Multiple computers locking up with 20.04LTS

[WB0HYQ]

I posted a while back about one of my computers locking up unexpectedly. Now I have 4 of them doing the same thing. One is a desktop (homebuilt) and the other three are laptops (Two models of Asus and an HP). All four are running Ubuntu (or XUbuntu in one case). All of them were upgraded around the same time from 18.04LTS. Previously, they ran solidly with no freezing. Since upgraded, they freeze randomly.

I downloaded an ISO copy of MEMTest from the source and burned it to a bootable CD. It ran in all four computers overnight with no failures, yet within ten minutes of starting up, two laptops froze. One was at the purple logon screen, the other was playing an MP3 song using VLC. The odd thing was the music never stopped, but everything else locked up: mouse, keyboard, and even the power button (which was supposed to power down when pressed). Sometimes, the mouse keeps going. There are times (on the XUbuntu machine, where Cairo-Dock and the mouse were moving, but the dock didn't respond to a mouse click.

This cannot be a coincidence. There has to be some commonality other than running the same OS. It is driving me around the bend in its randomness. I can find no clues in any of the log files. Right up the instant of failure, it was business as normal.

Can someone suggest a different point of attack here?

Bill

[dragonfly41]

There has to be some commonality other than running the same OS.

In your shoes I would come to the same conclusion. But I would then ask myself do these freezes all occur at the same time on all four systems? The only commonality I can think of is common power supply or LAN. Do they work separately (i.e. only one tested at a time)? A process of elimination is needed to pin it down further.

[WB0HYQ]

No. I usually run with only one or two of the computers powered up at any time. The time between boot and freeze may vary by as much as an hour or even a day, but eventually they will lock up. As far as connections go, they are all hooked into my house LAN either by wi-fi or hardwire. One of the laptops froze when I was at a book signing and totally away from my house. The other froze during a flight to San Francisco. The desktop runs 24/7/365. I did recently up the power supply from 450w to a 780w unit, but the freezes continued even after the upgrade.

I feel the freezes are due to something in the way Ubuntu handles RAM. If I am not doing anything, and the computer is simply idling on a desktop with the screensaver running, it tends not to freeze. When I run multiple apps, this increases the probability of a freeze. Note that I did run MEMTEST from a bootable CD and it found no flaws in RAM.

The desktop is a dual-boot machine, running Windows 7 and Ubuntu. Both have their own hard drives. When in Windows, I can do anything and everything I need to do with no failures. It will run for weeks and not freeze. It is something in Ubuntu that is causing the lockup.

One of the four, a laptop, was upgraded to Ubuntu 21.04 yesterday. I've left it running overnight and so far this morning it is still running. This will be my control computer. If it keeps running even after I subject it to some rather strenuous testing (such as running multiple VLC apps as well as opening many tabs (say, 10) in Firefox) to see if it will freeze.

Bill

[dragonfly41]

Intuitively .. you may have setup swap on all systems.
I would next explore that option.
And/or increase RAM.
I usually suggest Stacer for monitoring processes/memory usage.
...

[WB0HYQ]

Every computer is at its maximum RAM size. The laptops especially.

I read through the linked thread, but the terminology simply overwhelmed me. It was over my head mostly. I will investigate the swap situation. Actually, swap is the only thing (memory-wise) that is handled differently than Windows (I think). Since the computers all tend to freeze when a lot is going on, the swap sounds like the likely culprit since the MEMTEST went smoothly. BTW, the MEMTEST offered by GRUB will fail almost immediately, locking up the computer so that only a hard power down will let me recover. Does it use swap?

Bill

[Autodave]

I have 11 computers, some desktops and some laptops, all running Xubuntu 20.04. These range from very low powered machines to quite good ones. The only difference between yours and mine is this: all of mine had clean installs of 20.04: no upgrading to 20.04. None of mine have locked up at any time.

You may want to take one of those machines and boot it from a USB stick and run it off of that just to see if the problem persists.

I have been using Linux for about 13 years now and if there is one thing that I have learned it is this: don't waste time upgrading. Back up what you need and do a clean install.

[ajgreeny]

What graphics hardware do you have in these machines? Do they all use integrated graphic chips, eg in the CPU, or have you separate graphic cards that may need drivers to run properly?

It may be of interest if you could show us the output in all four machines of the command

inxi -Fzx

Please use Code-Tags for terminal output as it makes that output much more easily read and understood, formatting the text as seen in terminal, not as plain text when it is copied and pasted. See my signature below for a How-to

[TheFu]

Why guess. Just check the system log files.
journalctl -b -1 will show the prior boot and other hardware messages for that boot on each machine. If you know about the time it happened, you can use a filter to get close or put in a range.

If all the system have the same type of GPU, that could be a common issue too. What is common for all the systems. Lots of people use 18.04 and don't have any crashes or lockups. I generally go 2-4 weeks between reboots with my systems. No lockups during those times.

[WB0HYQ]

@Autodave: At one time or another all four machines were booted up with a Ubuntu Live DVD. Since the OS at that point is pretty bare bones, I haven't been able to get much running to try and overload it enough to cause the problem. It hasn't happened, but the time the systems were run off the DVD was rarely over a day. One of the laptops was a clean install of 20.04. In fact, the hard drive was a new one I had to format first.

@Aigreeny: I believe all four are running NVIDIA GPUs, but I will find out for sure. I used to run the NVIDIA drivers, but reverted all four to the default video drivers a while back. I will also run the command on the four and show the results. May have to wait for one of them as I loaned it out to a friend for his vacation trip and he won't be back for two weeks. He uses the Windows side, not the Linux.

