Xorg crashes when under memory pressure

padota1663 · 2024-12-04 22:31

Hi,

I have (sadly) another issue ongoing, as the title says in some occasions Xorg crashes abruptly, usually when my PC is under memory pressure, typically this happens when over 80% RAM & 30-40% swap simultaneously.

Is this a known bug of Xorg, or something unheard of? Should I avoid putting pressure on RAM, even if this means only opening no more than 3-4 Firefox tabs and nothing else? (I only have 4 GB and cannot install more, so I compensated extending swap to 8 GB to avoid OOM, really old PC, so I dunno how can I avoid this condition during my normal PC usage)

I attached the portion of /var/log/syslog which shows the error, if this bug should be reported to the Xorg guys and not here let me know.

TIA

Attachments:

crash xorg, Size: 9.4 KiB, Downloads: 26

bin · 2024-12-05 08:02

Possibly you have a bad chip.

Please copy and paste the output of

sudo inxi -Fxxxmprz

sudo is used to get the RAM report.

q4osteam · 2024-12-05 08:18

@padota1663
This sounds more like a hardware problem, such as bad memory or at least hardware incompatibility. A bug in Xorg is highly unlikely.

padota1663 · 2024-12-06 20:46

bin wrote:

Possibly you have a bad chip.
Please copy and paste the output of
sudo inxi -Fxxxmprz
sudo is used to get the RAM report.

