Debian 11 bullseye - will it work on 12th gen core i3-12100?

[R. Ramesh]

Hi,

  My age old intel NUC with pentium N3700 is sick and needs 
replacement. I am interested in building one based on core i3-12100 cpu. 
It has intel UHD 730 per my quick check. Is this CPU/GPU combo fully 
supported in debian 11/bullseye? I am looking to run the current setup 
as is on new hardware (simply swap the disk from NUC to new build)


Debian 11 bullseye with default kernel/packages for most part.
Grup_pc legacy BIOS boot.
X org with intel driver (I think) that is currently running on Pentium 
N3700/NUC.

mythtv 31/fixes from debian multimedia (I think)

Do you see any issues I should take care of before transfering the disk 
to the new system when I build. Let me know if I should go for older 
hardware to avoid any hiccups.


Regards
Ramesh

Which nvidia driver? xset +dpms issues. Is my video card dead/broken?

[R. Ramesh]

Hi,

  A few weeks ago after my usual monthly update/upgrade, xset +dpms 
stopped working. Using dpms works as expected initially, but Xorg hangs 
after about 15min in to power save state. Everything else except the 
monitor works. I can ssh and reboot. Since then I have been disabling dpms.


1. I have debian 11 bullseye installed.
2. Xorg tells me that I have this card: NVIDIA GPU GeForce GT 1030
   (GP108-A) at PCI:9:0:0 (GPU-0)
3.   'apt search' shows nvidia-vdpau-driver (version 460.91.03-1) is
   installed. Is this the correct driver? There are so many of them and
   I want to use the correct one. I see several other drivers also.

p  mir-graphics-drivers-nvidia - Mir Display Server - Nvidia driver 
metapackage
p  mir-graphics-drivers-nvidia:i386 - Mir Display Server - Nvidia 
driver metapackage

p  nvidia-driver - NVIDIA metapackage
p  nvidia-driver-bin - NVIDIA driver support binaries
p  nvidia-driver-libs - NVIDIA metapackage (OpenGL/GLX/EGL/GLES libraries)
p  nvidia-driver-libs:i386 - NVIDIA metapackage (OpenGL/GLX/EGL/GLES 
libraries)

p  nvidia-legacy-390xx-driver - NVIDIA metapackage (390xx legacy version)
p  nvidia-legacy-390xx-driver:i386 - NVIDIA metapackage (390xx legacy 
version)
p  nvidia-legacy-390xx-driver-bin - NVIDIA driver support binaries 
(390xx legacy version)
p  nvidia-legacy-390xx-driver-bin:i386 - NVIDIA driver support 
binaries (390xx legacy version)
p  nvidia-legacy-390xx-driver-libs - NVIDIA metapackage 
(OpenGL/GLX/EGL/GLES libraries) (390xx legacy version)
p  nvidia-legacy-390xx-driver-libs:i386 - NVIDIA metapackage 
(OpenGL/GLX/EGL/GLES libraries) (390xx legacy version)
p  nvidia-legacy-390xx-driver-libs-i386:i386 - NVIDIA metapackage 
(OpenGL/GLX/EGL/GLES 32-bit libraries) (390xx legacy)
p  nvidia-legacy-390xx-driver-libs-nonglvnd - NVIDIA metapackage 
(non-GLVND OpenGL/GLX/EGL/GLES libraries) (390xx legacy)
p  nvidia-legacy-390xx-driver-libs-nonglvnd:i386 - NVIDIA metapackage 
(non-GLVND OpenGL/GLX/EGL/GLES libraries) (390xx legacy)
p  nvidia-legacy-390xx-driver-libs-nonglvnd-i386:i386 - NVIDIA 
metapackage (non-GLVND OpenGL/EGL etc. 32-bit libraries) (390xx legacy)
p  nvidia-legacy-390xx-vdpau-driver - Video Decode and Presentation 
API for Unix - NVIDIA driver (390xx legacy)
p  nvidia-legacy-390xx-vdpau-driver:i386 - Video Decode and 
Presentation API for Unix - NVIDIA driver (390xx legacy)

p  nvidia-tesla-418-driver - NVIDIA metapackage (Tesla 418 version)
p  nvidia-tesla-418-driver-bin - NVIDIA driver support binaries (Tesla 
418 version)
p  nvidia-tesla-418-driver-libs - NVIDIA metapackage 
(OpenGL/GLX/EGL/GLES libraries) (Tesla 418 version)
p  nvidia-tesla-418-driver-libs:i386 - NVIDIA metapackage 
(OpenGL/GLX/EGL/GLES libraries) (Tesla 418 version)
p  nvidia-tesla-418-driver-libs-nonglvnd - NVIDIA metapackage 
(non-GLVND OpenGL/GLX/EGL/GLES libraries) (Tesla 418)
p  nvidia-tesla-418-driver-libs-nonglvnd:i386 - NVIDIA metapackage 
(non-GLVND OpenGL/GLX/EGL/GLES libraries) (Tesla 418)
p  nvidia-tesla-418-vdpau-driver - Video Decode and Presentation API 
for Unix - NVIDIA driver (Tesla 418)
p  nvidia-tesla-418-vdpau-driver:i386 - Video Decode and Presentation 
API for Unix - NVIDIA driver (Tesla 418)

p  nvidia-tesla-450-driver - NVIDIA metapackage (Tesla 450 version)
p  nvidia-tesla-450-driver-bin - NVIDIA driver support binaries (Tesla 
450 version)
p  nvidia-tesla-450-driver-libs - NVIDIA metapackage 
(OpenGL/GLX/EGL/GLES libraries) (Tesla 450 version)
p  nvidia-tesla-450-driver-libs:i386 - NVIDIA metapackage 
(OpenGL/GLX/EGL/GLES libraries) (Tesla 450 version)
p  nvidia-tesla-450-vdpau-driver - Video Decode and Presentation API 
for Unix - NVIDIA driver (Tesla 450)
p  nvidia-tesla-450-vdpau-driver:i386 - Video Decode and Presentation 
API for Unix - NVIDIA driver (Tesla 450)

p  nvidia-tesla-460-driver - NVIDIA metapackage (Tesla 460 version)
p  nvidia-tesla-460-driver-bin - NVIDIA driver support binaries (Tesla 
460 version)
p  nvidia-tesla-460-driver-libs - NVIDIA metapackage 
(OpenGL/GLX/EGL/GLES libraries) (Tesla 460 version)
p  nvidia-tesla-460-driver-libs:i386 - NVIDIA metapackage 
(OpenGL/GLX/EGL/GLES libraries) (Tesla 460 version)
p  nvidia-tesla-460-vdpau-driver - Video Decode and Presentation API 
for Unix - NVIDIA driver (Tesla 460)
p  nvidia-tesla-460-vdpau-driver:i386 - Video Decode and Presentation 
API for Unix - NVIDIA driver (Tesla 460)
p  nvidia-vdpau-driver:i386 - Video Decode and Presentation API for 
Unix - NVIDIA driver


Do I have a better driver that I should try? Is my video card gone bad?

Regards
Ramesh

Re: Cannot install vlc on bullseye

[R. Ramesh]

The best expert opinion is fairly well summed up on the Debian wiki at
https://wiki.debian.org/DontBreakDebian  
  . The advice given above is fairly
blunt and matter of fact because the users and developers on this list have
seen this sort of thing lots of times before.

> Regards
> Ramesh
> 


With every good wish, as ever,

Andy Cater



Thanks for everyone's help. I systematically went through every package 
and moved them to debian version if apt allowed me to do so. At the end 
I only have acroread from dmo and everything else is from debian. I 
removed mythtv also to get to this point. Since this is my desktop, I do 
not need to install mythtv and will keep it as close to debian as possible.


I do not have the same freedom for three other (debian based) machines I 
have (2 frontend and one backend) and it is going to be a bit of 
struggle to get it right. However, fixing my desktop helped me to 
understand dependencies better.  So... hopefully, I will find the right 
combo of packages to keep it close enough to debian.


Ramesh

Re: Cannot install vlc on bullseye

[R. Ramesh]

On 10/16/21 2:20 PM, Dan Ritter wrote:

R. Ramesh wrote:

from deb-multimedia to make things work. I have kept it since. I have no
desire to install older packages just because it is straight from debian vs.
deb-multimedia. I expect those two repositories to be already aware of
situation like mine and therefore handle these without my preferences.d
entries.  So, I am perfectly ok to remove my preferences.d/multimedia

Here is the part that may have escaped you:

deb-multimedia is not a Debian project, and does not coordinate
with Debian. It can break your system. It did break your system.

(It is relatively safe to use if and only if you know what you
are doing with pinning. Copying from someone on the Internet
without understanding does not qualify.)
Understood. Please also understand that we are users and not experts. I 
understand unix/linux and general SW principles. I will use debian as I 
trust this more than any other distribution. That does not mean I can 
learn everything about it. I will have no choice but to go with help 
from communities. I understand the cost.


I would recommend finding all the packages on your system with a
dmo source, removing them, removing the dmo apt repository,
fixing your system, and then... carefully consider whether you
want to re-add the dmo apt repository at all.
This is not acceptable to me. I will live broken system than a useless 
system. I need what I have and I cannot simply delete them for the sake 
making something pristine. Please understand we are users. No point in 
having a system that does nothing. Apart from mythtv, I use browser. So, 
if I take out dmo then I might as well install Win10.


Don't think I hate deb-multimedia. I have that repo enabled...
on one system. Not on a system that I depend on functioning all
the time.

-dsr-
Thanks for your comments. I understand where you come from. I have no 
hard feelings when someone tells me I am dumb. That is why I used 
debian-user. If I do not get any help, I will find a way as I have done 
that many times.  In fact, I was going to wipe my system and reinstall 
from scratch and add dmo as that is easier.


I was a linux user in 1991 and remained one. I am not going to be afraid 
to find issues and resolve them. I just want to try to get expert 
opinion first because I think that is the best approach.


Regards
Ramesh

Re: Cannot install vlc on bullseye

[R. Ramesh]

On 10/16/21 12:43 AM, piorunz wrote:

On 16/10/2021 06:32, R. Ramesh wrote:


I was suspecting this, but did not know how to go about finding versions
available vs. what is installed. As you point out my version of
libpostproc55 is from debian whereas I need the newer version in
deb-multimedia. I forced the version
with this

sudo aptitude install libpostproc55=10:4.4-dmo4+deb11u2

Still could not install vlc because libswscale5 was also older version
and needed to be upgraded

sudo aptitude install libswscale5=10:4.4-dmo4+deb11u2

After that vlc installs fine.

Let me know if I did anything wrong in the above.

Thanks for the help.

Regards
Ramesh


You are welcome :)
There is one more thing you need to fix.
By using this method, you will never get updates of packages installed
manually. If there is security update in Debian, it will get ignored,
because you have newer version already. And if multimedia releases new
version, you won't get that installed either, because you enforce lower
pin-priority.

--
With kindest regards, Piotr.

⢀⣴⠾⠻⢶⣦⠀
⣾⠁⢠⠒⠀⣿⡁ Debian - The universal operating system
⢿⡄⠘⠷⠚⠋⠀ https://www.debian.org
⠈⠳⣄


What is the fix I need to do? Remove my preferences.d/multimedia for 
deb-multimedia?


