Keyboard hangs after several minutes of uptime.

Forum Forums New users New Users and General Questions Keyboard hangs after several minutes of uptime.

Tagged: 

  • This topic has 15 replies, 6 voices, and was last updated May 10-4:46 am by 52midnight.
Viewing 15 posts - 1 through 15 (of 16 total)
  • Author
    Posts
  • #58496
    Member
    Avatar52midnight

    The new installation running LXDE seems to be working OK, but after several minutes of uptime the keyboard response starts to “hang”. It takes half a second or so for a keystroke in a terminal window to appear, and the delay increases until key repeats cease working, and individual keystrokes are required.

    The same effect is evident in e.g. File Open dialogues until the system becomes unusable and a reboot is required.

    Can anyone explain why this is happening and how to fix it? Any assistance much appreciated.

    #58515
    Member
    syboksybok

    I have no idea and you did not provide too much of a information to start with.
    What is missing system information (32/64-bit, antiX version, HW details); LXDE version and method of install; programs/processes/tasks running when the issue occurs etc.

    A quick internet search reveals few suggestions and potential causes.
    1) In any case, please do post inxi -Fxz
    to provide more information about the system and HW it runs on.
    2) Open terminal, type top, hit enter and observe what processes consume RAM/CPU.
    This could indicate the resources of your PC are consumed resulting into a delay, another link.
    3) In one case, sudo dmesg | tail revealed CPU was over-heating and slowed down.

    Hope that helps or at least the results of your observations and output from the steps 1)-3) will provide more information.

    • This reply was modified 1 week, 2 days ago by sybok.
    #58536
    Member
    Avatarseaken64

    The new installation running LXDE

    Hi @52midnight,

    antiX does not run a desktop environment like LXDE as default installation so we’ll need more info as sybok pointed out. Perhaps there are some forum members who run LXDE who can help. But please share the info sybok requested so we can get you pointed in the right direction.

    seaken64

    #58550
    Member
    Avatar52midnight

    Thanks for the replies. Output of inxi -Fxz:

    [code]System: Host: qt2 Kernel: 4.9.235-antix.1-amd64-smp x86_64 bits: 64 compiler: gcc v: 8.3.0 Desktop: LXDE 0.10.0
    Distro: antiX-19.3_x64-full Manolis Glezos 15 October 2020 base: Debian GNU/Linux 10 (buster)
    Machine: Type: Desktop Mobo: AMI model: Aptio CRB serial: N/A UEFI [Legacy]: American Megatrends v: 5.6.5 date: 02/25/2019
    CPU: Topology: Quad Core model: Intel Celeron J1900 bits: 64 type: MCP arch: Silvermont rev: 9 L2 cache: 1024 KiB
    flags: lm nx pae sse sse2 sse3 sse4_1 sse4_2 ssse3 vmx bogomips: 15993
    Speed: 1333 MHz min/max: 1333/2416 MHz Core speeds (MHz): 1: 1333 2: 1333 3: 1333 4: 1333
    Graphics: Device-1: Intel Atom Processor Z36xxx/Z37xxx Series Graphics & Display driver: i915 v: kernel bus ID: 00:02.0
    Display: x11 server: X.Org 1.20.4 driver: modesetting unloaded: fbdev,vesa resolution: 1920×1080~60Hz
    OpenGL: renderer: Mesa DRI Intel Bay Trail v: 4.2 Mesa 18.3.6 direct render: Yes
    Audio: Device-1: Intel Atom Processor Z36xxx/Z37xxx Series High Definition Audio driver: snd_hda_intel v: kernel
    bus ID: 00:1b.0
    Sound Server: ALSA v: k4.9.235-antix.1-amd64-smp
    Network: Device-1: Realtek RTL8111/8168/8411 PCI Express Gigabit Ethernet driver: r8169 v: 2.3LK-NAPI port: e000
    bus ID: 01:00.0
    IF: eth0 state: down mac: <filter>
    Device-2: Ralink MT7601U Wireless Adapter type: USB driver: mt7601u bus ID: 1-2.1:4
    IF: wlan1 state: up mac: <filter>
    Drives: Local Storage: total: 998.75 GiB used: 414.47 GiB (41.5%)
    ID-1: /dev/sda vendor: Seagate model: ST1000LM035-1RK172 size: 931.51 GiB
    ID-2: /dev/sdb vendor: Hoodisk model: SSD size: 59.63 GiB
    ID-3: /dev/sdc type: USB vendor: Verbatim model: Store n Go Drive size: 7.61 GiB
    Partition: ID-1: / size: 16.02 GiB used: 4.82 GiB (30.1%) fs: ext4 dev: /dev/sdb2
    ID-2: swap-1 size: 3.91 GiB used: 0 KiB (0.0%) fs: swap dev: /dev/sdb5
    Sensors: System Temperatures: cpu: 25.0 C mobo: N/A
    Fan Speeds (RPM): N/A
    Info: Processes: 195 Uptime: 10m Memory: 3.69 GiB used: 617.2 MiB (16.3%) Init: SysVinit runlevel: 5 Compilers:
    gcc: 8.3.0 Shell: bash v: 5.0.3 inxi: 3.0.36
    [/code]