root@scatorcio-di-merda:~# sudo inxi -Fxxxmprz
System:
  Kernel: 6.1.0-28-amd64 arch: x86_64 bits: 64 compiler: gcc v: 12.2.0 Console: pty pts/2 DM:
    1: TDM 2: XDM Distro: Q4OS 5.7.1-n1 base: Debian GNU/Linux 12 (bookworm)
Machine:
  Type: Desktop Mobo: ASUSTeK model: A8N-E v: 2.XX serial: <filter> BIOS: Phoenix v: ASUS A8N-E
    Revision 1013 date: 04/07/2006
Memory:
  RAM: total: 3.82 GiB used: 2.44 GiB (63.9%)
  Array-1: capacity: 4 GiB slots: 4 EC: None max-module-size: 1024 MiB
  Device-1: A0 info: double-bank type: N/A size: 1024 MiB speed: 400 MT/s volts: N/A
    width (bits): data: 64 total: 64 manufacturer: N/A part-no: N/A serial: N/A
  Device-2: A1 info: double-bank type: N/A size: 1024 MiB speed: 400 MT/s volts: N/A
    width (bits): data: 64 total: 64 manufacturer: N/A part-no: N/A serial: N/A
  Device-3: A2 info: double-bank type: N/A size: 1024 MiB speed: 400 MT/s volts: N/A
    width (bits): data: 64 total: 64 manufacturer: N/A part-no: N/A serial: N/A
  Device-4: A3 info: double-bank type: N/A size: 1024 MiB speed: 400 MT/s volts: N/A
    width (bits): data: 64 total: 64 manufacturer: N/A part-no: N/A serial: N/A
CPU:
  Info: single core model: AMD Athlon 64 3500+ bits: 64 type: UP smt: <unsupported> arch: K8
    rev: 2 cache: L1: 128 KiB L2: 512 KiB
  Speed (MHz): 2217 min/max: N/A volts: 1.4 V ext-clock: 216 MHz core: 1: 2217 bogomips: 4433
  Flags: lm nx pae sse sse2 sse3
Graphics:
  Device-1: AMD RV516 [Radeon X1300/X1550 Series] driver: radeon v: kernel arch: Rage-7 pcie:
    speed: 2.5 GT/s lanes: 16 ports: active: DVI-I-1 empty: DVI-I-2,SVIDEO-1 bus-ID: 01:00.0
    chip-ID: 1002:7183 class-ID: 0300
  Display: x11 server: X.org v: 1.21.1.7 driver: X: loaded: radeon
    unloaded: fbdev,modesetting,vesa dri: r300 gpu: radeon tty: 142x45
  Monitor-1: DVI-I-1 model: HP L1502 serial: <filter> res: 1024x768 dpi: 84
    size: 304x228mm (11.97x8.98") diag: 386mm (15.2") modes: max: 1024x768 min: 720x400
  API: OpenGL Message: GL data unavailable in console for root.
Audio:
  Device-1: NVIDIA CK804 AC97 Audio vendor: ASUSTeK K8N4/A8N Series Mainboard driver: snd_intel8x0
    v: kernel bus-ID: 00:04.0 chip-ID: 10de:0059 class-ID: 0401
  API: ALSA v: k6.1.0-28-amd64 status: kernel-api
  Server-1: JACK v: 1.9.21 status: off
  Server-2: PulseAudio v: 16.1 status: active (root, process)
Network:
  Device-1: NVIDIA CK804 Ethernet vendor: ASUSTeK K8N4/A8N Series Mainboard type: network bridge
    driver: forcedeth v: kernel port: b000 bus-ID: 00:0a.0 chip-ID: 10de:0057 class-ID: 0680
  IF: enp0s10 state: up speed: 1000 Mbps duplex: full mac: <filter>
Drives:
  Local Storage: total: 421.89 GiB used: 125.55 GiB (29.8%)
  ID-1: /dev/sda vendor: Western Digital model: WD800JD-75MSA3 size: 74.51 GiB speed: 3.0 Gb/s
    type: N/A serial: <filter> rev: 1E04 scheme: MBR
  ID-2: /dev/sdb vendor: Seagate model: ST3250820AS size: 232.89 GiB speed: 3.0 Gb/s type: N/A
    serial: <filter> rev: D scheme: GPT
  ID-3: /dev/sdc vendor: Maxtor model: 6Y120M0 size: 114.5 GiB speed: 1.5 Gb/s type: N/A
    serial: <filter> rev: 11W0 scheme: MBR
Partition:
  ID-1: / size: 65.45 GiB used: 43.6 GiB (66.6%) fs: ext4 dev: /dev/sda1
  ID-2: /home/<filter>/Scaricati/sdb size: 228.17 GiB used: 55.63 GiB (24.4%) fs: ext4
    dev: /dev/sdb1
  ID-3: /home/<filter>/Scaricati/sdc size: 112.14 GiB used: 26.11 GiB (23.3%) fs: ext4
    dev: /dev/sdc1
Swap:
  ID-1: swap-1 type: partition size: 7.45 GiB used: 217.6 MiB (2.9%) priority: -2 dev: /dev/sda2
Sensors:
  System Temperatures: cpu: 63.0 C mobo: 48.0 C
  Fan Speeds (RPM): cpu: 3013 case-1: 0
  Power: 12v: N/A 5v: N/A 3.3v: 3.28 vbat: N/A
Repos:
  Packages: 2803 pm: dpkg pkgs: 2771 pm: flatpak pkgs: 24 pm: snap pkgs: 8
  No active apt repos in: /etc/apt/sources.list
  Active apt repos in: /etc/apt/sources.list.d/10_q4os.list
    1: deb http://q4os.org/q4repo/ q4os-5-0-cn main
  Active apt repos in: /etc/apt/sources.list.d/12_qtde.list
    1: deb http://q4os.org/qtderepo/ bookworm main
  Active apt repos in: /etc/apt/sources.list.d/20_debian.list
    1: deb http://deb.debian.org/debian/ bookworm main contrib non-free non-free-firmware
    2: deb http://security.debian.org/ bookworm-security main contrib non-free non-free-firmware
    3: deb http://deb.debian.org/debian/ bookworm-updates main contrib non-free non-free-firmware
  No active apt repos in: /etc/apt/sources.list.d/30_debian_backports.list
  Active apt repos in: /etc/apt/sources.list.d/50_q4os-chrome.list
    1: deb [arch=amd64] http://q4os.org/qextrepo/ bookworm-chrome-cn main
  Active apt repos in: /etc/apt/sources.list.d/50_q4os-firefox.list
    1: deb [signed-by=/usr/share/keyrings/q4a-mozilla.gpg] https://packages.mozilla.org/apt/ mozilla main
    2: deb http://q4os.org/qextrepo/ bookworm-firefox-cn main
  Active apt repos in: /etc/apt/sources.list.d/archive_uri-https_boinc_berkeley_edu_dl_linux_stable_bookworm-bookworm.list
    1: deb https://boinc.berkeley.edu/dl/linux/stable/bookworm bookworm main
  Active apt repos in: /etc/apt/sources.list.d/google-chrome.list
    1: deb [arch=amd64] https://dl.google.com/linux/chrome/deb/ stable main
Info:
  Processes: 193 Uptime: 2d 18h 24m wakeups: 0 Init: systemd v: 252 target: graphical (5)
  default: graphical Compilers: gcc: 12.2.0 alt: 12 Shell: Bash (su) v: 5.2.15
  running-in: pty pts/2 inxi: 3.3.26
root@scatorcio-di-merda:~#