My apologies, I am not very good with apt and dependencies and this may 
be something straightforward. Can you give me more detail on what I need 
to fix?


BTW, I did not have preferences.d/multimedia at all to start with. 
However, there was an issue with mythtv installation that I could not 
explain. At that time someone (here) told me that I need to pin priority 
of packages from deb-multimedia to make things work. I have kept it 
since. I have no desire to install older packages just because it is 
straight from debian vs. deb-multimedia. I expect those two repositories 
to be already aware of situation like mine and therefore handle these 
without my preferences.d entries.  So, I am perfectly ok to remove my 
preferences.d/multimedia


Also did my install of specific version made those packages to be held 
in that version? My report from apt-mark seem to show a lot of packages 
to be installed manually even though I only installed two packages of 
specific versions. May be I am not using right flags for apt-mark to get 
my info.


How do I know my specific install made an auto installed package to be 
marked as manual? More importantly, is is possible to manually fix 
dependencies like I did, but still retain all flags as if they are 
installed normally by apt/aptitude?


Thanks again for your patience in dealing with my novice questions.

Regards
Ramesh

Re: Cannot install vlc on bullseye

[R. Ramesh]

On 16/10/2021 04:52, R. Ramesh wrote:

The following packages have unmet dependencies:
 libpostproc55 : Depends: libavutil56 (= 7:4.3.2-0+deb11u2) but
10:4.4-dmo4+deb11u2 is to be installed
 libswscale5 : Depends: libavutil56 (= 7:4.3.2-0+deb11u2) but
10:4.4-dmo4+deb11u2 is to be installed
E: Unable to correct problems, you have held broken packages.
I cannot understand why libpostproc55 will need a older version of
libavutil56.

libpostproc55 is from debian repo. It wants older, debian versions of
its deps. You want to force newer ones from multimedia repo.

Solution: Delete debian multimedia repo. That way you will have matching
versions for all packages. Or fix your pinned priorities, yours probably
don't work.

Verify this by:
apt-cache policy libavutil56

On my computer result is:
$ apt-cache policy libavutil56
libavutil56:
  Installed: 7:4.3.2-0+deb11u2
  Candidate: 7:4.3.2-0+deb11u2
  Version table:
 *** 7:4.3.2-0+deb11u2 500
    500 http://deb.debian.org/debian bullseye/main amd64 Packages
    100 /var/lib/dpkg/status

Your candidate will be most likely 10:4.4-dmo4+deb11u2, because you want
to force version from multimedia repo.

--
With kindest regards, Piotr.


I was suspecting this, but did not know how to go about finding versions 
available vs. what is installed. As you point out my version of 
libpostproc55 is from debian whereas I need the newer version in 
deb-multimedia. I forced the version

with this

sudo aptitude install libpostproc55=10:4.4-dmo4+deb11u2

Still could not install vlc because libswscale5 was also older version 
and needed to be upgraded


sudo aptitude install libswscale5=10:4.4-dmo4+deb11u2

After that vlc installs fine.

Let me know if I did anything wrong in the above.

Thanks for the help.

Regards
Ramesh

Cannot install vlc on bullseye

[R. Ramesh]

Hi,

  I am trying to install vlc and I get this error

n-lata [rramesh] 284 > apt -s install vlc
NOTE: This is only a simulation!
  apt needs root privileges for real execution.
  Keep also in mind that locking is deactivated,
  so don't depend on the relevance to the real current situation!
Reading package lists... Done
Building dependency tree... Done
Reading state information... Done
Some packages could not be installed. This may mean that you have
requested an impossible situation or if you are using the unstable
distribution that some required packages have not yet been created
or been moved out of Incoming.
The following information may help to resolve the situation:

The following packages have unmet dependencies:
 libpostproc55 : Depends: libavutil56 (= 7:4.3.2-0+deb11u2) but 
10:4.4-dmo4+deb11u2 is to be installed
 libswscale5 : Depends: libavutil56 (= 7:4.3.2-0+deb11u2) but 
10:4.4-dmo4+deb11u2 is to be installed

E: Unable to correct problems, you have held broken packages.
I cannot understand why libpostproc55 will need a older version of 
libavutil56.  My sources.list is this



deb http://ftp.us.debian.org/debian/ bullseye main contrib non-free
deb-src http://ftp.us.debian.org/debian/ bullseye main

deb http://www.deb-multimedia.org bullseye main non-free
deb http://www.deb-multimedia.org bullseye-backports main

deb http://security.debian.org/debian-security 
bullseye-security/updates main
deb-src http://security.debian.org/debian-security 
bullseye-security/updates main


# bullseye-updates
deb http://ftp.us.debian.org/debian/ bullseye-updates main
deb-src http://ftp.us.debian.org/debian/ bullseye-updates main


I also have these two rules in preferences.d/multimedia as in the past I 
had issues with same packages in both deb-multimedia and debian regular 
when I was told these rules will help.

Package: *
Pin: release o=Unofficial Multimedia Packages,n=bullseye
Pin-Priority: 332

Package: *
Pin: release o=Unofficial Multimedia Packages,n=bullseye-backports
Pin-Priority: 331


Any help appreciated.

Regards
Ramesh

Re: Minisforum X35G hangs on first boot. [Solved]

[R. Ramesh]

   a question.  are you installing using graphics mode or
not?  it not then the next bit might make a difference.

   all i can think of is that the hardware may be so new
that buster may not work for some hardware reason.

   i suggest trying a netinst iso from the debian installer
for the upcoming release.  there are versions that are
fairly recent (daily and/or weekly), but i don't check
often as to which actually run.

   if you are installing with graphics working during the
install but then for some reason it isn't working any
more, well that doesn't make sense.

   try not installing desktop at all to see if it will boot
that ways.  you can always add a desktop later.



Songbird, thanks for your help. I found out some more. Linux was booting 
after all. It appeared to be hung because I did not get a login prompt. 
I booted sysrescuecd and began inspecting logs and then hit ctrl-alt-1 
and voila, I get login prompt. After that some more digging showed 
lightdm is failing and that is why I was not getting login prompt. It 
failed because Xorg failed and Xorg failed because /dev/dri is missing. 
It appears that 10 gen intel cpu graphics is not supported in 4.19 
kernel. I added buster-backports and updated and installed 5.10 kernel. 
Everything works now. I am good.


I also started another thread on debian-user on above X server issue. I 
closed that also with this message.


Regards
Ramesh

Re: My X server not working. I think it cannot find /dev/dri/card0 [Solved]

[R. Ramesh]

On 3/26/21 4:50 PM, R. Ramesh wrote:

Hi,

  I installed debian buster on minisforum X35G with Core i3-1005G1. 
Install went through fine, but lightdm does not start up. Upon 
inspecting logs, I found out that X server does not startup for 
lightdm to use. Looking /var/log/Xorg.0.log, I see these error messages


   yoda-mini [rramesh] 43 > fgrep EE /var/log/Xorg.0.log
        (WW) warning, (EE) error, (NI) not implemented, (??) unknown.
   [  3134.273] (EE) open /dev/dri/card0: No such file or directory
   [  3134.273] (EE) open /dev/dri/card0: No such file or directory
   [  3134.274] (EE) Unable to find a valid framebuffer device
   [  3134.274] (EE) Screen 0 deleted because of no matching config
   section.
   [  3134.274] (EE) Screen 0 deleted because of no matching config
   section.
   [  3134.274] (EE)
   [  3134.274] (EE) Cannot run in framebuffer mode. Please specify
   busIDs    for all framebuffer devices
   [  3134.274] (EE)
   [  3134.274] (EE)
   [  3134.274] (EE) Please also check the log file at
   "/var/log/Xorg.0.log" for additional information.
   [  3134.274] (EE)
   [  3134.278] (EE) Server terminated with error (1). Closing log file.

Sure enough I cannot find /dev/dri. I double checked and udev is 
mounted on /dev


   yoda-mini [rramesh] 44 > df -h
   Filesystem  Size  Used Avail Use% Mounted on
   udev    7.7G 0  7.7G   0% /dev
   tmpfs   1.6G   18M  1.6G   2% /run
   /dev/nvme0n1p5  341G  3.6G  320G   2% /
   tmpfs   7.8G 0  7.8G   0% /dev/shm
   tmpfs   5.0M  4.0K  5.0M   1% /run/lock
   tmpfs   7.8G 0  7.8G   0% /sys/fs/cgroup
   /dev/nvme0n1p1   96M   51M   46M  53% /boot/efi
   tmpfs   1.6G 0  1.6G   0% /run/user/1000

My lspci shows this

   00:00.0 Host bridge: Intel Corporation Device 8a02 (rev 03)
   00:02.0 VGA compatible controller: Intel Corporation Device 8a56
   (rev 07)
   00:07.0 PCI bridge: Intel Corporation Device 8a1d (rev 03)
   00:0d.0 USB controller: Intel Corporation Device 8a13 (rev 03)
   00:0d.2 System peripheral: Intel Corporation Device 8a17 (rev 03)
   00:14.0 USB controller: Intel Corporation Ice Lake-LP USB 3.1 xHCI
   Host Controller (rev 30)
   00:14.2 RAM memory: Intel Corporation Device 34ef (rev 30)
   00:15.0 Serial bus controller [0c80]: Intel Corporation Ice Lake-LP
   Serial IO I2C Controller (rev 30)
   00:16.0 Communication controller: Intel Corporation Device 34e0 
(rev 30)

   00:17.0 RAID bus controller: Intel Corporation 82801 Mobile SATA
   Controller [RAID mode] (rev 30)
   00:1c.0 PCI bridge: Intel Corporation Device 34ba (rev 30)
   00:1c.4 PCI bridge: Intel Corporation Ice Lake-LP PCI Express Root
   Port (rev 30)
   00:1d.0 PCI bridge: Intel Corporation Ice Lake-LP PCI Express Root
   Port (rev 30)
   00:1d.4 PCI bridge: Intel Corporation Device 34b4 (rev 30)
   00:1e.0 Communication controller: Intel Corporation Ice Lake-LP
   Serial IO UART Controller (rev 30)
   00:1e.3 Serial bus controller [0c80]: Intel Corporation Ice Lake-LP
   Serial IO SPI Controller (rev 30)
   00:1f.0 ISA bridge: Intel Corporation Ice Lake-LP LPC Controller
   (rev 30)
   00:1f.3 Audio device: Intel Corporation Device 34c8 (rev 30)
   00:1f.4 SMBus: Intel Corporation Ice Lake-LP SMBus Controller (rev 30)
   00:1f.5 Serial bus controller [0c80]: Intel Corporation Ice Lake-LP
   SPI Controller (rev 30)
   2c:00.0 Network controller: Intel Corporation Device 2723 (rev 1a)
   2d:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd.
   RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller (rev 15)
   2e:00.0 Non-Volatile memory controller: Phison Electronics
   Corporation Device 5013 (rev 01)
   2f:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd.
   RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller (rev 15)

I thought 00:02.0 is the device X is looking for, but for some reason 
its equivalent /dev/dri items are missing. Not sure how to help my X. 
Please advise.