@Thefu: I've run journalctl on all four machines. The results were inconclusive with none showing anything out of sorts. The log simply ends then begins with the next boot. When I ran 18.04 I could run for weeks without rebooting except for the ones commanded by the software updater. With no rebootable updates, the systems ran great. Perhaps my upgrade to 21.04 might do the trick. Time will tell.

Bill

[WB0HYQ]

Here's the first "inxi" report. This one is from my HP G60"

bill@bill-hp-g60:~$ inxi -Fzx
System:
  Kernel: 5.11.0-37-generic x86_64 bits: 64 compiler: gcc v: 10.2.1 
  Desktop: Xfce 4.16.0 Distro: Ubuntu 21.04 (Hirsute Hippo)        <<<----------------------this is the one I just upgraded from 20.04
Machine:
  Type: Laptop System: Hewlett-Packard product: HP G60 Notebook PC v: F.25 
  serial: <filter> 
  Mobo: Wistron model: 303C v: 08.45 serial: <filter> BIOS: Hewlett-Packard 
  v: F.25 date: 10/03/2008                                                       <<<<<-------- I realize this is way out of date. I have no idea how to get/install a newer BIOS.
Battery:
  ID-1: BAT0 charge: 81.1 Wh condition: 90.3/97.7 Wh (92%) 
  model: Hewlett-Packard Primary status: Discharging 
CPU:
  Info: Dual Core model: AMD Turion RM-70 bits: 64 type: MCP 
  arch: Turion X2 Ultra rev: 1 L2 cache: 1024 KiB 
  flags: lm nx pae sse sse2 sse3 svm bogomips: 7999 
  Speed: 500 MHz min/max: 500/2000 MHz Core speeds (MHz): 1: 500 2: 500     <<<<<-------------- I'd love to figure out how to increase this speed.
Graphics:
  Device-1: NVIDIA C77 [GeForce 8200M G] vendor: Hewlett-Packard 
  driver: nouveau v: kernel bus ID: 02:00.0                                                <<<<<-------- NVIDIA card, but with nouveau driver.
  Device-2: Chicony Webcam type: USB driver: uvcvideo bus ID: 2-2:2 
  Display: x11 server: X.Org 1.20.11 driver: loaded: modesetting 
  unloaded: fbdev,vesa resolution: 1366x768~60Hz 
  OpenGL: renderer: NVAA v: 3.3 Mesa 21.0.3 direct render: Yes 
Audio:
  Device-1: NVIDIA MCP72XE/MCP72P/MCP78U/MCP78S High Definition Audio 
  vendor: Hewlett-Packard driver: snd_hda_intel v: kernel bus ID: 00:07.0 
  Sound Server: ALSA v: k5.11.0-37-generic 
Network:
  Device-1: NVIDIA MCP77 Ethernet vendor: Hewlett-Packard driver: forcedeth 
  v: kernel port: 30f8 bus ID: 00:0a.0 
  IF: enp0s10 state: down mac: <filter> 
  Device-2: Qualcomm Atheros AR928X Wireless Network Adapter 
  vendor: Hewlett-Packard driver: ath9k v: kernel port: 4000 bus ID: 07:00.0 
  IF: wlp7s0 state: up mac: <filter> 
Drives:
  Local Storage: total: 232.89 GiB used: 16.53 GiB (7.1%) 
  ID-1: /dev/sdb vendor: Toshiba model: MK2552GSX size: 232.89 GiB 
Partition:
  ID-1: / size: 227.74 GiB used: 16.53 GiB (7.3%) fs: ext4 dev: /dev/sdb5 
  ID-2: /boot/efi size: 511 MiB used: 4 KiB (0.0%) fs: vfat dev: /dev/sdb1 
Swap:
  ID-1: swap-1 type: file size: 2 GiB used: 0 KiB (0.0%) file: /swapfile     <<<<<------------ according to what I've read, this should be sufficient. However, I raised it to 4G
Sensors:
  System Temperatures: cpu: 58.2 C mobo: 52.0 C gpu: nouveau temp: 62.0 C 
  Fan Speeds (RPM): N/A 
Info:
  Processes: 205 Uptime: 20m Memory: 3.59 GiB used: 935.8 MiB (25.5%)     <<<---------- only has 4G of RAM and no room for more.
  Init: systemd runlevel: 5 Compilers: gcc: N/A 
  clang: 12.0.0-3ubuntu1~21.04.2 Packages: 2771 Shell: Bash v: 5.1.4 
  inxi: 3.3.01 
bill@bill-hp-g60:~$

I also tweaked the "swappiness" value from the default of 60 to 70, to see if it would cause the OS to use swapfile more. I'm hoping this will cause the computer to freeze more often. If it does, then I've probably found my culprit and can lower the value to 30 or 40.

All the above changes were also set into "fstab" and the system config file.

There was one puzzling bit in the fstab though:

bill@bill-hp-g60:~$ cat /etc/fstab
# /etc/fstab: static file system information.
#
# Use 'blkid' to print the universally unique identifier for a
# device; this may be used with UUID= as a more robust way to name devices
# that works even if disks are added and removed. See fstab(5).
#
# <file system> <mount point>   <type>  <options>       <dump>  <pass>
# / was on /dev/sda5 during installation
UUID=1b35c21c-a55d-428a-990a-c10e218bc5c8 /               ext4    errors=remount-ro 0       1
# /boot/efi was on /dev/sda1 during installation
UUID=4646-30CD  /boot/efi       vfat    umask=0077      0       1
/swapfile                                 none            swap    sw              0       0                <<<<< ------------- this is NOT what was indicated in the inxi report above. Should I change it to:
bill@bill-hp-g60:                                                                                                                                /swapfile      swap    swap     sw       0       0             ????

Bill