q4osteam wrote:

@padota1663
This sounds more like a hardware problem, such as bad memory or at least hardware incompatibility. A bug in Xorg is highly unlikely.

Should I try memtest or something else? Like some specific stress test? Or should I try reseating components inside the PC?

Last edited by padota1663 (2024-12-06 20:48)

bin · 2024-12-08 11:42

I don't think there is a single answer to this.
First of all - this is old hardware.
This is your mobo https://theretroweb.com/motherboards/s/asus-a8n-e
Your hard drives are not as young as they were.

So, reseating components is always a good start. Make sure you take antistatic precautions and do not handle connector surfaces - especially on the RAM chips. It depends on what your case is like inside as to what you need to do. Nornally with DIMMS I used to just eject, check the contact surfaces looked clean and re-insert.
If there is dust in the case then you have to be careful it doesn't get into the slots when the chips are ejected. I don't like using air cans for blowing out sockets - I use a small suck/blow vacuum cleaner that is used for nothing else except blowing! The power supply is usually full of dust, and the CPU cooling fan.
When it comes to cleaning the CPU fan and heat sink you should make sure the fan is not allowed to spin while you are blowing air through. A finger will do, stop, move the blades and then blow a bit more.
The SATA cable connectors on your mobo may or may not be coloured. Big mfrs used to have their slots colour coded to reduce assembly line and end use connection errors. Your boot drive should be on SATA 1 and then other drives as required. If you have a SATA optical drive put it on SATA 4 - just keeps it out of the way.

That lot may or may not have made any difference. The next area is a bit less easy.
I've just been running Q4OS TDE on an Acer laptop from around 2009 - it has 4gb RAM.
With Firefox and 6 tabs including Amazon, BBC home and a few others it shows a 1.5gb used under htop.
These days RAM usage has become harder to measure and so it is not easy to know what is true. Under inxi it shows as more and other tools show it differently.

You say that Xorg crashes. I take it you do not get any errors on screen - it just dumps you out to a command line and that's all. It's a good idea to try to copy .xsession-errors as that may tell more about what happened. That file is renamed to .xession-errors.old on restart so it will be available after a reboot. You should copy to a different place and rename it to xession-errors.txt.

I would install gsmartcontrol and use it to test/check your drives for bad sectors/errors/end of life. The WD boot drive is probably the original and is therefore in its late teens. Since you have your swap file on that drive there is room for problems if that is failing. I would be inclined to put it on the newest/least used drive to see if that helps.

padota1663 · 2024-12-09 02:07

bin wrote:

