Ubuntu
linux package

kswapd takes 100% CPU although there is no swap

Bug #1476211 reported by Andreas E. on 2015-07-20

This bug report is a duplicate of: Bug #1518457: kswapd0 100% CPU usage. Edit Remove

This bug affects 10 people

Affects		Status	Importance	Assigned to	Milestone
	linux (Ubuntu)	Confirmed	Medium	Unassigned

Bug Description

On a notebook with neither swap partition nor swap file (because of SSD and surplus of RAM), kswapd0 suddenly runs amok and takes 100% of CPU activity. When this happens, usually at most 5.5 of 8 GB RAM were in use, so there is no reason to swap.

This happens so suddenly, that no user input is possible anymore, eg. I can barely switch to tty to kill processes that take most RAM. The processor becomes extremely hot and I can only turn the computer off and reboot. I consider this as critical and harmful to my hardware.

I noticed that Ubuntu installs by default with swappiness 60. Now I changed that to 10 because I would never encounter the case that 8GB are in use, but it didn't have an effect.

Expected behavior:
- on systems without RAM, kswapd should never attempt to swap
- if there is not enough RAM, kernel should kill processes to free up RAM
- no process should be able to claim so much CPU activity, that the computer becomes unresponsive (it should always be possible to control/stop/kill processes over tty)

Actual behavior:
- kswapd tries to swap although there is no swap space

ProblemType: Bug
DistroRelease: Ubuntu 15.04
Package: linux-image-3.19.0-22-generic 3.19.0-22.22
ProcVersionSignature: Ubuntu 3.19.0-22.22-generic 3.19.8-ckt1
Uname: Linux 3.19.0-22-generic x86_64
ApportVersion: 2.17.2-0ubuntu1.1
Architecture: amd64
AudioDevicesInUse:
USER PID ACCESS COMMAND
/dev/snd/controlC1: andreas 2368 F.... pulseaudio
/dev/snd/controlC0: andreas 2368 F.... pulseaudio
CurrentDesktop: Unity
Date: Mon Jul 20 12:49:24 2015
EcryptfsInUse: Yes
InstallationDate: Installed on 2015-06-06 (43 days ago)
InstallationMedia: Ubuntu 15.04 "Vivid Vervet" - Release amd64 (20150422)
MachineType: LENOVO 20266
ProcFB: 0 inteldrmfb
ProcKernelCmdLine: BOOT_IMAGE=/boot/vmlinuz-3.19.0-22-generic root=UUID=8a0dc2f3-fcf3-4e5b-8e23-1441bd36e3dc ro noprompt persistent quiet splash acpi_backlight=vendor vt.handoff=7 init=/sbin/upstart
RelatedPackageVersions:
linux-restricted-modules-3.19.0-22-generic N/A
linux-backports-modules-3.19.0-22-generic N/A
linux-firmware 1.143.1
SourcePackage: linux
UpgradeStatus: No upgrade log present (probably fresh install)
dmi.bios.date: 10/31/2013
dmi.bios.vendor: LENOVO
dmi.bios.version: 76CN31WW
dmi.board.asset.tag: No Asset Tag
dmi.board.name: Yoga2
dmi.board.vendor: LENOVO
dmi.board.version: 31900058STD
dmi.chassis.asset.tag: No Asset Tag
dmi.chassis.type: 10
dmi.chassis.vendor: LENOVO
dmi.chassis.version: Lenovo Yoga 2 Pro
dmi.modalias: dmi:bvnLENOVO:bvr76CN31WW:bd10/31/2013:svnLENOVO:pn20266:pvrLenovoYoga2Pro:rvnLENOVO:rnYoga2:rvr31900058STD:cvnLENOVO:ct10:cvrLenovoYoga2Pro:
dmi.product.name: 20266
dmi.product.version: Lenovo Yoga 2 Pro
dmi.sys.vendor: LENOVO

See original description

Tags:

Revision history for this message

Andreas E. (andreas-e) wrote on 2015-07-20:

