Solved - Debian 9.6.0 amd64 freezes during boot

[jwsarge2]

Hello and Happy Veteran's Day

I installed Debian 9.6.0 amd64 (only OS in the machine), instead of the default partition scheme I went with the /home, /var and /tmp option and selected LXDE as the desktop environment during the install. On completion of the install the system rebooted successfully to the LXDE desktop. I logged in no problem then loaded all the backup files I had saved on an USB stick from a previous install of Lubuntu into a couple of my new /home sub-directories. After that, using Synaptic Package Manager I then installed xcompmgr followed by variety; I assumed that Synaptic would do all the configuring required for xcompmgr to function with Cairo-Dock to be installed later on. I decided to reboot the machine before configuring variety. The boot process now after numerous attempts only gets as far as where I can select Debian 9.6 or the second choice with additional options; I tried both but which ever I select causes the red HDD light to stay steady for a minute or so. After that it makes no difference which one I selected but the machine ends up with a black monitor display and I cannot use the keyboard at all (can't Ctl-Alt-F2, Ctl-Alt-Backspace, etc...). Nothing works, can't even get a GUI or a terminal login, the only thing that works is holding down the power button to shut it down. This makes it impossible to review the startup logs to see what, if any errors were encountered. I suspect the x system was negatively impacted by the installation of xcompmgr using Synaptic Packager Manager but not sure. I am considering re-installing it all but would like to avoid the same problem again when I move past the initial install. Any suggestions that you may be able to provide will be most appreciated. Thanks in advance...

R/jwsarge2

On the side of the box for OS requirements was written "WIN9X or better" so I installed Debian/GNU Linux.

[bw123]

Before you reinstall, try kernel parameter systemd.restore_state=0 -or- even turn acpi=off and see if you can get in.

I went thru something like this couple yrs ago, when moving to stretch on an older dell notebook. Installation fine, first boot fine, everything good. Went to a lot of trouble setting things up and worked several hours on it, etc. Then when I rebooted, a hung system. Booting to single user or isolating multi-user or rescue.target hung also, that rules out a problem with X. I installed probably ten times over 6 months in various different configurations, downloaded multiple different images, etc.. and the machine always hung after the 'second' or 2nd boot from install. Can't boot new install twice.

[jwsarge2]

Good evening bw123,

Thanks so very much for your feedback, most appreciated. You may have seen the same movie I am currently watching. I found that if I rebooted off the installation CD then instead of re-installing I went with an alternate route that allowed me to re-install GRUB in the existing system which afterwards would then boot up fine but any follow-on reboots I'd end up with the same problem, a system that would not boot up completely, a black screen and none of the special keyboard combinations I found online for alternatives to holding the power button down to shut it down did anything. I'll try and get a terminal the same way and use your suggestions. I am trying to avoid using DBAN, https://dban.org/, to clean up and prep the drive as it takes about a day to do that (hate to wait that long ). I had to do that once in the past to clear the road so to speak in order to successfully install a Linux system in this same computer that used to have a windoze OS when multiple attempts to do the desired install kept failing. Will advise how it goes, have a nice evening...

[bw123]

You don't need a terminal to try some kernel parameters, just hit 'e' when at the grub menu and add them to the end of the line that begins 'linux' and has 'ro quiet' at the end then F10 boots. Only persists for that one boot attempt. I don't have a link for all of them handy, they seem to change them or remove them pretty often.

[jwsarge2]

Progress of sort,

Good afternoon. I saw mention online of turning off ACPI in the bios so I tried that and so far the machine boots up. Flip side of the coin is my screen resolution is now 600x800. The Debian 9.6 Screen Monitor routine is not much help and I have so far been unable to get my 1920x1080 resolution back. Video Card is a Geoforce 1050LE and it worked no problem in the past so I'd rather not mess with drivers. If anyone has a suggestion on how to get this Debian 9.6 install with an LXDE desktop environment tuned up so I can maximize the card's resolution I'd be most happy. Thanks in advance...

[bw123]

Return BIOS to defaults, and use kernel parameters instead to control ACPI.

I was confused by your tagline about win9x or better, I guess this is a joke. The hardware is pretty new? Adding some info like lspci, or using pkg inxi to let people know some more info about the machine might get you a quick fix if it is a known issue.

