You are not logged in.
Pages: 1
Hello, I installed Arch with KDE on my Thinkpad t14S, it looks like everything is working, but I am noticing that after waking from sleep the system is slow/sluggish for a couple minutes, after this time, it starts working as usual. I am looking for ways to troubleshoot this issue, in order to apply a potential fix, can someone chime in to help me logging what might cause it?
I was not sure where to post this thread, I do apologize if I opened it in the wrong section
Offline
You should first and foremost isolate the "sluggish" component, https://wiki.archlinux.org/title/Benchmarking
CPU, GPU, Disk IO? Network?
Also check top/iotop to ee whether there's maybe an explanation.
Obvious contender would be https://wiki.archlinux.org/title/CPU_frequency_scaling
Offline
thanks for the support, I ran ubench and got the following
BYTE UNIX Benchmarks (Version 5.1.3)
System: : GNU/Linux
OS: GNU/Linux -- 6.13.8-arch1-1 -- #1 SMP PREEMPT_DYNAMIC Sun, 23 Mar 2025 17:17:30 +0000
Machine: x86_64 (unknown)
Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
CPU 0: 11th Gen Intel(R) Core(TM) i5-1145G7 @ 2.60GHz (5222.4 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 1: 11th Gen Intel(R) Core(TM) i5-1145G7 @ 2.60GHz (5222.4 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 2: 11th Gen Intel(R) Core(TM) i5-1145G7 @ 2.60GHz (5222.4 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 3: 11th Gen Intel(R) Core(TM) i5-1145G7 @ 2.60GHz (5222.4 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 4: 11th Gen Intel(R) Core(TM) i5-1145G7 @ 2.60GHz (5222.4 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 5: 11th Gen Intel(R) Core(TM) i5-1145G7 @ 2.60GHz (5222.4 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 6: 11th Gen Intel(R) Core(TM) i5-1145G7 @ 2.60GHz (5222.4 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 7: 11th Gen Intel(R) Core(TM) i5-1145G7 @ 2.60GHz (5222.4 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
15:04:28 up 7 days, 2 min, 1 user, load average: 1.26, 0.99, 0.92; runlevel
------------------------------------------------------------------------
Benchmark Run: Tue Apr 01 2025 15:04:28 - 15:32:29
8 CPUs in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 0.1 lps (10.0 s, 7 samples)
Double-Precision Whetstone 10275.8 MWIPS (9.9 s, 7 samples)
Execl Throughput 5416.1 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 1882761.3 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 533346.3 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 3813231.6 KBps (30.0 s, 2 samples)
Pipe Throughput 3082122.2 lps (10.0 s, 7 samples)
Pipe-based Context Switching 242637.5 lps (10.0 s, 7 samples)
Process Creation 5185.1 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 5188.2 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 4138.4 lpm (60.0 s, 2 samples)
System Call Overhead 2391764.0 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 0.1 0.0
Double-Precision Whetstone 55.0 10275.8 1868.3
Execl Throughput 43.0 5416.1 1259.6
File Copy 1024 bufsize 2000 maxblocks 3960.0 1882761.3 4754.4
File Copy 256 bufsize 500 maxblocks 1655.0 533346.3 3222.6
File Copy 4096 bufsize 8000 maxblocks 5800.0 3813231.6 6574.5
Pipe Throughput 12440.0 3082122.2 2477.6
Pipe-based Context Switching 4000.0 242637.5 606.6
Process Creation 126.0 5185.1 411.5
Shell Scripts (1 concurrent) 42.4 5188.2 1223.6
Shell Scripts (8 concurrent) 6.0 4138.4 6897.3
System Call Overhead 15000.0 2391764.0 1594.5
========
System Benchmarks Index Score 1282.1
------------------------------------------------------------------------
Benchmark Run: Tue Apr 01 2025 15:32:29 - 16:01:00
8 CPUs in system; running 8 parallel copies of tests
Dhrystone 2 using register variables 0.8 lps (10.0 s, 7 samples)
Double-Precision Whetstone 52661.4 MWIPS (12.6 s, 7 samples)
Execl Throughput 17187.6 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 4898189.1 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 1387661.9 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 13423996.1 KBps (30.0 s, 2 samples)
Pipe Throughput 9537508.0 lps (10.0 s, 7 samples)
Pipe-based Context Switching 1369209.8 lps (10.0 s, 7 samples)
Process Creation 44247.4 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 29738.5 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 4177.1 lpm (60.0 s, 2 samples)
System Call Overhead 9349770.7 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 0.8 0.0
Double-Precision Whetstone 55.0 52661.4 9574.8
Execl Throughput 43.0 17187.6 3997.1
File Copy 1024 bufsize 2000 maxblocks 3960.0 4898189.1 12369.2
File Copy 256 bufsize 500 maxblocks 1655.0 1387661.9 8384.7
File Copy 4096 bufsize 8000 maxblocks 5800.0 13423996.1 23144.8
Pipe Throughput 12440.0 9537508.0 7666.8
Pipe-based Context Switching 4000.0 1369209.8 3423.0
Process Creation 126.0 44247.4 3511.7
Shell Scripts (1 concurrent) 42.4 29738.5 7013.8
Shell Scripts (8 concurrent) 6.0 4177.1 6961.8
System Call Overhead 15000.0 9349770.7 6233.2
========
System Benchmarks Index Score 1539.0
I will rerun the benchamrk when the I feel the system slow, after waking up from sleep.