AlsaInfo.txt Edit (39.4 KiB, text/plain; charset="utf-8")
CRDA.txt Edit (262 bytes, text/plain; charset="utf-8")
CurrentDmesg.txt Edit (61.8 KiB, text/plain; charset="utf-8")
Dependencies.txt Edit (2.3 KiB, text/plain; charset="utf-8")
IwConfig.txt Edit (471 bytes, text/plain; charset="utf-8")
Lspci.txt Edit (7.5 KiB, text/plain; charset="utf-8")
Lsusb.txt Edit (438 bytes, text/plain; charset="utf-8")
ProcCpuinfo.txt Edit (3.9 KiB, text/plain; charset="utf-8")
ProcEnviron.txt Edit (103 bytes, text/plain; charset="utf-8")
ProcInterrupts.txt Edit (2.2 KiB, text/plain; charset="utf-8")
ProcModules.txt Edit (6.7 KiB, text/plain; charset="utf-8")
PulseList.txt Edit (23.2 KiB, text/plain; charset="utf-8")
RfKill.txt Edit (248 bytes, text/plain; charset="utf-8")
UdevDb.txt Edit (164.6 KiB, text/plain; charset="utf-8")
UdevLog.txt Edit (368.2 KiB, text/plain; charset="utf-8")
WifiSyslog.txt Edit (768.8 KiB, text/plain; charset="utf-8")

description:

updated

Revision history for this message

Brad Figg (brad-figg) wrote on 2015-07-20: Status changed to Confirmed

This change was made by a bot.

Changed in linux (Ubuntu):
status:	New → Confirmed

Revision history for this message

Joseph Salisbury (jsalisbury) wrote on 2015-07-20:

Would it be possible for you to test the latest upstream kernel? Refer to https://wiki.ubuntu.com/KernelMainlineBuilds . Please test the latest v4.2 kernel[0].

If this bug is fixed in the mainline kernel, please add the following tag 'kernel-fixed-upstream'.

If the mainline kernel does not fix this bug, please add the tag: 'kernel-bug-exists-upstream'.

Once testing of the upstream kernel is complete, please mark this bug as "Confirmed".

Thanks in advance.

[0] http://kernel.ubuntu.com/~kernel-ppa/mainline/v4.2-rc3-unstable/

Changed in linux (Ubuntu):
importance:	Undecided → Medium
status:	Confirmed → Incomplete

Revision history for this message

Andreas E. (andreas-e) wrote on 2015-07-25:

I put 4.2.0-040200rc3-generic under RAM stress. I still can not get to over 5GB (of 8GB) before kswapd turns the system unusable.

tags:	added: kernel-bug-exists-upstream
Changed in linux (Ubuntu):
status:	Incomplete → Confirmed

penalvch (penalvch) on 2015-07-28

tags:	added: bios-outdated-76cn43ww
Changed in linux (Ubuntu):
status:	Confirmed → Incomplete

Revision history for this message

Andreas E. (andreas-e) wrote on 2015-07-29:

[[Ubuntu does not provide an update path for (these) Lenovo BIOS. Maybe it would be nice to have a solution for this as well, to avoid users being bound to buggy and outdated BIOS. To add, the given instructions to the wiki seem very abandonned and do not (not all) apply to nowadays ROMs of >10MB. Figuring that out cost me a day. ]]

After installing the Win8/BIOS, the problem still occurs.
  sudo dmidecode -s bios-version && sudo dmidecode -s bios-release-date
  76CN43WW
  03/30/2015

When I intentionally launch so many applications to trigger the freezes, I noticed that `top` displays actually different numbers:
KiB Mem: 8085948 total, 8023384 used, 62564 free, 3564 buffers
KiB Swap: 0 total, 0 used, 0 free. 2976336 cached Mem
The before mentioned numbers were from Gnome system monitor (oh!), which always displays only 5GB / 65% of RAM. This time had also top opened before the freeze.
Do I interprete it right that under these conditions the memory is indeed full and that system monitor displays (used‒cached) ?

The laptop was still frozen about 10min until the Kernel decided (or was able to) kill processes and became repsonsive again. On a second test, it didn't become responsive even after 20min.