[jwsarge2]

Good morning bw123 and all,

Yes the sig line is a joke

I tried enabling the ACPI Function in the Power Section of the AMI Bios and the machine would not boot so I am using it now with that disabled again. I'm going to have to study up on adjusting kernel parameters. As requested here is some machine info:

john@DenLinuxPC:~$ lspci
00:00.0 RAM memory: NVIDIA Corporation C51 Host Bridge (rev a2)
00:00.1 RAM memory: NVIDIA Corporation C51 Memory Controller 0 (rev a2)
00:00.2 RAM memory: NVIDIA Corporation C51 Memory Controller 1 (rev a2)
00:00.3 RAM memory: NVIDIA Corporation C51 Memory Controller 5 (rev a2)
00:00.4 RAM memory: NVIDIA Corporation C51 Memory Controller 4 (rev a2)
00:00.5 RAM memory: NVIDIA Corporation C51 Host Bridge (rev a2)
00:00.6 RAM memory: NVIDIA Corporation C51 Memory Controller 3 (rev a2)
00:00.7 RAM memory: NVIDIA Corporation C51 Memory Controller 2 (rev a2)
00:02.0 PCI bridge: NVIDIA Corporation C51 PCI Express Bridge (rev a1)
00:03.0 PCI bridge: NVIDIA Corporation C51 PCI Express Bridge (rev a1)
00:04.0 PCI bridge: NVIDIA Corporation C51 PCI Express Bridge (rev a1)
00:05.0 VGA compatible controller: NVIDIA Corporation C51 [Quadro NVS 210S/GeForce 6150LE] (rev a2)
00:09.0 RAM memory: NVIDIA Corporation MCP51 Host Bridge (rev a2)
00:0a.0 ISA bridge: NVIDIA Corporation MCP51 LPC Bridge (rev a3)
00:0a.1 SMBus: NVIDIA Corporation MCP51 SMBus (rev a3)
00:0b.0 USB controller: NVIDIA Corporation MCP51 USB Controller (rev a3)
00:0b.1 USB controller: NVIDIA Corporation MCP51 USB Controller (rev a3)
00:0d.0 IDE interface: NVIDIA Corporation MCP51 IDE (rev a1)
00:0e.0 IDE interface: NVIDIA Corporation MCP51 Serial ATA Controller (rev a1)
00:0f.0 IDE interface: NVIDIA Corporation MCP51 Serial ATA Controller (rev a1)
00:10.0 PCI bridge: NVIDIA Corporation MCP51 PCI Bridge (rev a2)
00:10.1 Audio device: NVIDIA Corporation MCP51 High Definition Audio (rev a2)
00:14.0 Bridge: NVIDIA Corporation MCP51 Ethernet Controller (rev a3)
00:18.0 Host bridge: Advanced Micro Devices, Inc. [AMD] K8 [Athlon64/Opteron] HyperTransport Technology Configuration
00:18.1 Host bridge: Advanced Micro Devices, Inc. [AMD] K8 [Athlon64/Opteron] Address Map
00:18.2 Host bridge: Advanced Micro Devices, Inc. [AMD] K8 [Athlon64/Opteron] DRAM Controller
00:18.3 Host bridge: Advanced Micro Devices, Inc. [AMD] K8 [Athlon64/Opteron] Miscellaneous Control
04:08.0 FireWire (IEEE 1394): VIA Technologies, Inc. VT6306/7/8 [Fire II(M)] IEEE 1394 OHCI Controller (rev c0)

john@DenLinuxPC:~$ inxi
CPU~Dual core AMD Athlon 64 X2 4200+ (-MCP-) speed~0.00 MHz Kernel~4.9.0-8-amd64 x86_64 Up~27 min Mem~580.2/7924.4MB HDD~1000.2GB(1.4% used) Procs~144 Client~Shell inxi~2.3.5

Hope this helps and thanks again for your efforts, most appreciated...

[jwsarge2]

MB info if that will help:

john@DenLinuxPC:~$ sudo dmidecode -t baseboard
# dmidecode 3.0
Getting SMBIOS data from sysfs.
SMBIOS 2.5 present.

Handle 0x0002, DMI type 2, 15 bytes
Base Board Information
Manufacturer: MSI
Product Name: MS-7252
Version: 1.0
Serial Number: To be filled by O.E.M.
Asset Tag: To Be Filled By O.E.M.
Features:
Board is a hosting board
Board is replaceable
Location In Chassis: To Be Filled By O.E.M.
Chassis Handle: 0x0003
Type: Motherboard
Contained Object Handles: 0

Handle 0x002B, DMI type 10, 6 bytes
On Board Device Information
Type: Video
Status: Enabled
Description: To Be Filled By O.E.M.