Regards
Ramesh

I added "deb http://deb.debian.org/debian buster-backports main contrib 
non-free" to my sources.list and updated/installed 5.10 kernel and 
everything is fine per my quick check.


I think 10 generation intel cpu's graphics is not supported in 4.19 
kernel that is installed with stable per an article I read while googling.


Ramesh

My X server not working. I think it cannot find /dev/dri/card0

[R. Ramesh]

Hi,

  I installed debian buster on minisforum X35G with Core i3-1005G1. 
Install went through fine, but lightdm does not start up. Upon 
inspecting logs, I found out that X server does not startup for lightdm 
to use. Looking /var/log/Xorg.0.log, I see these error messages


   yoda-mini [rramesh] 43 > fgrep EE /var/log/Xorg.0.log
    (WW) warning, (EE) error, (NI) not implemented, (??) unknown.
   [  3134.273] (EE) open /dev/dri/card0: No such file or directory
   [  3134.273] (EE) open /dev/dri/card0: No such file or directory
   [  3134.274] (EE) Unable to find a valid framebuffer device
   [  3134.274] (EE) Screen 0 deleted because of no matching config
   section.
   [  3134.274] (EE) Screen 0 deleted because of no matching config
   section.
   [  3134.274] (EE)
   [  3134.274] (EE) Cannot run in framebuffer mode. Please specify
   busIDs    for all framebuffer devices
   [  3134.274] (EE)
   [  3134.274] (EE)
   [  3134.274] (EE) Please also check the log file at
   "/var/log/Xorg.0.log" for additional information.
   [  3134.274] (EE)
   [  3134.278] (EE) Server terminated with error (1). Closing log file.

Sure enough I cannot find /dev/dri. I double checked and udev is mounted 
on /dev


   yoda-mini [rramesh] 44 > df -h
   Filesystem  Size  Used Avail Use% Mounted on
   udev    7.7G 0  7.7G   0% /dev
   tmpfs   1.6G   18M  1.6G   2% /run
   /dev/nvme0n1p5  341G  3.6G  320G   2% /
   tmpfs   7.8G 0  7.8G   0% /dev/shm
   tmpfs   5.0M  4.0K  5.0M   1% /run/lock
   tmpfs   7.8G 0  7.8G   0% /sys/fs/cgroup
   /dev/nvme0n1p1   96M   51M   46M  53% /boot/efi
   tmpfs   1.6G 0  1.6G   0% /run/user/1000

My lspci shows this

   00:00.0 Host bridge: Intel Corporation Device 8a02 (rev 03)
   00:02.0 VGA compatible controller: Intel Corporation Device 8a56
   (rev 07)
   00:07.0 PCI bridge: Intel Corporation Device 8a1d (rev 03)
   00:0d.0 USB controller: Intel Corporation Device 8a13 (rev 03)
   00:0d.2 System peripheral: Intel Corporation Device 8a17 (rev 03)
   00:14.0 USB controller: Intel Corporation Ice Lake-LP USB 3.1 xHCI
   Host Controller (rev 30)
   00:14.2 RAM memory: Intel Corporation Device 34ef (rev 30)
   00:15.0 Serial bus controller [0c80]: Intel Corporation Ice Lake-LP
   Serial IO I2C Controller (rev 30)
   00:16.0 Communication controller: Intel Corporation Device 34e0 (rev 30)
   00:17.0 RAID bus controller: Intel Corporation 82801 Mobile SATA
   Controller [RAID mode] (rev 30)
   00:1c.0 PCI bridge: Intel Corporation Device 34ba (rev 30)
   00:1c.4 PCI bridge: Intel Corporation Ice Lake-LP PCI Express Root
   Port (rev 30)
   00:1d.0 PCI bridge: Intel Corporation Ice Lake-LP PCI Express Root
   Port (rev 30)
   00:1d.4 PCI bridge: Intel Corporation Device 34b4 (rev 30)
   00:1e.0 Communication controller: Intel Corporation Ice Lake-LP
   Serial IO UART Controller (rev 30)
   00:1e.3 Serial bus controller [0c80]: Intel Corporation Ice Lake-LP
   Serial IO SPI Controller (rev 30)
   00:1f.0 ISA bridge: Intel Corporation Ice Lake-LP LPC Controller
   (rev 30)
   00:1f.3 Audio device: Intel Corporation Device 34c8 (rev 30)
   00:1f.4 SMBus: Intel Corporation Ice Lake-LP SMBus Controller (rev 30)
   00:1f.5 Serial bus controller [0c80]: Intel Corporation Ice Lake-LP
   SPI Controller (rev 30)
   2c:00.0 Network controller: Intel Corporation Device 2723 (rev 1a)
   2d:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd.
   RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller (rev 15)
   2e:00.0 Non-Volatile memory controller: Phison Electronics
   Corporation Device 5013 (rev 01)
   2f:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd.
   RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller (rev 15)

I thought 00:02.0 is the device X is looking for, but for some reason 
its equivalent /dev/dri items are missing. Not sure how to help my X. 
Please advise.


Regards
Ramesh

Re: Minisforum X35G hangs on first boot.

[R. Ramesh]

On 3/25/21 11:02 PM, R. Ramesh wrote:

Hi,

  I am trying to install debian buster 
(debian-10.8.0-amd64-netinst.iso) on minisforum X35G. This comes 
installed with windows 10. It has  Core i3-1005G1 cpu 16GB memory and 
512 PCIe nvme SSD as far as I can tell.
Everything during the install goes fine (warnings about some realtec 
NIC firmware, but network is fine for install). However on the first 
reboot, it gets all the way where it complains about NIC firmware and 
hangs right there. The only thing I can do is to press power button 
and the kernel will get out of hang and shutdown and power off. 
Windows seem to dual boot fine.


  I do not know how to proceed further with debug. Any help is much 
appreciated.


Regards
Ramesh



I booted sysrescuecd and looked at couple of logs and figured out that 
my system is booting after all. However it fails on lightdm startup. I 
could switch to a different VT and get login prompt. This is what 
systemctl status lightdm shows


   lightdm.service - Light Display Manager
   Loaded: loaded (/lib/systemd/system/lightdm.service; indirect;
   vendor preset: enabled)
   Active: failed (Result: exit-code) since Fri 2021-03-26 15:56:26
   CDT; 1min 54s ago
 Docs: man:lightdm(1)
  Process: 932 ExecStartPre=/bin/sh -c [ "$(cat
   /etc/X11/default-display-manager 2>/dev/null)" = "/usr/sbin/lightdm"
   ] (code=exited, status=0/SUCCESS)
  Process: 934 ExecStart=/usr/sbin/lightdm (code=exited,
   status=1/FAILURE)
 Main PID: 934 (code=exited, status=1/FAILURE)

   Mar 26 15:56:26 yoda-mini systemd[1]: lightdm.service: Service
   RestartSec=100ms expired, scheduling restart.
   Mar 26 15:56:26 yoda-mini systemd[1]: lightdm.service: Scheduled
   restart job, restart counter is at 5.
   Mar 26 15:56:26 yoda-mini systemd[1]: Stopped Light Display Manager.
   Mar 26 15:56:26 yoda-mini systemd[1]: lightdm.service: Start request
   repeated too quickly.
   Mar 26 15:56:26 yoda-mini systemd[1]: lightdm.service: Failed with
   result 'exit-code'.
   Mar 26 15:56:26 yoda-mini systemd[1]: Failed to start Light Display
   Manager.
   Mar 26 15:56:26 yoda-mini systemd[1]: lightdm.service: Triggering
   OnFailure= dependencies.


Inspecting /var/log/lightdm/lightdm.log shows this:

   ...
   [+0.01s] DEBUG: Launching process 940: /usr/bin/X :0 -seat seat0
   -auth /var/run/lightdm/root/:0 -nolisten tcp vt7 -novtswitch
   [+0.01s] DEBUG: XServer 0: Waiting for ready signal from X server :0
   [+0.01s] DEBUG: Acquired bus name org.freedesktop.DisplayManager
   [+0.01s] DEBUG: Registering seat with bus path
   /org/freedesktop/DisplayManager/Seat0
   [+0.01s] WARNING: Error getting user list from
   org.freedesktop.Accounts:
   GDBus.Error:org.freedesktop.DBus.Error.ServiceUnknown: The name
   org.freedesktop.Accounts was not provided by any .service files
   [+0.01s] DEBUG: Loading user config from /etc/lightdm/users.conf
   [+0.01s] DEBUG: User rramesh added
   [+0.02s] DEBUG: Process 940 exited with return value 1
   [+0.02s] DEBUG: XServer 0: X server stopped
   [+0.02s] DEBUG: Releasing VT 7
   [+0.02s] DEBUG: XServer 0: Removing X server authority
   /var/run/lightdm/root/:0
   [+0.02s] DEBUG: Seat seat0: Display server stopped
   [+0.02s] DEBUG: Seat seat0: Stopping session
   [+0.02s] DEBUG: Seat seat0: Session stopped
   


Looking at /var/log/lightdm/x-0.log, we see

   X.Org X Server 1.20.4
   X Protocol Version 11, Revision 0
   Build Operating System: Linux 4.19.0-12-amd64 x86_64 Debian
   Current Operating System: Linux yoda-mini 4.19.0-14-amd64 #1 SMP
   Debian 4.19.171-2 (2021-01-30) x86_64
   Kernel command line: BOOT_IMAGE=/boot/vmlinuz-4.19.0-14-amd64
   root=UUID=38181352-3024-4f42-8bf9-3f0b889b4fc0 ro quiet
   Build Date: 01 December 2020  05:59:57PM
   xorg-server 2:1.20.4-1+deb10u2 (https://www.debian.org/support)
   Current version of pixman: 0.36.0
    Before reporting problems, check http://wiki.x.org
    to make sure that you have the latest version.
   Markers: (--) probed, (**) from config file, (==) default setting,
    (++) from command line, (!!) notice, (II) informational,
    (WW) warning, (EE) error, (NI) not implemented, (??) unknown.
   (==) Log file: "/var/log/Xorg.0.log", Time: Fri Mar 26 15:56:26 2021
   (==) Using system config directory "/usr/share/X11/xorg.conf.d"
   (EE)
   Fatal server error:
   (EE) Cannot run in framebuffer mode. Please specify busIDs for all
   framebuffer devices
   (EE)
   (EE)
   Please consult the The X.Org Foundation support
 at http://wiki.x.org
 for help.
   (EE) Please also check the log file at "/var/log/Xorg.0.log" for
   additional information.
   (EE)
   (EE) Server terminated with error (1). Closing log file.


Obviously, I did not invoke X and therefore fail to supply proper 
arguments. I tried reinstalling lightdm and xorg-xserver, but that did 
not work. I could wipe disk 

Minisforum X35G hangs on first boot.

[R. Ramesh]

Hi,

  I am trying to install debian buster 
(debian-10.8.0-amd64-netinst.iso) on minisforum X35G. This comes 
installed with windows 10. It has  Core i3-1005G1 cpu 16GB memory and 
512 PCIe nvme SSD as far as I can tell.
Everything during the install goes fine (warnings about some realtec NIC 
firmware, but network is fine for install). However on the first reboot, 
it gets all the way where it complains about NIC firmware and hangs 
right there. The only thing I can do is to press power button and the 
kernel will get out of hang and shutdown and power off. Windows seem to 
dual boot fine.