So concluding,
• there could be a bug in system monitor to display (used ‒ cached) RAM
• I still don't know what it is doing within these 10 – ∞ minutes. I still think the Kernel should be able to start killing processes (earlier), or there should be warnings that can take effect before the computer becomes unresponsive.

Changed in linux (Ubuntu):
status:	Incomplete → Confirmed

penalvch (penalvch) on 2015-08-01

tags:

added: latest-bios-76cn43ww
removed: bios-outdated-76cn43ww

Revision history for this message

Sam Soldatenko (sam-ngs-ru) wrote on 2015-10-04:

Download full text (10.6 KiB)

I have similar behaviour on my Ubuntu 15.04, Intel Atom D2500 Dual LAN, Dual COM Fanless Mini-ITX PC with 4 GB RAM and 128 GB SSD.

In few days after reboot kswapd0 starts to take 100% CPU (see highlighted row here https://gist.github.com/SamSoldatenko/83eaf196721bc51d7d55#file-gistfile1-txt-L29 ). No load. 3Gb free memory (out of 4), 10Gb swap space.

$ uname -a
Linux minisrv1 3.19.0-26-generic #28-Ubuntu SMP Tue Aug 11 14:16:45 UTC 2015 i686 i686 i686 GNU/Linux

$ sudo dmidecode -s bios-version && sudo dmidecode -s bios-release-date
CCCDT10N.86A.0032.2012.0323.1510
03/23/2012

$ uptime
22:39:01 up 8 days, 5:43, 1 user, load average: 1.34, 1.23, 1.23

$ sudo ps auxf
USER PID %CPU %MEM root 2 0.0 0.0 root 3 0.0 0.0 root 5 0.0 0.0 root 7 0.6 0.0 root 8 0.0 0.0 root 9 0.5 0.0 root 10 0.2 0.0 root 11 0.2 0.0 root 12 0.6 0.0 root 13 0.0 0.0 root 15 0.0 0.0 root 16 0.0 0.0 root 17 0.0 0.0 root 18 0.0 0.0 root 19 0.0 0.0 root 20 0.0 0.0 root 21 0.0 0.0 root 22 0.0 0.0 root 23 0.0 0.0 root 24 0.0 0.0 root 25 0.0 0.0 root 26 0.0 0.0 root 27 0.0 0.0 root 28 0.0 0.0 root 29 0.0 0.0 root 30 0.0 0.0 root 34 31.5 0.0 root 35 0.0 0.0 root 36 0.0 0.0 root 48 0.0 0.0 root 49 0.0 0.0 ... VSZ RSS TTY STAT START TIME COMMAND
0 0 ? S Sep26 0:15 [kthreadd]
0 0 ? S Sep26 4:39 \_ [ksoftirqd/0]
0 0 ? S< Sep26 0:00 \_ [kworker/0:0H]
0 0 ? S Sep26 76:29 \_ [rcu_sched]
0 0 ? S Sep26 0:00 \_ [rcu_bh]
0 0 ? S Sep26 64:30 \_ [migration/0]
0 0 ? S Sep26 23:57 \_ [watchdog/0]
0 0 ? S Sep26 26:02 \_ [watchdog/1]
0 0 ? S Sep26 75:46 \_ [migration/1]
0 0 ? S Sep26 4:38 \_ [ksoftirqd/1]
0 0 ? S< Sep26 0:00 \_ [kworker/1:0H]
0 0 ? S< Sep26 0:00 \_ [khelper]
0 0 ? S Sep26 0:00 \_ [kdevtmpfs]
0 0 ? S< Sep26 0:00 \_ [netns]
0 0 ? S< Sep26 0:00 \_ [perf]
0 0 ? S Sep26 0:05 \_ [khungtaskd]
0 0 ? S< Sep26 0:01 \_ [writeback]
0 0 ? SN Sep26 0:00 \_ [ksmd]
0 0 ? SN Sep26 0:40 \_ [khugepaged]
0 0 ? S< Sep26 0:00 \_ [crypto]
0 0 ? S< Sep26 0:00 \_ [kintegrityd]
0 0 ? S< Sep26 0:00 \_ [bioset]
0 0 ? S< Sep26 0:00 \_ [kblockd]
0 0 ? S< Sep26 0:00 \_ [ata_sff]
0 0 ? S< Sep26 0:00 \_ [md]
0 0 ? S< Sep26 0:00 \_ [devfreq_wq]
0 0 ? R Sep26 3748:10 \_ [kswapd0]
0 0 ? S Sep26 0:00 \_ [fsnotify_mark]
0 0 ? S Sep26 0:00 \_ [ecryptfs-kthrea]
0 0 ? S< Sep26 0:00 \_ [kthrotld]