    I’ve checked running processes and nothing is consuming system resources. The CPU is about 96% idle, 600MB of RAM are used from 3G6 available, and swap is unused.

    CPU core temperature is normal according to the last dmesg posts.

    #58554
    Member
    XecureXecure

    It is probably not related, but you can try editing /etc/hostname and /etc/hosts so they both match the same hostname (qt2).
    I saw that some user had issues in their computers (as real slowness) when running sudo commands, as they would check the hostname.

    Also, if not using a US keyboard, did you set the keyboard layout using a boot parameter, an LXDE program or one of keyboard tools in antiX?

    #58561
    Moderator
    christophechristophe

    It is probably not related, but you can try editing /etc/hostname and /etc/hosts so they both match the same hostname (qt2).
    I saw that some user had issues in their computers (as real slowness) when running sudo commands, as they would check the hostname.

    Yes, I can confirm this was (is!) happening to me. Work that angle, even if it only partially helps your situation. (Thanks for that tip, Xecure!)

    #58583
    Member
    Avatar52midnight

    > try editing /etc/hostname and /etc/hosts

    Interesting suggestion, but no, unfortunately. Both have the same Hostname.

    #58605
    Member
    syboksybok

    Hi, thanks for providing the additional information.

    CPU 96% idle = only LXDE running (and perhaps terminal/system-monitor), only 4% of CPU used… sounds good.
    May be, inspecting some logs in ‘/var/log/’ (‘dmesg’, ‘Xorg.0.log’, ‘syslog’ and their older versions) and finding at least one occurrence of the freeze would/could shed some light on this.
    1) Suggestion: copy the log, open in a text editor and try to find the time when you know the error has occurred.
    If you find anything suspicious/interesting (such as an error), please post.
    2) How did you install LXDE and what is the current version? (I assume that using package manager.)
    3) Did you have any other systems installed [on the same device] prior to antiX? If yes, how did they perform?
    Motivation:
    I had a mother board that was plagued by freezes once (upgrade after a failed flashing of BIOS); web-search suggested rearrange RAM in the slots and other things but I gave up and bought a new PC.
    4) Could you try the other desktops in antiX to see whether the freeze happens there as well?

    I did a quick web-search:
    LXDE "Intel Celeron" freeze
    with the following results (which do not seem to be of much help but you may try them anyway):
    A) Install intel-microcode suggested via this link.
    B) Freeze on shutdown/reboot, suggested change in EFI setting… probably not related.

    • This reply was modified 1 week, 1 day ago by sybok. Reason: Question 3 improved
    #58607
    Member
    Avatar52midnight

    > I had a mother board that was plagued by freezes once

    Thanks for an interesting reply sybok. Yes, I’ve struck this SOMEWHERE before, but after running various Lx distros across two decades I’ve no hope of remembering where. It LOOKS odd and behaves queerly. The only correlation I’ve (possibly) noticed is with using xset to reset keyboard rate and delay. I use a bash alias to reset these, since on AntiX and vanilla Debian they seem to get reset to system defaults at random, and I’ve not found time or motivation to find out how to change them at origin:

    alias kbd=’xset r rate 240 35′

    Your suggestions will take a few days to implement reliably, so give me that to come back with results.