  I do not know how to proceed further with debug. Any help is much 
appreciated.


Regards
Ramesh

Re: Re: aptitude safe-upgrade vs apt-get upgrade.

[R. Ramesh]

So, if you don't pin down the priority of deb-multimedia, virtually every
audio- and video-related package on your system will be replaced with the
deb-multimedia version, which for the sake of stability is very likely a
bad idea.
So it is safer to lower the priority of deb-multimedia and that of
deb-multimedia backports even a bit more, so that the official debian
packages remain the default, deb-multimedia packages the first
alternative if you actually want/need a newer version or a version with
some extra features and deb-multimedia backports the last choice if you
really, really need that version.



Thanks for the detailed explanation. I will pin down multi-media and 
then backports to ensure order among the repositories.


Regards
Ramesh

Re: Raid 1 borked

[R. Ramesh]

Hi folks,

So we're setting up a small server with a pair of 1 TB hard 
diskssectioned into 5x100GB Raid 1 partition pairs for data, with 
400GB+reserved for future uses on each disk.I'm not sure what 
happened, we had the five pairs of disk partitions setup properly 
through the installer without problems. However, now theRaid 1 pairs 
are not mounted as separate partitions but do show up assubdirectories 
under /, ie /datab, and they do seem to work as part ofthe regular / 
filesystem. df -h does not show any md devices or sda/bdevices, 
neither does mount. (The system partitions are on an nvme ssd).lsblk 
reveals sda and sdb with sda[1-5] and sdb[1-5] but no md[0-5].blkid 
reveals that sda[1-5] and sdb[1-5] are still listed as

TYPE="linux_raid_member".

So first of all I'd like to be able to diagnose what's going on. 
Whatcommands should I use for that? And secondly, I'd like to get the 
raidarrays remounted as separate partitions. How to do 
that?Fortunately, there is no data to worry about. However, I'd rather 
notreinstall as we've put in a bit of work installing and 
configuringthings. I'd prefer not to loose that. Can someone help us out?

Thanks in advance,

Bill


Did you create the md raid1s after partitioning the disks?

Normally when you install mdadm or when you install the system from 
usb/.iso for the first time, the respective mds are assembled and 
appropriately set up if you have already created them.


If you added and partitioned the disk after the main system has been 
installed and running, you will have to create md raid1s and enable 
automatic assembly through /etc/mdadm.conf file. You may need to update 
your initrd also, but this I am not sure. To access and use the md 
raid1s as file systems, You also need to add appropriate fstab entries 
to mount them.


Hope I am not trivializing your issues.

Regards
Ramesh

Re: Re: aptitude safe-upgrade vs apt-get upgrade.

[R. Ramesh]

To resolve this, you might consider to create a file
like e.g. /etc/apt/preferences.d/multimedia .

Here the content of that file looks like:

Package: *
Pin: release o=Unofficial Multimedia Packages,n=buster
Pin-Priority: 332

Package: *
Pin: release o=Unofficial Multimedia Packages,n=buster-backports
Pin-Priority: 331


Michael,

  Thanks for explaining why the difference exists. However, I do not 
know what these two rules mean to upgrade decisions.  So, I ask the 
following based on sources.list lines.



with the respective entries in my sources.list:

debhttp://www.deb-multimedia.org    buster main 
non-free
debhttp://www.deb-multimedia.org    
buster-backports main non-free

I have these exact lines in my sources.list also. I thought we have 
backports so that we can get the newer version of packages. For example, 
buster multimedia has mythtv 0.30 and backports has mythtv 0.31 (the 
last time I checked). I have installed 0.31 in my system (using -t 
stable-backports, I think) as my frontend need to be compatible with 
backend that runs mythtv 0.31 (on xubuntu 20.04). In that case, would 
the above pinning rule prevent proper upgrade as it puts backports at a 
lower priority? (I assume that is what your pinning rules imply unless 
priorities are increased with lower numerical value)


In other words, should I stick to aptitude's decision?

Why priority 331 and 332? Why not some other two with same relationship? 
Sorry, if I asked a simple question. I will be more than happy to read 
documentation, if I will find it there.


Regards
Ramesh

Re: Re: aptitude safe-upgrade vs apt-get upgrade.

[R. Ramesh]

To begin with, which distribution is it? In general, with Stable, it
pretty much doesn't matter which tool is used. The kind of problems you
have indicate Unstable or Testing.

First, apt is pretty much apt-get, with different syntax and a few
extra features. Aptitude can generally do a better job of resolving
'difficult' dependencies, but if used with a great many packages (>100)
is likely to be very slow. With more than about 500 packages to deal
with, it may grind to a halt. Don't use it on an Unstable which hasn't
been upgraded for six months. Don't use it for a version upgrade of
Stable, unless the release notes for the upgrade explicitly recommend
it.

Synaptic, the GUI tool, is a front end to apt-get. All the apt tools
are a front end to dpkg, which does all the work but does no dependency
checking and is therefore not safe to be used directly.

All of the tools have a 'safe' mode which is guaranteed not to remove
packages. However, many upgrades do require the removal of some
packages, so the safe mode will only get you so far in these cases.

If you are using Stable, there should never be any real problem in
upgrading. With Testing or Unstable, problems do occur, and you need to
be willing to understand and solve them. Generally, if you keep the
system up to date regularly, aptitude will usually do a good job.

With Unstable or Testing, it is often the case that new versions of a
few packages from a large related set become available before the rest.
The new packages will be mostly incompatible with the existing set, so
upgrading them will cause the removal of some of the rest of the set,
along with other applications which depend on them. If you see that
half your applications are about to be removed, say 'no'. If you do
this with apt-get or apt in safe mode, then packages will be 'held
back'. Aptitude full-upgrade will usually give you a number of options,
with decreasing numbers of removals proposed. If you keep declining
options, it will eventually get to 'keep everything as it is', but you
may find one of the options worth doing.

Generally time will sort this problem out, as the rest of the set is
released. It is usually possible to upgrade some packages which do not
include the problem ones. This can be done with any of the tools, but
on a graphical system I prefer Synaptic. Other methods may involve a lot
of typing.
My bad, I did not think releases mattered to figure out the 
differentiation between the two. I am on current stable release

installed within the last month. Here is my /etc/os-release

myth2 [rramesh] 102 > cat /etc/os-release
PRETTY_NAME="Debian GNU/Linux 10 (buster)"
NAME="Debian GNU/Linux"
VERSION_ID="10"
VERSION="10 (buster)"
VERSION_CODENAME=buster
ID=debian
HOME_URL="https://www.debian.org/;
SUPPORT_URL="https://www.debian.org/support;
BUG_REPORT_URL="https://bugs.debian.org/;
Nothing fancy. Installed debian 10 from USB and added multi-media and 
installed mythfrontend. That is all I have done.

This is a NUC Pentium (N3700) box and not fancy at all.  Here is my kernel

myth2 [rramesh] 104 > uname -a
Linux myth2 4.19.0-11-amd64 #1 SMP Debian 4.19.146-1 (2020-09-17) 
x86_64 GNU/Linux
My apt-get/aptitude output showed clear differences between the two. So, 
I am not convinced about your claim that they should do the same thing 
on a stable release unless stable release itself was broken when 
installing with debian-10.6.0-amd64-netinst.iso.


Regards
Ramesh

Re: Re: Re: Trouble with upgrading debian 7 (wheezy) (solved)

[R. Ramesh]

The expired keys do complicate life but my understanding of the
rationale is that the limited key lifetime serves as a sort of
contract regarding the integrity of the files. Once a release has
been archived it does not fall under that promise from the project
any more and so the expired keys serve as a way to underscore that.
Without the key expiry, users may think that Debian has promised
to take care of the integrity of those files forever, but it hasn't.

I'm trying to remember what I had to do. It may have been:

# apt-get -o Acquire::Check-Valid-Until=false update

but I admit it could also have been me manually downloading the .deb
files from archive,debian.org and installing them with dpkg.

Cheers,
Andy

Andy,

  Thanks for the detailed message. My aim here is to document any steps 
needed for unfortunate or lazy users who are still with wheezy. People 
like you are the reason why I posted it so that each one will add a 
trickle to the solution. Hopefully, the next time someone does a search 
on the archives will find this and other related articles to help them 
get over upgrade hump. I feel happy that some of my goal is achieved and 
thank you very much for taking the time to respond.


  My only wish is apt is updated to say something about  the fact that 
this is unsupported and users are on their own, but still provide the 
download/install without we having to manually intervene. However, like 
I said before, this is free SW and whatever we get is a gift and no 
questions about that.


Best to you.

Regards
Ramesh

Re: Re: Trouble with upgrading debian 7 (wheezy) (solved)

[R. Ramesh]

That said, Wheezy is  *way*  too far away from Buster!  For example,  
"Update in Place" is likely to fail, because of SystemD changes.  
(Heck, I had trouble updating Stretch to Buster in Place, but some of 
that was User Error).


I did think about fresh install, but every method has its drawback. 
There are subtle changes that I could not get right in the past, so I 
chose upgrade path.


I thought name of the game with Debian is stability. If so, I really do 
not see why an upgrade should not work, if it worked in the past (I mean 
as soon as the next release was made) Nothing has changed in my system 
to break. It would have been the same had I tried to upgrade as soon as 
Jesse came. I can understand that I cannot find the archive servers, but 
that is not the problem. The keys expire on an archive that is supposed 
to be accessible.


Anyway I can complain, but that is not my aim. I am getting this free 
and anything and everything is acceptable in my book when free. It is I 
who should adapt. I was merely pointing out what I saw just to find out 
if anyone has seen that and resolved it. I saw wheezy to Jessy upgrade 
documents written at late as Sep 2019, so assumed that I am not too late 
to the game.  In any case, I did followup and mention what I did and how 
it worked. This is  in the hope that some one else will find it useful. 
Part of my intention in posting the original message was this.


Regards
Ramesh

Re: Trouble with upgrading debian 7 (wheezy) (solved)

[R. Ramesh]

My attempts to upgrade to Jesse (8) did not work at all and all the 
information on the net seem old and not current that Buster is out and 
jesse is already archived. Anyway, after looking around, I made ddrescue 
copy of the current install disk (only 64GB, so easy) first and then 
directly upgraded to Stretch (9).