I have similar behaviour on my Ubuntu 15.04, Intel Atom D2500 Dual LAN, Dual COM Fanless Mini-ITX PC with 4 GB RAM and 128 GB SSD.

In few days after reboot kswapd0 starts to take 100% CPU  (see highlighted row here https://gist.github.com/SamSoldatenko/83eaf196721bc51d7d55#file-gistfile1-txt-L29 ). No load. 3Gb free memory (out of 4), 10Gb swap space.

$ uname -a
Linux minisrv1 3.19.0-26-generic #28-Ubuntu SMP Tue Aug 11 14:16:45 UTC 2015 i686 i686 i686 GNU/Linux

$ sudo dmidecode -s bios-version && sudo dmidecode -s bios-release-date
CCCDT10N.86A.0032.2012.0323.1510
03/23/2012

$ uptime
 22:39:01 up 8 days,  5:43,  1 user,  load average: 1.34, 1.23, 1.23

$ sudo ps auxf
USER       PID %CPU %MEM    VSZ   RSS TTY      STAT START   TIME COMMAND
root         2  0.0  0.0      0     0 ?        S    Sep26   0:15 [kthreadd]
root         3  0.0  0.0      0     0 ?        S    Sep26   4:39  \_ [ksoftirqd/0]
root         5  0.0  0.0      0     0 ?        S<   Sep26   0:00  \_ [kworker/0:0H]
root         7  0.6  0.0      0     0 ?        S    Sep26  76:29  \_ [rcu_sched]
root         8  0.0  0.0      0     0 ?        S    Sep26   0:00  \_ [rcu_bh]
root         9  0.5  0.0      0     0 ?        S    Sep26  64:30  \_ [migration/0]
root        10  0.2  0.0      0     0 ?        S    Sep26  23:57  \_ [watchdog/0]
root        11  0.2  0.0      0     0 ?        S    Sep26  26:02  \_ [watchdog/1]
root        12  0.6  0.0      0     0 ?        S    Sep26  75:46  \_ [migration/1]
root        13  0.0  0.0      0     0 ?        S    Sep26   4:38  \_ [ksoftirqd/1]
root        15  0.0  0.0      0     0 ?        S<   Sep26   0:00  \_ [kworker/1:0H]
root        16  0.0  0.0      0     0 ?        S<   Sep26   0:00  \_ [khelper]
root        17  0.0  0.0      0     0 ?        S    Sep26   0:00  \_ [kdevtmpfs]
root        18  0.0  0.0      0     0 ?        S<   Sep26   0:00  \_ [netns]
root        19  0.0  0.0      0     0 ?        S<   Sep26   0:00  \_ [perf]
root        20  0.0  0.0      0     0 ?        S    Sep26   0:05  \_ [khungtaskd]
root        21  0.0  0.0      0     0 ?        S<   Sep26   0:01  \_ [writeback]
root        22  0.0  0.0      0     0 ?        SN   Sep26   0:00  \_ [ksmd]
root        23  0.0  0.0      0     0 ?        SN   Sep26   0:40  \_ [khugepaged]
root        24  0.0  0.0      0     0 ?        S<   Sep26   0:00  \_ [crypto]
root        25  0.0  0.0      0     0 ?        S<   Sep26   0:00  \_ [kintegrityd]
root        26  0.0  0.0      0     0 ?        S<   Sep26   0:00  \_ [bioset]
root        27  0.0  0.0      0     0 ?        S<   Sep26   0:00  \_ [kblockd]
root        28  0.0  0.0      0     0 ?        S<   Sep26   0:00  \_ [ata_sff]
root        29  0.0  0.0      0     0 ?        S<   Sep26   0:00  \_ [md]
root        30  0.0  0.0      0     0 ?        S<   Sep26   0:00  \_ [devfreq_wq]
root        34 31.5  0.0      0     0 ?        R    Sep26 3748:10  \_ [kswapd0]
root        35  0.0  0.0      0     0 ?        S    Sep26   0:00  \_ [fsnotify_mark]
root        36  0.0  0.0      0     0 ?        S    Sep26   0:00  \_ [ecryptfs-kthrea]
root        48  0.0  0.0      0     0 ?        S<   Sep26   0:00  \_ [kthrotld]
root        49  0.0  0.0      0     0 ?        S<   Sep26   0:00  \_ [acpi_thermal_pm]
root        51  0.0  0.0      0     0 ?        S<   Sep26   0:00  \_ [ipv6_addrconf]
root        70  0.0  0.0      0     0 ?        S<   Sep26   0:00  \_ [deferwq]
root        71  0.0  0.0      0     0 ?        S<   Sep26   0:00  \_ [charger_manager]