I don't think there is a single answer to this.
First of all - this is old hardware.
This is your mobo https://theretroweb.com/motherboards/s/asus-a8n-e
Your hard drives are not as young as they were.
So, reseating components is always a good start. Make sure you take antistatic precautions and do not handle connector surfaces - especially on the RAM chips. It depends on what your case is like inside as to what you need to do. Nornally with DIMMS I used to just eject, check the contact surfaces looked clean and re-insert.
If there is dust in the case then you have to be careful it doesn't get into the slots when the chips are ejected. I don't like using air cans for blowing out sockets - I use a small suck/blow vacuum cleaner that is used for nothing else except blowing! The power supply is usually full of dust, and the CPU cooling fan.
When it comes to cleaning the CPU fan and heat sink you should make sure the fan is not allowed to spin while you are blowing air through. A finger will do, stop, move the blades and then blow a bit more.
The SATA cable connectors on your mobo may or may not be coloured. Big mfrs used to have their slots colour coded to reduce assembly line and end use connection errors. Your boot drive should be on SATA 1 and then other drives as required. If you have a SATA optical drive put it on SATA 4 - just keeps it out of the way.
That lot may or may not have made any difference. The next area is a bit less easy.
I've just been running Q4OS TDE on an Acer laptop from around 2009 - it has 4gb RAM.
With Firefox and 6 tabs including Amazon, BBC home and a few others it shows a 1.5gb used under htop.
These days RAM usage has become harder to measure and so it is not easy to know what is true. Under inxi it shows as more and other tools show it differently.
You say that Xorg crashes. I take it you do not get any errors on screen - it just dumps you out to a command line and that's all. It's a good idea to try to copy .xsession-errors as that may tell more about what happened. That file is renamed to .xession-errors.old on restart so it will be available after a reboot. You should copy to a different place and rename it to xession-errors.txt.
I would install gsmartcontrol and use it to test/check your drives for bad sectors/errors/end of life. The WD boot drive is probably the original and is therefore in its late teens. Since you have your swap file on that drive there is room for problems if that is failing. I would be inclined to put it on the newest/least used drive to see if that helps.

Hi bin,

first of all, thanks for taking the time to write your lengthy answer, mine will be too.

I know how to reseat components and in general, how to assemble / disassemble a PC and clean it as I've been doing it since I was young, in fact this one was assembled in April with old pieces lying around (yeah, I know it's quite old, but it still works so I tried to put it to doing something meaningful till it's able to run a modern and patched / up to date distro); the inside of the machine is spotless as is the room where I have all my PCs, I clean all of them once a month with an industrial grade PC air duster (I used my air compressor before buying the duster a year ago), obviously taking care to block fans, I know the kind of damage an overspinning PC fan can do to the electronics it's attached to; I also cleaned every piece when I assembled it some months ago, even the inside of the PSU by opening it (the warranty is void since long ago) and blocking its fan too; in some cases, on the contacts I spray isopropyl alcohol to fully clean them, then let them dry completely (not on these low value components for now).

The mobo didn't have swollen capacitors, so should be fine, but I should recheck.

As for the hard drives, I already have installed gsmartcontrol since the beginning and did check it regularly, and I also launched SMART self test on all drives.

The WDC boot drive, connected at SATA 1 and mounted at / (the 80 Gb one) is perfectly healthy at 35321 power on hours, passed SMART extended self-test two months ago, and has no errors at all in the attributes section and in the error log, even the temp is inside the limit, so is the best of the three.

The second drive, the 250 GB Seagate connected at SATA 2 and mounted at /home/a/Scaricati/sdb/ has 39509 hours, the only thing to note is some old errors dating back to when it had 9707 hours (before I owned it), which are titled "Uncorrectable error in data" and their output is like this:

Error 1445 [4] occurred at disk power-on lifetime: 9707 hours (404 days + 11 hours)
  When the command that caused the error occurred, the device was active or idle.

  After command completion occurred, registers were:
  ER -- ST COUNT  LBA_48  LH LM LL DV DC
  -- -- -- == -- == == == -- -- -- -- --
  40 -- 51 00 00 00 00 51 5e 60 5e e0 00  Error: UNC at LBA = 0x515e605e = 1365139550