 I had to iterate a couple of times looking to fix broken packages, but 
at the end, I had stretch working fine. The biggest trouble I had was 
dist-upgrade fail as it could not remove some packages. I had to force 
remove and reboot before dist-upgrade worked. After that it was very 
easy to upgrade to Buster. However there was some hickups with dbus 
connections and commands took a long time. A couple of reboots and force 
removal and reinstall of a couple of packages, everything worked (I think)


Regards
Ramesh

Trouble with upgrading debian 7 (wheezy)

[R. Ramesh]

I finally decided to move from debian 7 to 10. As a first step I wanted 
to upgrade to debian 8 (jesse)


I changed all ftp.us.debian.org part in /etc/apt/sources.list to 
archive.debian.org and tried aptitude update and got the following error
W: GPG error: http://archive.debian.org wheezy Release: The following 
signatures were invalid: KEYEXPIRED 1587841717 KEYEXPIRED 1587841717 
KEYEXPIRED 1587841717 KEYEXPIRED 1587841717 KEYEXPIRED 1587841717 
KEYEXPIRED 1557241909 The following signatures couldn't be verified 
because the public key is not available: NO_PUBKEY 7638D0442B90D010
W: GPG error: http://archive.debian.org wheezy-backports Release: The 
following signatures were invalid: KEYEXPIRED 1587841717 KEYEXPIRED 
1587841717 KEYEXPIRED 1587841717 KEYEXPIRED 1587841717 KEYEXPIRED 
1587841717 The following signatures couldn't be verified because the 
public key is not available: NO_PUBKEY 7638D0442B90D010


So I proceeded to update the keys. I did the following

sudo apt-key list | grep expired
pub 4096R/B98321F9 2010-08-07 [expired: 2017-08-05]
pub 4096R/46925553 2012-04-27 [expired: 2020-04-25]
pub 4096R/65FFB764 2012-05-08 [expired: 2019-05-07]


After that I did

foreach i (B98321F9 46925553 65FFB764)
apt-key adv --keyserver keys.gnupg.net --recv-keys $i
end


It updated the keys, but even after that I got the same error and the 
dates on the expired keys do not change anymore.


What am I missing?

Ramesh

Re: 64bit ext4 and kernel version compatibility

[R. Ramesh]

On 1/9/20 2:02 AM, Klaus Singvogel wrote:

R. Ramesh wrote:

I want to make sure
that my current kernel version does not have any limitation to support 64bit
ext4.

Please consult the Kernel Wiki regarding Ext4:

https://ext4.wiki.kernel.org/index.php/Main_Page

You will notice that Linux 2.6.28 was the first suppored kernel with ext4,
which was released a quiet long time ago (around 2009). As your
distribution is running 3.13.0-132-generic, the support of ext4 should be
no problem (never tested it to be 100% sure).

Another important information I was missing of: the machine architecture
of your processor; whether you're running a 64bit kernel or 32bit kernel.

"uname -m" will tell you, if you have 64bit, like in "x86_64", or not.

I'm very unsure, if your processor is able to address such files. PAE
extension (32 bit) of Intel architecture supports it, but not sure about
other archs, like Raspberry Pi v1 has.

Nevertheless, I stronlgy recommend to upgrade to 16.04 LTS or better, if
this machine connects or is connectable from the internet, due to security
reasons.

Regards,
Klaus.

Klaus,

  Thanks for your help. The last time I tried to upgrade to 16.04, my 
mythtv broke. I am sure I made some mistakes. I will ry once 20.04 is 
released to do a fresh install rather than upgrade from 14.04. That way 
I will manually install mythtv and do a database backup and restore. I 
have been lazy and not doing it right. This time I will be more thorough.


  May be I should delay 64bit conversion until then. Let me see, if I 
can manage delaying it till summer.


Regards
Ramesh

Re: Re: 64bit ext4 and kernel version compatibility

[R. Ramesh]

On Thu, Jan 09, 2020 at 05:06:29PM +, Tixy wrote:
> On Thu, 2020-01-09 at 14:35 +0300, Reco wrote:
> > On Wed, Jan 08, 2020 at 06:08:16PM -0600, R. Ramesh wrote:
> > > Before I get the source and build and update e2fsprogs and then the
> > > file system, I want to make sure that my current kernel version does
> > > not have any limitation to support 64bit ext4.
> > 
> > You kernel should support the feature, as it was introduced back at the

> > version 3.6 of the kernel - [1].
> > 
> > [1] https://ext4.wiki.kernel.org/index.php/Ext4_Metadata_Checksums
> 
> That web page is talking about metadata checksums I can't see anything

> about '64-bit'.

Let me read it for you (note kernel version):

Install Linux 3.6+ and e2fsprogs 1.43-WIP.
modprobe crc32c-intel
mkfs.ext4 -O metadata_csum,64bit /dev/path/to/disk


To enable "metadata_csum" in its full glory one needs to enable "64bit"
fs feature:

In a perfect world, one could simply enable metadata checksums ...
Therefore, it is best to start by formatting a fresh
filesystem with 64-bit support enabled ...


>What I'm assuming the OP is interested is 64-bit block numbers because
>they said they want to "convert ext4 fs on this server to 64bit so that
>I can grow it past 16TB limit". Note from me, 4kB sized blocks * 2^32 =
>16TB, so block numbers whould need to be more than 32-bit for bigger
>drives.

Same page says:

Therefore, it is best to start by formatting a fresh filesystem with
64-bit support enabled, since it is not possible to upgrade a 32-bit
filesystem to a 64-bit filesystem.


So if "convert" = "mkfs", then OP is fully set.
But then again, it's nothing that can't be solved by pre-existing
backup.

Reco


Reco,

  Thanks for the details. I do not know why, for a moment I thought 3.6 
> 3.13 (must have been treating version numbers as floating point 
numbers :-)


  No I do not plan to convert/grow in place. I have 2x8tb disks (and 
several other 4TB disks) for making backup. I will back up each md that 
is less than 16TB (separately) and make them 64bit first. After that I 
will grow them as needed.


Regards
Ramesh

64bit ext4 and kernel version compatibility

[R. Ramesh]

Hi,

  On my DVR, I am running fairly old version of kernel 
(3.13.0-132-generic, mythbuntu 14.04.5 LTS). I want to convert ext4 fs 
on this server to 64bit so that I can grow it past 16TB limit. At that 
time of installation (2014), e2fsprogs did not support 64bit fs.  Now it 
does. Before I get the source and build and update e2fsprogs and then 
the file system, I want to make sure that my current kernel version does 
not have any limitation to support 64bit ext4.   My google searches do 
not mention any kernel limitations. I think this is due to my inability 
to ask the right question to google. Appreciate any help I can get on 
answering my question.


At this time it is difficult for me to upgrade the kernel without 
updating the distribution (and hence mythtv) and that path is beyond me 
at this time due to time constraints.


Ramesh

Re: Re: kvm win8 guest audio is terrible.

[R. Ramesh]

More info needed:

- HW (cpu, graphics, soundcard)

Motherboard audio. I think it is realtek alc 887.


- BIOS (cpu power-states)
Guest cpu? Not sure how to answer this. I did not do anything special. 
My command looks something like this


/usr/bin/kvm -usbdevice tablet -full-screen -localtime\
?? -vnc $vnc -k en-us \
?? -drive file=$diskC,index=0,media=disk,cache=writeback \
?? -vga qxl\
?? -m 2047 -net nic,model=virtio,macaddr=$mac \
?? -net tap,ifname=tap2,script=no\
?? -soundhw all &



- Host setup (power states,audio setup)

Host is core-i5 cpu. Here is the first few lines of cpuinfo
processor : 0
vendor_id : GenuineIntel
cpu family?? : 6
model : 42
model name?? : Intel(R) Core(TM) i5-2500K CPU @ 3.30GHz
stepping?? : 7
microcode : 0x18
cpu MHz : 1651.812
cache size?? : 6144 KB




- VM config.
I think my kvm command line answers this. If not, please let me know 
what you are looking for.


Regards
Ramesh

Re: Re: kvm win8 guest audio is terrible.

[R. Ramesh]

On 5/20/19 10:35 PM, David Christensen wrote:

On 5/20/19 2:55 PM, R. Ramesh wrote:
I created a fresh install of debian stretch amd64. In that I created 
a qemu/kvm guess install of win8. It only accepted hda as a valid 
sound card. All others show us without drivers. So, I am limited to 
only HDA as -soundhw. Further HDA sounds so broken if I try to test 
with any sound file or youtube video. It is grainy/distorted and 
outright horrible.


Google searches mentions something about MSI and those approaches 
are simply not accepted by win8. So, I could not use them. I am 
wondering if there is something fundamentally wrong with windows 
guests or is there a setting that I missing.


Ramesh
Have you tried spice under Virtual Machine manager??? I used to have 
similar problems.?? Let all 6 cores run at maximum speed helped some 
but still occasional dropouts in Win7 and Win10.?? I switched to spice 
and audio and video improved greatly.

--


I invoke qemu-kvm directly. Do I have to use virtsh or virt manager? I 
am not set up for that. Also, googling spice does not get to the correct 
links.?? Can you help?


Ramesh

kvm win8 gues audio is terrible.

[R. Ramesh]

I created a fresh install of debian stretch amd64. In that I created a 
qemu/kvm guess install of win8. It only accepted hda as a valid sound 
card. All others show us without drivers. So, I am limited to only HDA 
as -soundhw. Further HDA sounds so broken if I try to test with any 
sound file or youtube video. It is grainy/distorted and outright horrible.


Google searches mentions something about MSI and those approaches are 
simply not accepted by win8. So, I could not use them. I am wondering if 
there is something fundamentally wrong with windows guests or is there a 
setting that I missing.


Ramesh

kvm win8 guest audio is terrible.

[R. Ramesh]

I created a fresh install of debian stretch amd64. In that I created a 
qemu/kvm guess install of win8. It only accepted hda as a valid sound 
card. All others show us without drivers. So, I am limited to only HDA 
as -soundhw. Further HDA sounds so broken if I try to test with any 
sound file or youtube video. It is grainy/distorted and outright horrible.


Google searches mentions something about MSI and those approaches are 
simply not accepted by win8. So, I could not use them. I am wondering if 
there is something fundamentally wrong with windows guests or is there a 
setting that I missing.


Ramesh

kvm win8 gues audio is terrible.

[R. Ramesh]

I created a fresh install of debian stretch amd64. In that I created a 
qemu/kvm guess install of win8. It only accepted hda as a valid sound 
card. All others show us without drivers. So, I am limited to only HDA 
as -soundhw. Further HDA sounds so broken if I try to test with any 
sound file or youtube video. It is grainy/distorted and outright horrible.


Google searches mentions something about MSI and those approaches are 
simply not accepted by win8. So, I could not use them. I am wondering if 
there is something fundamentally wrong with windows guests or is there a 
setting that I missing.


Ramesh

debian stretch tigervnc issues

[R. Ramesh]

Recently I got a nvme ssd and did a fresh install of debian stretch 
(amd64). In the new install, tigervnc is acting up. Commands like 
"vncconfig" and "vncserver" do not always work. Randomly, about half the 
time, commands just hang waiting for something and at other times they 
will exit without doing anything.  I do not have any idea why this is 
happening.


I usually fire off one vnc for each user account and we simply connect 
to one of them to use. This has become a pain now. We use a script vnc0 
:N below on an xterm to switch between different user's screen and it 
nolonger works reliably now



#!/bin/tcsh -f

if ($?DISPLAY) then
    vncconfig -disconnect
    set olddisplay = "$DISPLAY"
    setenv DISPLAY :0
    xvncviewer -FullScreen -FullColor $*
    xvncviewer -FullScreen \
    -passwd ~/.vnc/passwd -FullColor $olddisplay >& /dev/null &
else
    echo 'No current $DISPLAY. Trying :0'
    xvncviewer -display :0 -FullScreen -FullColor $* || \
    echo "display :0 not found"
endif


My older install  (jesse, i386) on the very same machine (dual boot) 
works just fine, although I think it is xvnc4.  I need amd64 to install 
windows 8 guest.  So, I am in a bit of a fix.  Please help if you know 
what I am missing in my tigenvnc setup.