    #58728
    Member
    Avatar52midnight

    I’ve found something rather unusual: several kworker processes that can’t be removed by either slay or kill. I’m not savvy enough to know just what these might be (I assume kernel daemons of some sort) but if they’re orphaned and still active they might be the cause of the problem, especially if they were or are keyboard-related.

    I discovered them after pursuing sybok’s suggestion to look through the system logs, which showed up nothing unusual, although I’m not sufficiently familiar with them to be certain as yet. Here are the last few lines of ps ax. You can see the eight terminal tabs open in xfce4-terminal and the three instances of Firefox that are part of my initial Desktop setup, then two instances of Midnight Commander, then a series of kworkers for which I have no explanation:

    2716 pts/0 Ss 0:00 bash
    2717 pts/1 Ss 0:00 bash
    2728 pts/2 Ss 0:00 bash
    2730 pts/3 Ss+ 0:00 bash
    2731 pts/4 Ss+ 0:00 bash
    2747 pts/5 Ss+ 0:00 bash
    2763 pts/6 Ss+ 0:00 bash
    2769 pts/7 Ss+ 0:00 bash
    2809 ? S 0:00 /usr/lib/bluetooth/obexd
    2843 ? Sl 3:11 /usr/lib/firefox-esr/firefox-esr …
    2984 ? Sl 0:36 /usr/lib/firefox-esr/firefox-esr …
    3018 ? Sl 0:05 /usr/lib/firefox-esr/firefox-esr …
    5365 ? Ss 0:00 /usr/sbin/cupsd -C /etc/cups/cups…
    6727 ? Sl 0:00 /usr/lib/firefox-esr/firefox-esr …
    6762 pts/0 S+ 0:00 mc -b /ace/www /ace
    6764 pts/8 Ss+ 0:00 bash -rcfile .bashrc
    6910 pts/1 S+ 0:01 mc -S mc46.ini /ace/www /ace
    6912 pts/9 Ss+ 0:00 bash -rcfile .bashrc
    6931 ? S 0:00 [kworker/0:0]
    6945 ? S 0:00 [kworker/u8:2]
    6956 ? S 0:00 [kworker/1:0]
    6960 ? S 0:00 [kworker/3:0]
    7011 ? S 0:00 [kworker/2:1]
    7021 ? Sl 0:00 /usr/lib/firefox-esr/firefox-esr …
    7099 ? S 0:00 [kworker/u8:1]
    7104 ? S 0:00 [kworker/1:1]
    7111 ? S 0:00 [kworker/0:1]
    7116 ? S 0:00 [kworker/2:0]
    7117 ? S 0:00 [kworker/3:2]
    7121 ? S 0:00 [kworker/1:2]
    7122 ? S 0:00 [kworker/u8:0]
    7124 ? S 0:00 [kworker/0:2]
    7133 pts/2 R+ 0:00 ps ax

    #58734
    Member
    XecureXecure

    If you suspect that the kernel may be the reason (as kworker is a kernel worker process), try updating to a 4.19.xxx or 5.10.xxx kernel. It may be related to the kernel taking too long to process info from a pci/usb port, which may lead to input-lag after some up time.
    You can use the packageinstaller, synaptic or cli-aptiX to install a different kernel.

    #58752
    Member
    Avatar52midnight

    Thanks. Yes, sounds likely. At present I’m using 4.9.235-antix.1-amd64-smp.