I saw that when it feels sluggy, it has got no hogging processes, but in system monitor I can see the cpu being loaded. I believe it has something to do with the cpu scheduler/power profile
Offline
it has got no hogging processes, but in system monitor I can see the cpu being loaded
Check "top" for what actually uses the CPU - cpu load after waking up sound udev-related.
Offline
here's a top dump when the problem appeared, it's randomic
top - 22:16:40 up 8 days, 7:14, 1 user, load average: 2.45, 0.90, 0.59
Tasks: 341 total, 3 running, 337 sleep, 1 d-sleep, 0 stopped, 0 zombie
top - 22:17:04 up 8 days, 7:14, 1 user, load average: 2.18, 0.97, 0.62
Tasks: 342 total, 2 running, 340 sleep, 0 d-sleep, 0 stopped, 0 zombie
%Cpu(s): 5.4 us, 3.4 sy, 0.0 ni, 90.3 id, 0.0 wa, 0.5 hi, 0.4 si, 0.0 st
MiB Mem : 15691.0 total, 994.5 free, 7503.2 used, 8662.2 buff/cache
MiB Swap: 4096.0 total, 4096.0 free, 0.0 used. 8187.8 avail Mem
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
2976777 edge33 20 0 23.1g 566408 170428 S 13.8 3.5 12:29.02 Isolated Web Co
130147 edge33 20 0 231616 6524 6140 S 12.5 0.0 20:57.90 ksgrd_network_h
2976628 edge33 20 0 11.8g 855256 449576 S 6.3 5.3 33:26.71 firefox
130074 edge33 20 0 2248804 326060 214648 S 4.3 2.0 19:16.01 plasma-systemmo
2988985 edge33 20 0 2698148 287180 175604 S 3.6 1.8 3:32.55 Isolated Web Co
921 edge33 20 0 5896456 568640 191456 S 3.3 3.5 23:52.09 plasmashell
3014528 edge33 20 0 2953600 573516 194792 S 3.3 3.6 1:35.67 Isolated Web Co
2978983 edge33 20 0 3013716 627496 188152 S 3.0 3.9 3:20.68 Isolated Web Co
144571 edge33 20 0 4912812 352416 192864 S 2.6 2.2 2:30.86 spotify
743 edge33 -2 0 2562712 272736 176888 S 2.3 1.7 93:04.83 kwin_wayland
992 edge33 20 0 258488 38056 31656 S 2.3 0.2 10:14.05 ksystemstats
2978947 edge33 20 0 2558336 210516 149284 S 2.3 1.3 0:03.91 Isolated Web Co
2976900 edge33 20 0 22.9g 359964 148048 S 1.3 2.2 1:12.98 WebExtensions
3019412 edge33 20 0 1053100 223788 206184 S 1.3 1.4 0:02.31 konsole
2976773 edge33 20 0 2913352 415204 180884 S 1.0 2.6 1:22.96 Isolated Web Co
3019882 edge33 20 0 10904 7844 5668 R 1.0 0.0 0:00.24 top
532 root 20 0 414768 24820 21364 S 0.7 0.2 1:55.98 NetworkManager
3014352 edge33 20 0 2658052 260136 167720 S 0.7 1.6 0:05.03 Isolated Web Co
19 root -2 0 0 0 0 I 0.3 0.0 3:24.43 rcu_preempt
82 root -51 0 0 0 0 S 0.3 0.0 5:27.63 irq/9-acpi
Offline
firefox - and google throws a lot of "temporary cpu hogging" at me when I ask for "ksgrd_network_helper", seems to be part of ksysguard.