Ramesh

Re: Re: Please help me resize my ext4 file system to size > 16TB

[R. Ramesh]

>   I am just looking for path of least resistance. That is why I asked if it

That would be organizing your videos into N subdirectories and using new
filesystems for some of those.  Each single filesystem will be smaller
than 16TB.

If that works with your use case, you could do that now.


Thanks. In that case, is mke2fs -O 64bit is enough? Or do I need a bunch 
of other options to go with it like  extent, extra_isize, etc. Can you 
tell me why xfs is better? Is it as stable as ext4? Will I be able to 
add dmcache or something similar as I create the new fs with xfs?


I have about 14TB over 4 disks. I will backup on to them and recreate a 
newfs and copy them back. I do not like the downtime as
that means my DVR does not work for 4 days, at least, as copying will 
have to be over usb3.


Ramesh

Re: Please help me resize my ext4 file system to size > 16TB

[R. Ramesh]

On 06/13/2017 10:48 AM, Doug wrote:


On 06/12/2017 11:33 PM, R. Ramesh wrote:

You implied you don't even have a backup of that data, which means you
have exactly one chance of getting it right.  This is a non-starter.

First: you are warned to NEVER proceed with a filesystem resize before
you have a valid, current, and *tested* backup.

You are also warned that the ext4 *conversion* to 64-bit block numbers
so as to be able to span more than 16TB is NOT your typical filesystem
resize operation in the first place.  It touches a *lot more* of the
filesystem and the risks are much higher than just adding an extent.
Even if it works flawlessly, it is not going to result in an optimal
filesystem.

The recommended procedure is to create a *new* filesystem and restore
from backup (or copy from the legacy filesystem if you find a way to
have both at the same time).  And for such large filesystems, the 
use of

"xfs" instead of "ext4" should be seriously considered.

Second: whatever reasons you had, or excuses you gave yourself, nothing
is going to get your data back if the filesystem ends up damaged beyond
repair and you don't have a backup.

"it would require updates that I am not ready do to yet"? Then DON'T.
Find another way to solve your immediate problem, and postpone the
larger filesystem for when you update everything.

You *have* been warned.


Now, you did not give us any idea of what is in that filesystem and 
what

you use it for, but as an alternative to resizing it, maybe you could
create several extra filesystems instead of enlarging the one you have
and attaching these extra ones to wherever you need more space? Any
subdirectory can be made an entirely separate filesystem...

This wouldn't work for everything (e.g. you can't hardlink or
fast-rename across filesystems), but maybe it would work for whatever
you use that big filesystem for?

--
   Henrique Holschuh

Henrique,

  Let us not worry about backup. The data is just videos and not 
worth the effort. They are all recreatable with some effort. So, it 
is ok to try resize2fs and if that dies, I will simply get the data 
from my DVDs/Blurays.


  I don't think there is in place ext4 to xfs conversion. So, 
converting to xfs is less possible that resize2fs.


  I am just looking for path of least resistance. That is why I asked 
if it possible to resize without any upgrade. I will eventually 
upgrade and eventually figure out something. I just need to hear that 
there isn't a simple solution I have overlooked *today*. That is all.


Ramesh


It would seem to me that the simplest way to solve this problem would 
be to buy a second drive that is big enough, format it with xfs, and 
copy everything
you want to keep onto the new drive. Then when everything is done to 
your satisfaction, you can keep the original drive as a backup, or if 
you really
don't care about backups, reformat the original drive and then use it, 
for whatever you want. (Make sure you copy the boot setup onto the new 
drive!)


--doug

Thanks. I have to spend at least $500 to get 12+TB disk. Not a good 
solution. I will wait it out till a stable resize2fs appears.


Ramesh

Re: Re: Please help me resize my ext4 file system to size > 16TB

[R. Ramesh]

You implied you don't even have a backup of that data, which means you
have exactly one chance of getting it right.  This is a non-starter.

First: you are warned to NEVER proceed with a filesystem resize before
you have a valid, current, and *tested* backup.

You are also warned that the ext4 *conversion* to 64-bit block numbers
so as to be able to span more than 16TB is NOT your typical filesystem
resize operation in the first place.  It touches a *lot more* of the
filesystem and the risks are much higher than just adding an extent.
Even if it works flawlessly, it is not going to result in an optimal
filesystem.

The recommended procedure is to create a *new* filesystem and restore
from backup (or copy from the legacy filesystem if you find a way to
have both at the same time).  And for such large filesystems, the use of
"xfs" instead of "ext4" should be seriously considered.

Second: whatever reasons you had, or excuses you gave yourself, nothing
is going to get your data back if the filesystem ends up damaged beyond
repair and you don't have a backup.

"it would require updates that I am not ready do to yet"? Then DON'T.
Find another way to solve your immediate problem, and postpone the
larger filesystem for when you update everything.

You *have* been warned.


Now, you did not give us any idea of what is in that filesystem and what
you use it for, but as an alternative to resizing it, maybe you could
create several extra filesystems instead of enlarging the one you have
and attaching these extra ones to wherever you need more space?  Any
subdirectory can be made an entirely separate filesystem...

This wouldn't work for everything (e.g. you can't hardlink or
fast-rename across filesystems), but maybe it would work for whatever
you use that big filesystem for?

--
   Henrique Holschuh

Henrique,

  Let us not worry about backup. The data is just videos and not worth 
the effort. They are all recreatable with some effort. So, it is ok to 
try resize2fs and if that dies, I will simply get the data from my 
DVDs/Blurays.


  I don't think there is in place ext4 to xfs conversion. So, 
converting to xfs is less possible that resize2fs.


  I am just looking for path of least resistance. That is why I asked 
if it possible to resize without any upgrade. I will eventually upgrade 
and eventually figure out something. I just need to hear that there 
isn't a simple solution I have overlooked *today*. That is all.


Ramesh

Re: Re: How to delay resume after suspend to get disks ready, using kernel command line switch?

[R. Ramesh]

> Problem:
>
> I am having a problem with ubuntu 14.04 resume from suspend. I suspect a
> race condition in boot process. I have a mpt2sas (LSISAS2008:
> FWVersion(20.00.07.00)) host adapter to which several of my disks are
> attached. On occasions, there is a delay before these devices become
> available. However, kernel tries to assemble RAID6 that includes these
> disks before they become available. As a result the array gets assembled
> in a degraded mode.
>
> My question:
>
> I like to tell kernel to delay resume until LSI card finishes its card
> identifying devices. Can I use resumedelay/rootdelay switches in
> grub.cfg to accomplish this? If so, which one I should use?
>
> Ramesh

I also have mpt2sas and it is slow in disk initialization but I never use
suspend/resume. I only use shutdown/reboot. I use rootdelay=15.
Interesting. I did not have to use rootdelay for reboots. This is 
because I don't get to grub screen until mpt2sas finishes its thing.
So, disks are ready. This is not true in suspend because normal bios 
sequence is not run for resume, I think.


Ramesh

Re: Samba help.

[R. Ramesh]

On 02/01/2017 08:02 PM, R. Ramesh wrote:

On 1/29/2017 11:25 AM, John Darrah wrote:

 On 1/29/2017 12:25 AM, R. Ramesh wrote:

 I recently upgraded my debian jesse to 8.6. All of a sudden
 all samba guest access to this box stopped working. I 
did not

 update smb.conf file any time before or after. I vaguely
 remember that there was a flash of notes flying by when 
samba
 was upgraded, but do not recall what it is. BTW, I can 
mount

 if I provide an actual valid user with password. Only guest
 access does not work. Linux host 3.16.0-4-686-pae #1 SMP
 Debian 3.16.36-1+deb8u2 (2016-10-19) i686 GNU/Linux

 Samba Version: 2:4.2.14+dfsg-0+deb8u2

 Guest access to the following share fails after the 
upgrade:


 [data]
  path = /data
  create mask = 0755
  directory mask = 0755
  public = yes
  browseable = yes
  writable = yes
  guest ok = yes

 If it is not too much trouble, please copy me your 
responses


 Thanks

 Ramesh


 Check to see that smb.conf was not replaced. If it was, the
 previous version will be called smb.conf.dpkg-dist or
 smb.conf.ucf-dist. You will have to compare them with your
 smb.conf file and merge any differences.

 -- john


Also check the following in [global] settings:

Change:

 map to guest = Bad User

to:

  map to guest = nobody

-- john

Thanks for your help.

I remember that apt-get asked if I wanted my smb.conf to be replaced. 
I said "no." Quick tkdiff between smb.conf and dist version shows 
that I miss these

  server role = standalone server
  map to guest = bad user
  usershare allow guests = yes

Of course, I will fix the second line and add all three and see what 
happens.

Ramesh



Adding the three lines with "bad user" replaced by "nobody" prevents 
nmbd from being restarted. It fails to restart and the syslong shows 
that it does not like nobody and commenting out the line allows 
restart of service. However I do not have guest mount permitted as 
before.


I tried putting a valid user instead of nobody and nmbd does not like 
that either. So, I am lost.


Ramesh


Actually using "map to guest = bad user" is needed for guest mount to 
work. I read it on samba blog as part of an error report 
https://samba.plus/blog/  dated (5/3/16)


So, problem solved for now (until I upgrade again :-)

Ramesh

Re: Re: Samba help.

[R. Ramesh]

On 1/29/2017 11:25 AM, John Darrah wrote:

 On 1/29/2017 12:25 AM, R. Ramesh wrote:

 I recently upgraded my debian jesse to 8.6. All of a sudden
 all samba guest access to this box stopped working. I did not
 update smb.conf file any time before or after. I vaguely
 remember that there was a flash of notes flying by when samba
 was upgraded, but do not recall what it is. BTW, I can mount
 if I provide an actual valid user with password. Only guest
 access does not work. Linux host 3.16.0-4-686-pae #1 SMP
 Debian 3.16.36-1+deb8u2 (2016-10-19) i686 GNU/Linux

 Samba Version: 2:4.2.14+dfsg-0+deb8u2

 Guest access to the following share fails after the upgrade:

 [data]
  path = /data
  create mask = 0755
  directory mask = 0755
  public = yes
  browseable = yes
  writable = yes
  guest ok = yes

 If it is not too much trouble, please copy me your responses

 Thanks

 Ramesh


 Check to see that smb.conf was not replaced. If it was, the
 previous version will be called smb.conf.dpkg-dist or
 smb.conf.ucf-dist. You will have to compare them with your
 smb.conf file and merge any differences.

 -- john


Also check the following in [global] settings:

Change:

 map to guest = Bad User

to:

  map to guest = nobody

-- john

Thanks for your help.

I remember that apt-get asked if I wanted my smb.conf to be replaced. 
I said "no." Quick tkdiff between smb.conf and dist version shows that 
I miss these

  server role = standalone server
  map to guest = bad user
  usershare allow guests = yes

Of course, I will fix the second line and add all three and see what 
happens.