  Commands leading to the command that caused the error were:
  CR FEATR COUNT  LBA_48  LH LM LL DV DC  Powered_Up_Time  Command/Feature_Name
  -- == -- == -- == == == -- -- -- -- --  ---------------  --------------------
  25 00 d8 00 08 00 00 51 00 60 57 e0 00     00:21:39.502  READ DMA EXT
  25 00 d8 00 01 00 00 00 00 00 00 e0 00     00:21:39.502  READ DMA EXT
  25 00 d8 00 08 00 00 51 00 60 57 e0 00     00:21:39.501  READ DMA EXT
  25 00 d8 00 08 00 00 9e 00 00 17 e0 00     00:21:39.499  READ DMA EXT
  25 00 d8 00 08 00 00 9e 00 00 0f e0 00     00:21:39.499  READ DMA EXT

I dont' know what that error code means, googling around gives back nothing, but I think they're not relevant, having them took place so backward in time; the drive passed SMART complete self test a month ago so I take it's fine at least for testing stuff or temporary storage of non important data.

The third drive, the Maxtor 120 GB connected at SATA 3 and mounted at /home/a/Scaricati/sdc/ can maybe considered a bit concerning, it behaves strangely in gsmartcontrol, the power on time value resets itself at random times, right now it shows 532 hours of power on time, that doesn't seem right to me honestly, as the drives, like the other ones, has at least 15 years of lifetime, if not nearing 20, and the PC has been on 24/7 since April or May; it shows a warning under "Soft read error rate" which is low, but not 0, and the value changes at every refresh; it also has old errors in the error log, named "Interface CRC error", like it was connected with a bad cable before I owned it, their content is:

Error 22 occurred at disk power-on lifetime: 2856 hours (119 days + 0 hours)
  When the command that caused the error occurred, the device was in an unknown state.

  After command completion occurred, registers were:
  ER ST SC SN CL CH DH
  -- -- -- -- -- -- --
  84 51 08 98 79 03 40  Error: ICRC, ABRT at LBA = 0x00037998 = 227736

  Commands leading to the command that caused the error were:
  CR FR SC SN CL CH DH DC   Powered_Up_Time  Command/Feature_Name
  -- -- -- -- -- -- -- --  ----------------  --------------------
  ca ff 08 98 79 03 40 00      07:44:09.680  WRITE DMA
  40 00 00 00 01 00 40 00      07:44:09.664  READ VERIFY SECTOR(S)
  ef 03 46 00 00 00 00 00      07:44:06.592  SET FEATURES [Set transfer mode]
  ef 03 0c 00 00 00 00 00      07:44:06.592  SET FEATURES [Set transfer mode]
  ec 00 00 00 00 00 00 00      07:44:06.592  IDENTIFY DEVICE

I only use this drive for testing, it's always empty otherwise, as I don't trust it, even if it did pass a complete SMART self test without errors like the other two; in fact, I don't trust fully even the other two, maybe a bit more the WDC as is the most sane one, cause they didn't belong to me from the start, they were gifted from a colleague; I use this PC for non important stuff and testing so I don't fear losing data, all the important stuff is elsewhere; I don't have an optical drive in none of my PCs since at least 10 years ago, and even back then I wasn't using it since about 5 years prior; SATA cables are fine, I used them before in other builds and for other tests, they are well inserted on both sides.

I'm quite confident RAM sticks are well inserted, in fact I had to reseat them at least a dozen of times when I assembled the PC, as the full capacity of some sticks wasn't recognized at first, dunno why honestly; I also checked the contacts on the mobo and the sticks back then, they were fine, no bent pins or something alike; ran memtest for 48 hours afterwards and no errors; I'd avoid to reseat them unless absolutely necessary (tell me if you think it absolutely needs to be done), I fear I'll find myself in the same situation of needing to reseat them multiple times for them to be read correctly.

Regarding Xorg, I didn't found the .xsession-errors.old, I only have the .xsession-errors, I copied it and renamed it .xsession-errors.txt, I attached it to this message for you and others to check.

I wasn't thrown out to a command line when the screen froze, so what you say about Xorg crashing makes me think it maybe didn't crash but something else happened, I have to point out that sometimes, when everything came to a standstill, I still saw activity on the HDD LED of the PC, some other times I didn't.

I don't have (never had, to be precise) installed SSHD at the moment, so when the screen froze I couldn't check if the PC was still SSHable from outside (only the screen was frozen) and maybe TDM or Xorg could be restarted, or if it was gone completely haywire.