Edit: maybe also part of plasma-systemonitor?
Another thing is that you're getting short-ish on RAM, maybe zswap/zram related?
Last edited by seth (2025-04-02 20:29:36)
Offline
firefox - and google throws a lot of "temporary cpu hogging" at me when I ask for "ksgrd_network_helper", seems to be part of ksysguard.
Edit: maybe also part of plasma-systemonitor?Another thing is that you're getting short-ish on RAM, maybe zswap/zram related?
not sure about cpu hogging, seems pretty normal to me,
about zswap, I don't know, how would I troubleshoot? I see In the screnshot 900 Mb free but half is buff/cache, I have 4096 MB of swap partition
Offline
Disable it (it's enabled by default, but I think the archinstall script disables it and uses zram instead, so have a look at your situation)
https://wiki.archlinux.org/title/Zswap#Toggling_zswap
zswap can explode when decompressed onto the physical swap (though it would indeed be odd if that ran you OOM and killing the offending process would wipe the *entire* swap)
The "ksgrd_network_helper" cpu hogging is reported to be temporarily, so it might be in decline when you ran top, but 12% cpu load from a process that reads the network traffic stats is rather not "normal" tbw (unless your computer is potato)
Do you have the benchmark results from when the system is "sluggish" so we can look for the bottleneck?
Offline
here's the result, I started benchamrking when system was sluggy:
BYTE UNIX Benchmarks (Version 5.1.3)
System: : GNU/Linux
OS: GNU/Linux -- 6.13.8-arch1-1 -- #1 SMP PREEMPT_DYNAMIC Sun, 23 Mar 2025 17:17:30 +0000
Machine: x86_64 (unknown)
Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
CPU 0: 11th Gen Intel(R) Core(TM) i5-1145G7 @ 2.60GHz (5222.4 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 1: 11th Gen Intel(R) Core(TM) i5-1145G7 @ 2.60GHz (5222.4 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 2: 11th Gen Intel(R) Core(TM) i5-1145G7 @ 2.60GHz (5222.4 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 3: 11th Gen Intel(R) Core(TM) i5-1145G7 @ 2.60GHz (5222.4 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 4: 11th Gen Intel(R) Core(TM) i5-1145G7 @ 2.60GHz (5222.4 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 5: 11th Gen Intel(R) Core(TM) i5-1145G7 @ 2.60GHz (5222.4 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 6: 11th Gen Intel(R) Core(TM) i5-1145G7 @ 2.60GHz (5222.4 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 7: 11th Gen Intel(R) Core(TM) i5-1145G7 @ 2.60GHz (5222.4 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
11:14:39 up 8 days, 20:12, 1 user, load average: 3.34, 1.90, 1.23; runlevel
------------------------------------------------------------------------
Benchmark Run: Thu Apr 03 2025 11:14:39 - 11:50:55
8 CPUs in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 0.1 lps (10.0 s, 7 samples)
Double-Precision Whetstone 10037.6 MWIPS (10.1 s, 7 samples)
Execl Throughput 5334.0 lps (29.7 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 1880822.4 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 531483.5 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 3692367.1 KBps (30.0 s, 2 samples)
Pipe Throughput 3093669.6 lps (10.0 s, 7 samples)
Pipe-based Context Switching 240114.5 lps (10.0 s, 7 samples)
Process Creation 5593.7 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 1856.9 lpm (305.9 s, 2 samples)
Shell Scripts (8 concurrent) 4457.2 lpm (60.0 s, 2 samples)
System Call Overhead 2386692.5 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 0.1 0.0
Double-Precision Whetstone 55.0 10037.6 1825.0
Execl Throughput 43.0 5334.0 1240.5
File Copy 1024 bufsize 2000 maxblocks 3960.0 1880822.4 4749.6
File Copy 256 bufsize 500 maxblocks 1655.0 531483.5 3211.4
File Copy 4096 bufsize 8000 maxblocks 5800.0 3692367.1 6366.2
Pipe Throughput 12440.0 3093669.6 2486.9
Pipe-based Context Switching 4000.0 240114.5 600.3