root       118  0.0  0.0      0     0 ?        S<   Sep26   0:00  \_ [kpsmoused]
root       119  0.0  0.0      0     0 ?        S    Sep26   0:00  \_ [scsi_eh_0]
root       120  0.0  0.0      0     0 ?        S<   Sep26   0:00  \_ [scsi_tmf_0]
root       121  0.0  0.0      0     0 ?        S    Sep26   0:00  \_ [scsi_eh_1]
root       122  0.0  0.0      0     0 ?        S<   Sep26   0:00  \_ [scsi_tmf_1]
root       123  0.0  0.0      0     0 ?        S    Sep26   0:00  \_ [scsi_eh_2]
root       124  0.0  0.0      0     0 ?        S<   Sep26   0:00  \_ [scsi_tmf_2]
root       125  0.0  0.0      0     0 ?        S    Sep26   0:00  \_ [scsi_eh_3]
root       126  0.0  0.0      0     0 ?        S<   Sep26   0:00  \_ [scsi_tmf_3]
root       138  0.0  0.0      0     0 ?        S    Sep26   0:57  \_ [jbd2/sda1-8]
root       139  0.0  0.0      0     0 ?        S<   Sep26   0:00  \_ [ext4-rsv-conver]
root       181  0.0  0.0      0     0 ?        S    Sep26   0:00  \_ [kauditd]
root       297  0.0  0.0      0     0 ?        S<   Sep26   0:00  \_ [hd-audio0]
root       310  0.0  0.0      0     0 ?        S<   Sep26   0:50  \_ [kworker/1:1H]
root       311  0.0  0.0      0     0 ?        S<   Sep26   0:56  \_ [kworker/0:1H]
root       602  0.0  0.0      0     0 ?        S<   Sep26   0:00  \_ [cfg80211]
root      5464  0.1  0.0      0     0 ?        S    15:14   0:51  \_ [kworker/1:2]
root      6065  0.3  0.0      0     0 ?        S    17:12   1:13  \_ [kworker/0:1]
root      7237  0.4  0.0      0     0 ?        S    20:54   0:29  \_ [kworker/1:0]
root      7648  0.0  0.0      0     0 ?        S    22:15   0:00  \_ [kworker/0:0]
root      7732  1.0  0.0      0     0 ?        S    22:15   0:15  \_ [kworker/u8:0]
root      7839  0.4  0.0      0     0 ?        S    22:31   0:02  \_ [kworker/u8:2]
root         1  0.1  0.4  20180 18464 ?        Ss   Sep26  12:28 /sbin/init text
root       234  0.0  0.0   9416  3428 ?        Ss   Sep26   4:27 /lib/systemd/systemd-journald
root       237  0.0  0.0  13224  3428 ?        Ss   Sep26   0:00 /lib/systemd/systemd-udevd
systemd+   378  0.0  0.0  12424  2200 ?        Ssl  Sep26   0:22 /lib/systemd/systemd-timesyncd
whoopsie   547  0.1  0.2  73256 11208 ?        Ssl  Sep26  14:21 /usr/bin/whoopsie -f
root       570  0.0  0.0   5572  2768 ?        Ss   Sep26   0:59 /usr/sbin/cron -f
root       573  0.0  0.2  41140  8560 ?        Ssl  Sep26   1:20 /usr/sbin/ModemManager
syslog     574  0.0  0.5  50220 22236 ?        Ssl  Sep26   1:03 /usr/sbin/rsyslogd -n
avahi      575  0.0  0.0   3736  2816 ?        Ss   Sep26   0:53 avahi-daemon: running [minisrv1-3.local]
avahi      583  0.0  0.0   3616   212 ?        S    Sep26   0:00  \_ avahi-daemon: chroot helper
root       585  1.1  0.3  82884 13480 ?        Ssl  Sep26 141:59 /usr/sbin/NetworkManager --no-daemon
root       684  0.0  0.1   9300  4564 ?        S    Sep26   0:00  \_ /sbin/dhclient -d -q -sf /usr/lib/NetworkManager/nm-dhcp-helper -pf /run/sendsigs.omit.d/network-manager.dhclient-eth0.pid -lf /var/lib/NetworkManager/dhclient-eba3cda0-1d2d-4db7-a1dd-a90fde49ac32-eth0.lease -cf /var/lib/NetworkManager/dhclient-eth0.conf eth0
nobody     687  0.0  0.0   6724  3104 ?        S    Sep26   3:53  \_ /usr/sbin/dnsmasq --no-resolv --keep-in-foreground --no-hosts --bind-interfaces --pid-file=/run/sendsigs.omit.d/network-manager.dnsmasq.pid --listen-address=127.0.1.1 --conf-file=/var/run/NetworkManager/dnsmasq.conf --cache-size=0 --proxy-dnssec --enable-dbus=org.freedesktop.NetworkManager.dnsmasq --conf-dir=/etc/NetworkManager/dnsmasq.d