    Before making the change, is there any way of determining what launched these processes? Here’s ps ax just after boot:

    2653 ? Rl 0:01 xfce4-terminal –tab –tab –tab –tab –tab –tab –tab –hide-menubar
    2657 ? Sl 0:00 cmst -m -w5
    2660 ? Sl 0:00 /usr/lib/at-spi2-core/at-spi-bus-launcher –launch-immediately
    2665 ? S 0:00 /usr/bin/dbus-daemon –config-file=/usr/share/defaults/at-spi2/accessibility.conf …
    2680 ? Sl 0:00 /usr/lib/at-spi2-core/at-spi2-registryd –use-gnome-session
    2682 ? Sl 0:00 /usr/lib/menu-cache/menu-cached /run/user/0/menu-cached-:0
    2714 pts/0 Ss 0:00 bash
    2715 pts/1 Ss 0:00 bash
    2718 pts/2 Ss+ 0:00 bash
    2719 pts/3 Ss+ 0:00 bash
    2721 pts/4 Ss+ 0:00 bash
    2735 pts/5 Ss+ 0:00 bash
    2738 pts/6 Ss+ 0:00 bash
    2742 pts/7 Ss+ 0:00 bash
    2803 ? S 0:00 /usr/lib/bluetooth/obexd
    2839 ? Sl 0:10 /usr/lib/firefox-esr/firefox-esr -contentproc -childID 1 -isForBrowser …
    2959 ? Sl 0:00 /usr/lib/firefox-esr/firefox-esr -contentproc -childID 4 -isForBrowser …
    2983 ? Sl 0:03 /usr/lib/firefox-esr/firefox-esr -contentproc -childID 5 -isForBrowser …
    3016 ? Sl 0:01 /usr/lib/firefox-esr/firefox-esr -contentproc -childID 6 -isForBrowser …
    3253 ? Sl 0:00 /usr/lib/dconf/dconf-service
    3259 pts/0 S+ 0:00 mc -b /ace/www /ace
    3261 pts/8 Ss+ 0:00 bash -rcfile .bashrc
    3296 ? S 0:00 [scsi_eh_2]
    3297 ? S< 0:00 [scsi_tmf_2]
    3299 ? S 0:00 [usb-storage]
    3301 ? S< 0:00 [uas]
    3303 ? S< 0:00 [bioset]
    3343 ? S 0:00 [jbd2/sdc1-8]
    3344 ? S< 0:00 [ext4-rsv-conver]
    3358 pts/1 R+ 0:00 ps ax

    #58755
    Member
    XecureXecure

    launch htop and use F5 option to see tree-mode. Maybe you can see a hierarchy for the specific processes you mention.
    If you don’t want to see it real-time, then pipe to a file this command
    ps auxf > processes.txt
    This should also display process hierarchy much better than htop

    #58760
    Moderator
    Brian MasinickBrian Masinick

    I’m not 100% certain of this (so it is not necessarily accurate). One of the recent Version 5 kernels – not positive, but I THINK that it may have been something earlier in the 5.10 kernel series, maybe something around 5.10.15 to 5.10.22 – I definitely have NOT seen this at all since then, and never in the history of antiX for any other kernel or system configuration, I did see a few system freezes that I don’t attribute to just the keyboard, they literally froze the system. Whatever the version or symptoms WERE, they are now completely gone.

    I have two hard drive instances of antiX running on my primary laptop and I have several USB copies ranging over time and releases. Those FEW instances went away before I had a chance to get a convincing, repeatable, reproducible scenario to share in a defect report.

    Take this for what it’s worth, frankly not much since I have been unable to substantiate anything. Just noting that while RARE, failures are sometimes possible. More often than not, they represent hardware starting to fail, but software failures are also POSSIBLE. If we can diagnose, and come up with steps that reproduce a problem, it helps to further reduce any possibility of frequent failures.

    Brian Masinick

    #58786
    Member
    Avatar52midnight

    Most grateful for ongoing assistance. The new box is not yet in full-time use (second machine) so I’ll take my time and try to capture something useful. May then UNDERSTAND the problem instead of just making it go away.

    Nothing worse than applying a ‘fix’ only to find that the problem’s still there, just shows up less frequently.

Viewing 15 posts - 1 through 15 (of 16 total)
  • You must be logged in to reply to this topic.