To check RAM usage I'm comfortable using KSensors which was preinstalled and to me is extremely clear to read and interpret, especially for RAM and swap usage; ATM I'm between 2375 and 2570 MBs approx. of 3915 available to the OS (4096 total installed) for RAM and 2321 MBs of 7629 total for swap, I have firefox open with four tabs, this one of the Q4OS Forum, WhatsApp Web (quite heavy for this old machine), my webmail and another forum, then a root terminal with journalctl --follow to check if weird stuff happens in the background, Tartube (graphical front end for yt-dlp) downloading a whole channel of more than 5000 videos right now and a bunch of other stuff including BOINC in the background which pegs my CPU at 100% all the time (and helps me heating up my room in the winter ); right now all is good, the PC runs fine, only I dunno when it will crap itself, as it happens completely random.

I have to admit I'm running a slight overclock since I built the PC and before installing Q4OS, but it ran fine since then, in fact when I installed everything I managed to reach 1625 hours (which amounts to about a month and a half, IIRC) without rebooting a single time since I finished installing, then it froze up.

I'll try putting pressure on RAM again by opening more stuff in FF and programs and see if it hangs again, if it does I'll remove the overclock and repeat the test, to rule it out.

If there are other suggestions other than check that everything is well seated (I'll recheck everything as soon as I find the time) they're welcome.

Good night everyone

Last edited by padota1663 (2024-12-09 02:18)

Attachments:

xsession-errors.txt, Size: 181.78 KiB, Downloads: 19

bin · 2024-12-09 08:36

I know the kind of damage an overspinning PC fan can do to the electronics it's attached to

So few people seem to know about this........

I think you've got the hardware covered pretty well here! The fact you can run 1625 hours is impressive. If it crashes again the question is whether you can switch to a different tty session - normally X runs on tty7. This is either via Alt+f2 or Ctrl+Alt+f2. This would at least allow you to access logs of the crash state.

Bit puzzled about the .xsession-errors.old as this is standard behaviour in all linux operations. When X starts it looks for .xsession-errors and copies it to .xsession-errors.old If that is not happening then that is a puzzle. The other thing is that TDE and KDE are really, really verbose when it comes to .xsession-errors and write tons of output to the file. In 1625 hours I do wonder if that file got so big it was part of the reason for the freeze. These days I have that file write to /tmp and /tmp is mounted in RAM via tmpfs so in other words it doesn't touch the hard drive. OK it gets lost on reboot but there's just so much dross in it I don't worry on a day top day basis.

Beyond that, swap file on a spinner is going to be so slow I wonder if the system just gets tripped up. The obvious answer is to bung in a couple of SSDs and see how it goes - reserve one just for the swap file.

padota1663 · 2024-12-10 00:41

Hi bin,

thanks for reassuring me about my hardware coverage and run time, in fact I always think I'm not detailed enough when explaining stuff to people!

Regarding the .xsession-errors.old file, I checked my other machine (FYI slightly more recent by 1-2 years and powerful hardware, 2 core AMD Athlon X64, 8 GB DDR2, integrated nVidia GeForce 7025 GPU inside the nForce 630a chipset, in contrast to this older one which has no iGPU but a PCI-E ATI X1300 Pro), which runs Q4OS with TDE just like this and the situation is the same, no .xsession-errors.old file in it either, but in both machines I notice an empty .xsession-errors-:1 file, dunno what it's about.

Thanks for the tip on the verbosity of TDE on that file, the file size reached in 1625 hours can be one of the many possible causes which lead to the crash / freeze, even considering that the main drive is only 80 GB (effective size for / is 67 GB, 18,8 free right now); can this file be limited / truncated in size, just like system logs, to avoid it reaching monstrous sizes?