message+   597  0.1  0.0   5836  3616 ?        Ss   Sep26  13:29 /usr/bin/dbus-daemon --system --address=systemd: --nofork --nopidfile --systemd-activation
postgres   614  0.0  0.3 193456 14268 ?        S    Sep26   1:15 /usr/lib/postgresql/9.4/bin/postgres -D /var/lib/postgresql/9.4/main -c config_file=/etc/postgresql/9.4/main/postgresql.conf
postgres   645  0.0  0.0 193456  3184 ?        Ss   Sep26   0:01  \_ postgres: checkpointer process                                                  
postgres   646  0.0  0.0 193456  2336 ?        Ss   Sep26   2:02  \_ postgres: writer process                                                        
postgres   647  0.0  0.0 193456  3260 ?        Ss   Sep26   2:03  \_ postgres: wal writer process                                                    
postgres   648  0.0  0.1 193756  5480 ?        Ss   Sep26   1:18  \_ postgres: autovacuum launcher process                                           
postgres   649  0.0  0.0  50968  3576 ?        Ss   Sep26   0:24  \_ postgres: stats collector process                                               
root       621  0.0  0.2  39420 11204 ?        Ssl  Sep26   2:12 /usr/lib/policykit-1/polkitd --no-debug
root       622  0.0  0.1  21832  6580 ?        Ssl  Sep26   1:37 /usr/sbin/cups-browsed
root       661  0.0  0.1   8580  4880 ?        Ss   Sep26   4:43 /usr/sbin/sshd -D
root      9944  0.0  0.1  11872  5640 ?        Ss   Oct02   0:00  \_ sshd: git [priv]
root      7641  0.0  0.1  11872  5860 ?        Ss   22:15   0:00  \_ sshd: sam [priv]
sam       7716  0.2  0.0  12004  3204 ?        S    22:15   0:03      \_ sshd: sam@pts/0
sam       7717  0.1  0.1   7572  5088 pts/0    Ss   22:15   0:02          \_ -bash
root      7882  0.0  0.0   6800  3648 pts/0    S+   22:40   0:00              \_ sudo ps auxf
root      7883  0.0  0.0   5316  2388 pts/0    R+   22:40   0:00                  \_ ps auxf
root       670  0.0  0.0   4740  1928 tty1     Ss+  Sep26   0:00 /sbin/agetty --noclear tty1 linux
kernoops   879  0.0  0.0   7116  2616 ?        Ss   Sep26   0:23 /usr/sbin/kerneloops
root      1704  0.0  0.0  42560   992 ?        Ss   Sep26   0:00 nginx: master process /usr/sbin/nginx -g daemon on; master_process on;
www-data  1705  0.1  0.0  42692  2912 ?        S    Sep26  17:17  \_ nginx: worker process
www-data  1706  0.1  0.0  42692  2912 ?        S    Sep26  17:31  \_ nginx: worker process
www-data  1707  0.0  0.0  42692  2912 ?        S    Sep26   0:00  \_ nginx: worker process
www-data  1708  0.1  0.0  42692  2912 ?        S    Sep26  17:41  \_ nginx: worker process
root      6137  0.0  0.0   4400  2044 ?        Ss   Sep26   3:52 /usr/sbin/irqbalance --pid=/var/run/irqbalance.pid
root      6487  0.0  0.1  46964  6536 ?        Ssl  Sep26   4:35 /usr/sbin/thermald --no-daemon --dbus-enable
git       7916  0.0  0.0   5684  3912 ?        Ss   Sep26   0:00 /lib/systemd/systemd --user
git       7917  0.0  0.0   7012   852 ?        S    Sep26   0:00  \_ (sd-pam)
root      5378  0.0  0.1  12496  5108 ?        Ss   Sep27   0:00 /usr/sbin/cupsd -l
root      8072  0.0  0.1   8468  6796 ?        Ss   Oct02   0:34 /lib/systemd/systemd-logind
sam       7645  0.0  0.1   5684  4120 ?        Ss   22:15   0:00 /lib/systemd/systemd --user
sam       7646  0.0  0.3  21148 15116 ?        S    22:15   0:00  \_ (sd-pam)