Process Creation 126.0 5593.7 443.9
Shell Scripts (1 concurrent) 42.4 1856.9 438.0
Shell Scripts (8 concurrent) 6.0 4457.2 7428.6
System Call Overhead 15000.0 2386692.5 1591.1
========
System Benchmarks Index Score 1183.4
------------------------------------------------------------------------
Benchmark Run: Thu Apr 03 2025 11:50:55 - 12:32:22
8 CPUs in system; running 8 parallel copies of tests
Dhrystone 2 using register variables 0.8 lps (10.0 s, 7 samples)
Double-Precision Whetstone 52501.1 MWIPS (12.6 s, 7 samples)
Execl Throughput 16924.5 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 4925928.2 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 1377608.1 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 5316261.1 KBps (30.0 s, 2 samples)
Pipe Throughput 10403652.1 lps (10.0 s, 7 samples)
Pipe-based Context Switching 1342376.6 lps (10.0 s, 7 samples)
Process Creation 41667.0 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 28540.6 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 4057.2 lpm (60.0 s, 2 samples)
System Call Overhead 9002509.4 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 0.8 0.0
Double-Precision Whetstone 55.0 52501.1 9545.7
Execl Throughput 43.0 16924.5 3935.9
File Copy 1024 bufsize 2000 maxblocks 3960.0 4925928.2 12439.2
File Copy 256 bufsize 500 maxblocks 1655.0 1377608.1 8323.9
File Copy 4096 bufsize 8000 maxblocks 5800.0 5316261.1 9166.0
Pipe Throughput 12440.0 10403652.1 8363.1
Pipe-based Context Switching 4000.0 1342376.6 3355.9
Process Creation 126.0 41667.0 3306.9
Shell Scripts (1 concurrent) 42.4 28540.6 6731.3
Shell Scripts (8 concurrent) 6.0 4057.2 6761.9
System Call Overhead 15000.0 9002509.4 6001.7
========
System Benchmarks Index Score
Offline
"System Benchmarks Index Score" ?
However:
Shell Scripts (1 concurrent) 42.4 1856.9 438.0
File Copy 4096 bufsize 8000 maxblocks 13423996.1 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 5800.0 13423996.1 23144.8
Shell Scripts (1 concurrent) 42.4 5188.2 1223.6
File Copy 4096 bufsize 8000 maxblocks 5316261.1 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 5800.0 5316261.1 9166.0
The single-shell script run between 15:00 and 15:30 and the 4k file copying between 15:30 and 16:00 are the only drop-offs in that benchmark
I started benchamrking when system was sluggy
And was it all throughout (or just the first couple of minutes)?
Because w/ the massive timeframe that benchmark spans, it would become a rather useless tool for this
Did you try to disable the network indicators?
Offline
"System Benchmarks Index Score" ?
However:
good wrote:Shell Scripts (1 concurrent) 42.4 1856.9 438.0
File Copy 4096 bufsize 8000 maxblocks 13423996.1 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 5800.0 13423996.1 23144.8bad wrote:Shell Scripts (1 concurrent) 42.4 5188.2 1223.6
File Copy 4096 bufsize 8000 maxblocks 5316261.1 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 5800.0 5316261.1 9166.0The single-shell script run between 15:00 and 15:30 and the 4k file copying between 15:30 and 16:00 are the only drop-offs in that benchmark
I started benchamrking when system was sluggy
And was it all throughout (or just the first couple of minutes)?
Because w/ the massive timeframe that benchmark spans, it would become a rather useless tool for thisDid you try to disable the network indicators?
it was slow only for the first couple minutes, so yeah, I was thinking maybe is not the right tool.
how would I disable network indicators? would putting airplane mode work from the network manager kde panel?
Offline
The helper is likely invoked from ksysguard or some plasma widget-thingy/network indicator.
Your goal is to make the "ksgrd_network_helper" process disappear.
Please avoid bloating the thread w/ full quotes, the previous posts are readily available right above the last one.
Offline
Pages: 1