Ramesh



Adding the three lines with "bad user" replaced by "nobody" prevents 
nmbd from being restarted. It fails to restart and the syslong shows 
that it does not like nobody and commenting out the line allows restart 
of service. However I do not have guest mount permitted as before.


I tried putting a valid user instead of nobody and nmbd does not like 
that either. So, I am lost.


Ramesh

Re: Re: Samba help.

[R. Ramesh]

On 1/29/2017 11:25 AM, John Darrah wrote:

On 1/29/2017 12:25 AM, R. Ramesh wrote:

I recently upgraded my debian jesse to 8.6. All of a sudden
all samba guest access to this box stopped working. I did not
update smb.conf file any time before or after. I vaguely
remember that there was a flash of notes flying by when samba
was upgraded, but do not recall what it is. BTW, I can mount
if I provide an actual valid user with password. Only guest
access does not work. Linux host 3.16.0-4-686-pae #1 SMP
Debian 3.16.36-1+deb8u2 (2016-10-19) i686 GNU/Linux

Samba Version: 2:4.2.14+dfsg-0+deb8u2

Guest access to the following share fails after the upgrade:

[data]
 path = /data
 create mask = 0755
 directory mask = 0755
 public = yes
 browseable = yes
 writable = yes
 guest ok = yes

If it is not too much trouble, please copy me your responses

Thanks

Ramesh


Check to see that smb.conf was not replaced. If it was, the
previous version will be called smb.conf.dpkg-dist or
smb.conf.ucf-dist. You will have to compare them with your
smb.conf file and merge any differences.

-- john


Also check the following in [global] settings:

Change:

map to guest = Bad User

to:

 map to guest = nobody

-- john



Thanks for your help.

I remember that apt-get asked if I wanted my smb.conf to be replaced. I 
said "no." Quick tkdiff between smb.conf and dist version shows that I 
miss these


 server role = standalone server
 map to guest = bad user
 usershare allow guests = yes

Of course, I will fix the second line and add all three and see what 
happens.


Ramesh

Samba help.

[R. Ramesh]

I recently upgraded my debian jesse to 8.6. All of a sudden all samba 
guest access to this box stopped working. I did not update smb.conf file 
any time before or after.  I vaguely remember that there was a flash of 
notes flying by when samba was upgraded, but do not recall what it is.


BTW, I can mount if I provide an actual valid user with password. Only 
guest access does not work.


Linux host 3.16.0-4-686-pae #1 SMP Debian 3.16.36-1+deb8u2 (2016-10-19) 
i686 GNU/Linux

Samba Version: 2:4.2.14+dfsg-0+deb8u2

Guest access to the following share fails after the upgrade:

[data]
path = /data
create mask = 0755
directory mask = 0755
public = yes
browseable = yes
writable = yes
guest ok = yes

If it is not too much trouble, please copy me your responses

Thanks

Ramesh

Re: Re: Growing md0 by larger disk - which is best method.

[R. Ramesh]

On 21/04/13 01:11 PM, R. Ramesh wrote:
   


I have a mdadm implemented raid5 with 3 disks (2x 3TiB and 1x
2TiB). Each disk has one partition only for the full size of
that disk and the partitions are then combined in to md0. I like
to swap out the 2TiB with a new 3TiB. While I do not expect
issue with this, a lot of reading about unrecoverable errors
spooked me a little on the rebuild/resync. Further, I think it
is silly that we need to read all the disks in the original
array as part of resync (in a fail+remove+add method of changing
disks) Is there a way to avoid resync by doing a dd from 2TB on
to the *new* 3TB and then reassembling the array? I mean this

   1. shutdown
   2. add 3tb to the PC
   3. boot using rescue disk and do not assemble the md0.
   4. Copy partition table from one of  old 3TiB on to the *new* 3TiB
   5. dd 2TiB-part1 to *new* 3TiB-part1
   6. shutdown
   7. disconnect 2TiB disk
   8. restart to original OS on the disk to find the 3TiB as part of the
  reassemble
   9. Grow the array to full size


 


Since the above is not suggested anywhere I could find, I like
to know what I am missing as this seems too easy to me.

Ramesh

 

The problem is the location of the superblocks. If you're prepared 
to have your system shut down for a while to copy the data, I 
suggest that you simply remove the 2T disk and add the 3T disk. You 
will still have your 2T disk so if something fails, you should be 
able to recover using it. This of course means that you cannot be 
using your system. The md array cannot be written to or the 2T drive 
will no longer be usable. So boot from a rescue disk and rebuild the 
array without mounting it or any partition on it until the rebuild 
has completed. 


Thanks. I was thinking copy will be faster than rebuild. Further, it 
will avoid reading 4TB disk (of course, it will read 2TB when 
copying) I guess you are saying that the simple copying will not work.


I was going to try experimenting with files and loop devices (I tried 
2x 3G + 1x 2G files plus a new 3G file representing the new 3TB 
drive). However, after copying the files, loop thinks that the file 
representing new 3TB disk (or the new 3GB file) is really a 2TB disk. 
So, mdadm does not grow the array when using the new loop device. 
However, the file system was intact after copying and I could mount 
it etc. I just could not grow

it to full 6TB.

Ramesh

Reading 4T in parallel is not slower than reading 2T. However, the 
rebuild is probably slower than a simple copy. Unfortunately any kind 
of playing around with the array may have resulted in the 2T disk no 
longer being a viable member of the array. For safety you may want to 
restart the array with the 2T disk in it to make sure that it is OK 
before replacing it with the 3T.


However, if it looks like it is doing a full rebuild then stop the 
array and switch to the 3T. If the array is going to be rebuilt 
anyway, then you might just as well build it on the 3T disk.


Make sure you have all your important files backed up beforehand.


Thanks Gary for thinking this through for me. After our exchanges and 
failure to successfully experiment with simple files and loop devices, I 
decided to let mdadm rebuild/resync instead of copying.  I plan to do 
the following.


  1. Add the new 3TB as spare
  2. Manually fail the 2TB to start resync in place on to 3TB.
  3. When finished. Simply remove 2TB from the array

Just curious. If I do not write to this array (ie, unmount it before 
syncing), will the 2TB be still usable should

the resync to 3TB fail?

Ramesh

Growing md0 by larger disk - which is best method.

[R. Ramesh]

I have a mdadm implemented raid5 with 3 disks (2x 3TiB and 1x 2TiB). 
Each disk has one partition only for the full size of that disk and the 
partitions are then combined in to md0.


I like to swap out the 2TiB with a new 3TiB. While I do not expect issue 
with this, a lot of reading about unrecoverable errors spooked me a 
little on the rebuild/resync. Further, I think it is silly that we need 
to read all the disks in the original array as part of resync (in a 
fail+remove+add method of changing disks)  Is there a way to avoid 
resync by doing a dd from 2TB on to the *new* 3TB and then reassembling 
the array? I mean this


  1. shutdown
  2. add 3tb to the PC
  3. boot using rescue disk and do not assemble the md0.
  4. Copy partition table from one of  old 3TiB on to the *new* 3TiB
  5. dd 2TiB-part1 to *new* 3TiB-part1
  6. shutdown
  7. disconnect 2TiB disk
  8. restart to original OS on the disk to find the 3TiB as part of the
 reassemble
  9. Grow the array to full size


Since the above is not suggested anywhere I could find, I like to know 
what I am missing as this seems too easy to me.


Ramesh


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Archive: http://lists.debian.org/51741da5.5010...@verizon.net

Re: Re: Growing md0 by larger disk - which is best method.

[R. Ramesh]

On 21/04/13 01:11 PM, R. Ramesh wrote:
   


I have a mdadm implemented raid5 with 3 disks (2x 3TiB and 1x
2TiB). Each disk has one partition only for the full size of that
disk and the partitions are then combined in to md0. I like to
swap out the 2TiB with a new 3TiB. While I do not expect issue
with this, a lot of reading about unrecoverable errors spooked me
a little on the rebuild/resync. Further, I think it is silly that
we need to read all the disks in the original array as part of
resync (in a fail+remove+add method of changing disks) Is there a
way to avoid resync by doing a dd from 2TB on to the *new* 3TB and
then reassembling the array? I mean this

   1. shutdown
   2. add 3tb to the PC
   3. boot using rescue disk and do not assemble the md0.
   4. Copy partition table from one of  old 3TiB on to the *new* 3TiB
   5. dd 2TiB-part1 to *new* 3TiB-part1
   6. shutdown
   7. disconnect 2TiB disk
   8. restart to original OS on the disk to find the 3TiB as part of the
  reassemble
   9. Grow the array to full size


 


Since the above is not suggested anywhere I could find, I like to
know what I am missing as this seems too easy to me.

Ramesh

 

The problem is the location of the superblocks. If you're prepared to 
have your system shut down for a while to copy the data, I suggest 
that you simply remove the 2T disk and add the 3T disk. You will still 
have your 2T disk so if something fails, you should be able to recover 
using it. This of course means that you cannot be using your system. 
The md array cannot be written to or the 2T drive will no longer be 
usable. So boot from a rescue disk and rebuild the array without 
mounting it or any partition on it until the rebuild has completed. 


Thanks. I was thinking copy will be faster than rebuild. Further, it 
will avoid reading 4TB disk (of course, it will read 2TB when copying) I 
guess you are saying that the simple copying will not work.


I was going to try experimenting with files and loop devices (I tried 2x 
3G + 1x 2G files plus a new 3G file representing the new 3TB drive). 
However, after copying the files, loop thinks that the file representing 
new 3TB disk (or the new 3GB file) is really a 2TB disk. So, mdadm does 
not grow the array when using the new loop device. However, the file 
system was intact after copying and I could mount it etc. I just could 
not grow

it to full 6TB.

Ramesh

Re: Re: Re: Please help me to evaluate flash/ssd life using vmstat -d

[R. Ramesh]

On Fri, 2011-02-04 at 18:30 -0600, R. Ramesh wrote:

 I do not have SSD. I have a USB flash drive - went cheap on this :-)
 
 Regardless of the above, still every write by the kernel has to be 
 translated in to NAND writes. I have read in more than one place that 
 these writes will be in units of erase-block size regardless of the 
 kernel IO size.


I could be wrong, but I've always understood that you could write less
than an erase-block size of data, just that the part of NAND written to
can't have previously been used since the last erase. E.g.

- Erase 128kB block
- Write 64kB to first half of block
- Later write 64kB to second half of block
- Need to erase whole 128kB block before it can be written to again.

  I am simply trying to map the X kernel writes in to Y 
 erase-block writes. Note that I do not map it specific erase-block, but 
 to some erase-block. So I only worry about the block counts not the 
 block addresses. That is why my calculations are based on total number 
 of erase-blocks writes available (= 16G/512k*1 = 32768) before 
 the device goes bad. So to me the life of flash is 32768 erase-block 
 writes. Now how many hours is it? To answer this, I need to understand 
 what vmstat -d prints.