I'm also well aware of the slowness of a spinning drive vs. a SSD when they deal with highly random read/write like paging, I went with spinning rust because it was free and, testing initially this machine with a spare SSD, back then I noticed the only benefits were slightly faster boot times and programs loading, anything else, especially web browsing and video download and conversion via Tartube / yt-dlp is handicapped by the CPU lack of power, even network file transfers don't reach the full Gigabit speed of the integrated NIC (they almost do on the other Q4OS PC, in that case the limit are the drives' transfer speeds); on top of that, Linux is notorious (at least in my experience) to deal badly with the combination of heavy swapping and spinning drives, to the point the system completely halts because it dedicates all its resources to swapping, making even SSHing into the affected machine impossible; maybe I should try dedicating an SSD to paging if the machine crashes again and disabling temporarily the swap partition on the mail drive; ATM everything is fine, I opened up more stuff since yesterday and nothing else (aside a bit more slowness) is happening, nothing is present in dmesg or /var/log/syslog; I'll keep you updated on how it goes.

Thanks again for your patience.

Last edited by padota1663 (2024-12-10 01:04)

bin · 2024-12-10 08:53

There are various ways of disabling or managing .xsession-errors.
This may help https://dev1galaxy.org/viewtopic.php?id=6129
Potentially you could either just prevent it being written in the first place, or mount it on one of your larger drives where it can grow. The file with the :1 indicates an incremented version of the file - in other words it got very big.

padota1663 · 2024-12-14 17:52

Thanks bin,

I read that article, and also https://dev1galaxy.org/viewtopic.php?pid=12265#p12265 and https://www.daniloaz.com/en/how-to-prev … huge-size/ which were contained inside the discussion.

I also recalled the probable cause of a possible (I didn't check at the time) gigantic increase in size of the .xsession-errors file: back then, I installed Boinc via the default repos, but it turned out, after a lot of googling about it spamming:

"Authorization required, but no authorization protocol specified."

every second in syslog, that the version contained in the default repos was old and meant for Android, so I followed the Boinc instructions to install the most up to date version from their own repo, and that message went away (it was hogging the syslog, making it unreadable and causing it to rotate frequently because of its size).

Probably, being the aforementioned message about a display issue (Boinc trying to access the display without success), it was also spamming .xsession-errors, and as it doesn't rotate by default, it literally exploded.

I also verified on both my machines that .xsession-errors:-1 was last modified on 24/8/2024 on both PCs (which is when they crashed the first time), on this PC at 03:24 and on the other at 18:41, this is another clue that probably .xsession-errors exploded in size until it filled my / partition (yes, aside the separate /swap partition and the other two disks mounted under /home/a/Scaricati/sd* only for file storage I have everything under /, not so smart I know) and everything crashed.

Right now my .xsession-errors files are at 480 Kbytes on this PC and at 42,5 MB on the other one (it's full of messages about a Chromium zombie process; Chromium was uninstalled automatically some weeks ago, during an auto update round, but I didn't use it, so I don't know how can it be running); I'll keep an eye on both of them, and evaluate to implement one of the solutions listed on the URLs above if they start growing excessively (42,5 MB for the second PC is already concerning to me, being it a text file).

For now, I'll consider the issue resolved then, I'll report back if it crashes again; I'll also install OpenSSH to check if the PC is still responsive in case of a crash, and only the GUI went haywire.

Thanks bin for all your support.

All the best
padota1663

Q4OS Forum

#1 2024-12-04 22:31

Xorg crashes when under memory pressure

#2 2024-12-05 08:02

Re: Xorg crashes when under memory pressure

#3 2024-12-05 08:18

Re: Xorg crashes when under memory pressure

#4 2024-12-06 20:46

Re: Xorg crashes when under memory pressure

#5 2024-12-08 11:42

Re: Xorg crashes when under memory pressure

#6 2024-12-09 02:07

Re: Xorg crashes when under memory pressure

#7 2024-12-09 08:36

Re: Xorg crashes when under memory pressure

#8 2024-12-10 00:41

Re: Xorg crashes when under memory pressure

#9 2024-12-10 08:53

Re: Xorg crashes when under memory pressure

#10 2024-12-14 17:52

Re: Xorg crashes when under memory pressure

Board footer