[jwsarge2]

Good afternoon bw123,

Solved possibly

I turned acpi back on in the power management section of the bios then tried your first suggestion to inject acpi=off while the system was coming on. The use of acpi=off made no changes as the resolution was still 600x800 so it definitely requires acpi to run properly. I rebooted the system and this time inserted systemd.restore_state=0 at the end of the linux line and the machine booted up correctly with the 1920x1080 resolution. Also I noted with acpi off either by the bios or when i put acpi=off at the end of the linux line in the boot process the machine would not respond correctly to my using the shutdown icon; it would exit the OS but not shutdown the machine without my using the machine's power button; that is working properly now as well. I did find out if I use the terminal or the run box and enter shutdown I get an error stating the command cannot be found, I'm guessing I need to add it to the system. So, hopefully the problem is fixed but after rebooting what I entered at the end of the linux line is no longer there as you mentioned. So hopefully setting it to systemd.restore_state=0 will have corrected it after that one use. I'll try it out for a few days and if all is well I'll label this post SOLVED in a follow-on reply. Thanks again, it was getting frustrating and I really did not want to revert back to Lubuntu as a couple of programs I tried to use with it had problems. Have a great day...

[bw123]

systemd.restore_state=0 isn't a fix, it's a workaround. I had to make it permanent for awhile in /etc/default/grub until I learned what was going on.

The fix will probably be in finding out what thing it is trying to reset that hangs the kernel when booting, and mask it. In my case it was a (non-existent)keyboard backlight. I think this may be one good example where systemd got involved in something for no good reason.

This report https://bugzilla.redhat.com/show_bug.cgi?id=1253523 was interesting, and it helped me solve my issue. Thankfully there are very persistent people using linux like Daniel Kian Mc Kiernan that leave breadcrumbs about their struggles. Three years to pursue a bug stubbornly is really amazing.

If this thing is affecting other hardware, then I'd use an installation report and let people know, I did. https://bugs.debian.org/cgi-bin/bugrepo ... bug=868085

Good luck.

p.s. it takes "two boots" to verify the problem. The first good boot is where the state is saved, the 2nd boot is where it gets hung. That is why it is truly evil. You can waste hours of time on a good newly installed system, and then on reboot you are locked out hard, no warning, no nothing.

[jwsarge2]

Good evening bw123,

Well on the 3rd re-boot it went back to square one so as you advise it was just a temporary fix. AAARRRRGGGGGHHHH... Going to watch the USA vs
Scotland soccer match. Tomorrow I will consider the install report vs going back to ubuntu.

[jwsarge2]

Good morning bw123,

Where in the /etc/default/grub file did you insert the systemd.restore_state=0, at the end below the last line or does it make any difference?
GRUB_DEFAULT=0
GRUB_TIMEOUT=5
GRUB_DISTRIBUTOR=`lsb_release -i -s 2> /dev/null || echo Debian`
GRUB_CMDLINE_LINUX_DEFAULT="quiet"
GRUB_CMDLINE_LINUX=""

[bw123]

AFAIK, you'll need to edit the line GRUB_CMDLINE_LINUX and enclose the kernel param in the empty quotes to make it active for all boots, including single user or maintenace mode. Run update-grub as root afterwards.

If you want to dig more at some point, then man systemd-backlight now has some additional info I never noticed before. It may have been added recently...

...the brightness is clamped to a value of at least 1 or 5% of maximum brightness, whichever is greater.
...

See, this would obviously be a problem for a device where maxbrightness was 0, and would surely not be wise. We can't call it a bug in systemd, just call it an 'issue' ok?

You could also just mask the systemd-backlight service, or the one that is hanging. I believe they are all auuto-magically generated. You can see them all like this:

Code: Select all

systemctl status systemd-backlight*


The reason you should use an installation report is to help other users of debian that might have the same problem on your hardware. Also, someday debian may have devs and leadership that actually want to solve issues like this. Currently, they seem to all be sycophants living in the same cathedral.
https://en.wikipedia.org/wiki/File:Sand ... A02854.jpg
https://en.wikipedia.org/wiki/Hubris

[jwsarge2]

Afternoon bw123,

Thanks for the clarification and input. If I enter systemctl status systemd-backlight* in the terminal I get nothing after enter, am I missing something? LMK and thanks again.

sycophants - good one....

[bw123]

Thanks for the clarification and input. If I enter systemctl status systemd-backlight* in the terminal I get nothing after enter, am I missing something? LMK and thanks again.

Because you used systemd.restore_state=0 when you booted? Probably that prevents the service files from being created, but the save states will still show up in /var/lib/systemd/backlight (or other places in var/lib/systemd) with a filename usually starting with pci address and a clue in the filename about what they are.

It has been over a yr since I messed with it, so things may have been reorganized. Take your time and let people know if you're interested in leaving your own breadcrumbs, thanks.