766988dc boot_dir=/ wkg_uuid=ad2e0c34-3466-4fba-9c56-217aab2cfb70 wkg_dir=/Easy"
}

Airdale wrote: Sun Jan 09, 2022 2:16 pm

Airdale wrote: Sun Jan 09, 2022 2:48 am

BarryK wrote: Sun Jan 09, 2022 2:37 am

I'll checkout that cups add new printer problem.

I installed Avidemux from the package manager. This is a simple, but very nice, video editor. I discovered that it requires one extra dependency, 'fdk-aac' --this is also installable from the package manager.

Avidemux is a PET package, and when I created it, I left out that dep. from the list of deps. Have now fixed the PET, but not yet uploaded it.

Have done a bit of playing with LiVES, still very buggy, despite the author's recent fixes. I have just now reported a bug to his github site.

Alfonso sent an email, OBS Studio doesn't work for him, that incompatible-gpu error message at startup. Perhaps we need to go to their project site, complain that OBS won't work on so many computers. Will wait for more feedback from testers first. I have so far tested on 3 computers; 2 work, 1 fail.

Of note, OBS starts up without issues on my Dell Vostro laptop.

Concerning the CUPS problem, I think that I confirmed that it is a permissions problem of some type.

Since EasyOS now comes with HPLIP, I tried to install my printer with the command hp-setup.

That installed the printer without difficulty. However, it did not actually print.

When I opened CUPS it showed the printer had been installed. However, the following errors were shown.

capture17387.png

So I set permissions for both /usr/libexec/cups/backend and /usr/libexec/cups/filter to permissions 555 recursively.
Now CUPS works perfectly and I can add printers and print

It seems CUPS does not like everyone to have write access. By default it looks like the permissions for these folders were 777.

May be something to look into for future versions of Easy, though. I would think this problem would come up for others as well.

Thanks!

Ha ha, I just came online to suggest a workaround, while I figured out the permissions problem. Edit /usr/sbin/cups_shell and insert this at line 33:

USEBROWSER="seamonkey.bin --no-remote"

...that will run SM as user 'root' and the "--no-remote" will prevent a new tab being started from the "www" on the desktop -- that will start SM as a separate window running as user 'seamonkey'.

Interesting that this bug has flown under the radar for awhile, ever since SM moved to running as user 'seamonkey'. I never noticed, because I was only testing printing from my Lenovo after doing an update, and printing was already setup. Only printed from the Evince pdf viewer.

retiredt00 wrote: Sun Jan 09, 2022 6:31 pm

Code: Select all

menuentry “Easy3.2” {
    volume test2
    loader /vmlinuz
    initrd /initrd
    options "rw  vga=792 boot_uuid=8f9b1091-626b-49c5-bdc2-5219766988dc boot_dir=/ wkg_uuid=ad2e0c34-3466-4fba-9c56-217aab2cfb70 wkg_dir=/Easy"
}

You can also have boot_label and wkg_label:

boot_dev or boot_uuid or boot_label
boot_dir
wkg_dev or wkg_uuid or wkg_label
wkg_dir

Where the labels are got from 'blkid'.
Note that boot_dev and wkg_dev would be something like "sda8" without the "/dev/".

CUPS setup has been overhauled!

https://bkhome.org/news/202201/cups-web ... fixed.html

@BarryK Yes, I know, I am late.

I looked into CUPS printing - using my 'old' Samsung SCX-3405.
I used 'printer-driver-splix' and 'libjbig.so.0' to make this printer happy.

The reason I went this way is "portability" to 'arm64'.

'printer-driver-splix' is a debian package (Splix for amd x86-64), 44K xz,
https://packages.debian.org/bullseye/pr ... iver-splix
https://packages.debian.org/bullseye/am ... x/download

'libjbig.so.0' is a debian package (libjbig0 for amd x86-64), 31K xz,
https://packages.debian.org/bullseye/libjbig0
https://packages.debian.org/bullseye/am ... 0/download

The available printers are listed at /usr/share/cups/drv.

Rem: Samsung printers using 'rastertospl' can be ported by the user via /usr/share/cups/drv/splix-samsung.drv

Portability
Using 'splix' and 'libjbig0' on the Raspberry Pi-4 (Dunfell-2.8.3) works great. There is only one point to watch out for.
libjbig.so.0 will be placed at /usr/lib/aarch64-linux-gnu. You'll need to create a symlink to /usr/lib or /usr/lib/libjbig.so.0 to make everything work.

libjbig0 for arm64, 28K xz
https://packages.debian.org/bullseye/libjbig0
https://packages.debian.org/bullseye/ar ... 0/download

Splix for arm64, 43K xz
https://packages.debian.org/bullseye/pr ... iver-splix
https://packages.debian.org/bullseye/ar ... x/download

BarryK wrote: Tue Jan 11, 2022 2:27 am

CUPS setup has been overhauled!

https://bkhome.org/news/202201/cups-web ... fixed.html

Cups broke between 3.1.19 and 3.2?
setting up my printer in 3.1.19 worked fine, as it has for a while now since the foomatic drivers were added a while back.

williwaw wrote: Wed Jan 12, 2022 3:16 am

BarryK wrote: Tue Jan 11, 2022 2:27 am

CUPS setup has been overhauled!

https://bkhome.org/news/202201/cups-web ... fixed.html

Cups broke between 3.1.19 and 3.2?
setting up my printer in 3.1.19 worked fine, as it has for a while now since the foomatic drivers were added a while back.

Yes, you are correct, cups works, it didn't need to be fixed. I am in 3.2 right now, and brought up http://127.0.0.1:631 in Firefox, which is running as user 'firefox', no problem. So why the file permissions problem reported earlier in this thread?

Ditto works in SM running as user 'seamonkey'. Only looked at the UI, didn't try to setup a printer.

Anyway, the "bare window" for the cups UI is nice.

BarryK wrote: Thu Jan 13, 2022 3:41 pm

williwaw wrote: Wed Jan 12, 2022 3:16 am

BarryK wrote: Tue Jan 11, 2022 2:27 am

CUPS setup has been overhauled!

https://bkhome.org/news/202201/cups-web ... fixed.html

Cups broke between 3.1.19 and 3.2?
setting up my printer in 3.1.19 worked fine, as it has for a while now since the foomatic drivers were added a while back.

Yes, you are correct, cups works, it didn't need to be fixed. I am in 3.2 right now, and brought up http://127.0.0.1:631 in Firefox, which is running as user 'firefox', no problem. So why the file permissions problem reported earlier in this thread?

Ditto works in SM running as user 'seamonkey'. Only looked at the UI, didn't try to setup a printer.

Anyway, the "bare window" for the cups UI is nice.

I don't really know.
CUPS didn't work for me at all after the upgrade.
Changing permissions on the folders I listed above fixed it for me.
But glad it seems to be working for everybody else.
Strange.