If I'm right about partial writes to a NAND erase-block, then I don't
think that you need to try and factor in erase block sizes, just use the
fact that for each byte written a byte must have been previously erased.
Assuming that the flash controller only erases full blocks (it won't be
that efficient though) then you just calculate the total amount of data
that can be written to the disk. In your case, 16GB disk * 1 erases
= 160TB data before the disk expires.

In my previous job when we were thinking about wear on MMC cards due to
demand paging, the calculations showed that if you wrote continuously to
the card at it's maximum rate supported it would last several years. It
was at that point I stopped worrying about flash wear :-)

  
I also think that you can write less than an erase block. But, 
calculations become harder if I take that approach because I need to 
figure out which kernel writes will  require erase and which will not.
I do not know how to come up with that magic. So I wanted to use vmstat 
and figure out in a different way.


Before that, let us try your approach above (at the end). It suggests 
about writing at full speed and see how long  it takes for 160TB. My 
flash has advertised rate of 15MB/sec (sequential, so the best it can 
do). So this gives us 160*1024*1024/15 sec or 129.5 days. That does seem 
right. So this method is unrealistically conservative.


So what does vmstat -d tell me? Is the number of IO under total column 
supposed to be the number of IOs
issued to the controller with each IO being contiguous N sectors? If not 
then my 24 years is the best I can come up with.


Ramesh


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Archive: http://lists.debian.org/4d4df4ea.5060...@verizon.net

Please help me to evaluate flash/ssd life using vmstat -d

[R. Ramesh]

All,

 I run a debian firewall on an atom PC running of a 16G flash drive. I 
am trying understand the amount of disk write performed in order to 
understand how long my flash is likely to last. I have two vmstat -d 
information on the flash


   2011-01-31 00:35
   disk- reads writes--- 
-IO--
  total merged sectors  ms  total merged sectors  ms
cursec
   sda10624   4770  352041   16848   2532464   70880  
546908  0 36
   sda1   10621   4767  351993   16844   2532464   70880  
546908  0 36


   2011-02-03 16:01
   disk- reads writes--- 
-IO--
  total merged sectors  ms  total merged sectors  ms
cursec
   sda10736   4827  357265   17084   9649   1621  206984 
1057928  0 76
   sda1   10733   4824  357217   17080   9649   1621  206984 
1057928  0 76


Note that the kernel writes do not translate into equivalent size writes 
into the flash. This is because flash has to be always written in 
multiples of erase block units (or write allocation size or some such 
thing). So I like to know how to translate above writes into number of 
erase-block writes. I could simply use number of sectors as the number 
of erase blocks and get a very conservative estimate. This way I get 
136104 erase blocks in 88 hours. Assuming erase block is 512K and flash 
can deal with 10K writes per block before it dies, I calculate 
(16G*1)/(136104*512K)*88 hours of life. This is 211866 hours or 24 
years. Plenty for me.


But, I really do not like being this conservative. So, I like to know if 
the number of writes under the column titled total refers to the 
number of kernel writes in which each write is a single disk IO for 
contiguous group of sectors. If so, I can assume that each such write 
will cause at most two erase-block writes on the flash giving me a life 
of (16G*1)/((9649-2512)*2*512K) * 88 hours. This is 2020165 hours or 
230 years. While I do not need this much life out of the flash, I like 
to know which one is closer.


I have not done any optimization for the flash like aligned partitioning 
and use of tmpfs. However, I have used noatime mount option which 
reduced the number of writes by an order of magnitude or more.


Ramesh


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Archive: http://lists.debian.org/4d4c46cf.9050...@verizon.net

Re: Re: Please help me to evaluate flash/ssd life using vmstat -d

[R. Ramesh]

On Fri, 2011-02-04 at 12:34 -0600, R. Ramesh wrote:
 All,
 
   I run a debian firewall on an atom PC running of a 16G flash drive. I 
 am trying understand the amount of disk write performed in order to 
 understand how long my flash is likely to last.


[..]

 Note that the kernel writes do not translate into equivalent size writes 
 into the flash. This is because flash has to be always written in 
 multiples of erase block units (or write allocation size or some such 
 thing). So I like to know how to translate above writes into number of 
 erase-block writes.


Assuming you have a standard SSD drive attached via SATA or the like,
then you can't make any association between writes issued by the kernel
and the actual NAND device writes inside the drive. This is all
abstracted by the SSD controller which will have a fancy flash
translation layer to do wear levelling and bad block management.

  

I do not have SSD. I have a USB flash drive - went cheap on this :-)

Regardless of the above, still every write by the kernel has to be 
translated in to NAND writes. I have read in more than one place that 
these writes will be in units of erase-block size regardless of the 
kernel IO size. I am simply trying to map the X kernel writes in to Y 
erase-block writes. Note that I do not map it specific erase-block, but 
to some erase-block. So I only worry about the block counts not the 
block addresses. That is why my calculations are based on total number 
of erase-blocks writes available (= 16G/512k*1 = 32768) before 
the device goes bad. So to me the life of flash is 32768 erase-block 
writes. Now how many hours is it? To answer this, I need to understand 
what vmstat -d prints.


Obviously I want to know how many writes are done by the kernel in a 
given hour (on the average).
Assuming that the flash controller is (very) smart to distribute the 
writes 100% evenly, you calculate the life dividing one by the other.  
Obviously, I do not know how to translate the output of vmstat -d to 
actual disk IO seen by the flash controller. Does it see it as X IOs 
from the total column or as Y IOs from the sectors column? Grossly 
taking each IO as a erase-block write I can calculate the life.


Of course the above assumes that the controller is so smart that it 
makes every one of the 32768 writes available to kernel before it 
declares flash is dead. I need this assumption to abstract the 
underlying controller algorithm.


Ramesh


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Archive: http://lists.debian.org/4d4c9a1b.3020...@verizon.net

Re: How to get rid of this firewall error.

[R. Ramesh]

Ramasubramanian Ramesh wrote:


top posting - iihh ugly ;-p

  well, it's the netfilter code from the kernel instructed by an 
iptables rule, that spits out that message.

most likely this is a message informing you about a blocked packet.
  the question is: who/what set this iptables rule to tell the kernel 
to discard such packets?

my cristal ball is currently somehow foggy, so I can't tell.
maybe you know what firewall frontend you are using?

what to do?
  well - to prohibit logging of those packets in the short term a 
rule like this may do the job:

iptables -I INPUT -i eth1 -s 192.168.1.1 -p icmp --icmp-type 8 -j DROP

for the long term:
learn how to configure your firewall frontend or iptables.

but it's you modem/router right?
  so it might just check if you online. so it could be a 'good 
packet', which you might want to accept. - same thing as above just 
with the ACCEPT target.

best regards

Mart

  
My mail tool some times refuses to bottom post, especially, when I 
reply to myself. I have to jump through hoops and some time it is 
easier to submit to its demands :-)


Agreed. I set up the tables to drop and log messages whenever an 
unsolicited message comes from outside. But, I did not think
I setup the router to send periodic pings. I guess the verizon router 
(yes, it is mine because verizon gave it to me free) does that because 
it is windows centric and it likes to by user friendly by inquiring 
are you there? I greet you, etc.


It may be a good packet or it may be a bad packet. I still do not want 
to look at unsolicited packets.  I am worried that some one from 
outside can spoof as my router. I will just drop this specific packet 
without logging it.


Specifically, what does icmp_type = 8 mean?

Ramesh





My bad, I googled icmp_type 8. It seems harmless and required to be 
implemented. So I am going to accept.
After filtering out this one, I notice another one coming from my own 
firewall and need to figure out who is sending it.


[2731831.967429] IN=eth1 OUT= MAC= SRC=192.168.1.47 DST=192.168.1.255 
LEN=233 TOS=0x00 PREC=0x00 TTL=64 ID=0 DF PROTO=UDP SPT=138 DPT=138 LEN=213


I think this from my nmbd on the outgoing port. I am going to 
investigate interfaces option in smb.conf. Please tell me if I am on the 
wrong path.


Ramesh


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Archive: http://lists.debian.org/4c526491.20...@verizon.net

Re: Is it possible to get Nvidia 8500GT Vid. card to work with debian stable?

[R. Ramesh]

On Sun, Jan 13, 2008 at 12:24:18AM -0600, R. Ramesh wrote:
 Per debian package search and apt-cache info the version of Xorg in 
stable  is 7.1.  Open source nv driver in Xorg 7.1 seem to support 
only upto  Geforce 7xxx. I want to buy nvidia 8500GT cart and it is 
based on Geforce  8xxx. My understanding is that this card is 
supported by nv driver in Xorg  7.2. Are there any back port of Xorg 
7.2 available for Debian stable/etch  that I could use

 without having to dist-upgrade to lenny?

Holger Levsen recently posted about this on his blog:

http://layer-acht.org/blog/debian/#1-148

HTH.

Kumar
--
Kumar Appaiah,
458, Jamuna Hostel,
Indian Institute of Technology Madras,
Chennai - 600 036
 
**





Thanks. But, unfortunately, it appears that there is no amd64 port in 
http://layer-acht.org/blog/debian/#1-148.


Any idea about amd64 backports, if any?

Regards
Ramesh


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Is it possible to get Nvidia 8500GT Vid. card to work with debian stable?

[R. Ramesh]

Hi,

Per debian package search and apt-cache info the version of Xorg in 
stable is 7.1.  Open source nv driver in Xorg 7.1 seem to support only 
upto Geforce 7xxx. I want to buy nvidia 8500GT cart and it is based on 
Geforce 8xxx. My understanding is that this card is supported by nv 
driver in Xorg 7.2. Are there any back port of Xorg 7.2 available for 
Debian stable/etch that I could use

without having to dist-upgrade to lenny?

Regards
Ramesh


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Re: Re: How to make boot CD to run your curent hard disk installed linux?

[R. Ramesh]

On Sat, Oct 22, 2005 at 11:57:47PM -0500, Ramasubramanian Ramesh wrote:

Hi,

I have installed stable release of debian (using the netinst CD) on a 
headless machine (no kb, mouse or monitor) . The machine also does not 
have a floppy drive. I like to make a bood cd of the installed kernel so 
that I can bypass the grub boot. Specifically the grub setup boots win 
XP by default and I need to have something that can boot linux on 
demand.  (Note that without KB and monitor I am blind to grub 
interaction an cannot ask it to boot the non default selection which is 
linux) 


Why would you install two operating systems on a machine where you 
cannot choose the OS at boot time?



--
Nothing will dispel enthusiasm like a small admission fee.
-- Kim Hubbard

This is my HTPC and going to sit next to my stereo system (it has a HD 
tuner card and all the intersting win MCE like stuff). However, it will 
be the backup firewall should my main firewall linux machine fail. Thus 
I want the linux side to exist and upto date. The upto date part 
requires periodic boot into linux and I do not want to physically move 
the machine each time next to a monitor keyboard and go through all the 
connectivity just to do that. Also each time I upgrade windows I  want a 
quick way of booting into installed linux and just do grub-install to 
restore the mutiboot. If you have a better suggestion, let me know.


Most important of all, I have been saved several time by this type of 
floppy whenever I screw up my lilo run or want to swap /dev/hda when 
linux is in /dev/hdb.


Regards
Ramesh


--
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]