$ free -m
             total       used       free     shared    buffers     cached
Mem:          4020       1027       2993         54          2        124
-/+ buffers/cache:        900       3120
Swap:        10220         10      10210

$ sudo swapon -s
Filename                                Type            Size    Used    Priority
/dev/sda5                               partition       2077692 10952   -1
/swapfile                               file            8388604 0       -2

Revision history for this message

Andreas E. (andreas-e) wrote on 2015-12-04:

This is still the case in Ubuntu 15.10.
It would really be cool if this could be fixed, because it causes data loss and regularly interrupts my work. I cannot imagine this same software runs on servers, IoT and systems that need reliability.

Revision history for this message

Ski (kacoroski) wrote on 2015-12-10:

I am on xbuntu 15.10 and this affected me after I upgraded. I have 16GB ram and kswapd kicks in, locks it up for 10 - 30 seconds multiple times an hour. vmstat 5 shows:

1 0 1208112 569056 69376 7019032 0 0 84 347 1634 2951 25 1 73 0 0
2 0 1208072 164192 2212 6619100 14 6 163 10 2177 5029 26 3 70 0 0
0 20 2366584 204456 316 6367792 48 234926 25103 235206 6928 2748 2 8 5 86 0
1 5 3294060 151216 548 6413108 27001 193908 88306 193970 5196 6590 2 10 13 76 0
2 1 4065264 172748 3172 6488812 72246 163202 106635 163531 8388 11018 2 22 43 33 0
5 0 4806852 136160 4444 6469352 76391 153371 98650 154074 8352 9759 2 26 37 35 0
1 1 5573372 145112 816 6440936 73558 157316 87340 157662 6802 8592 2 21 44 33 0
0 4 6383744 168976 1800 6463780 74014 166143 88715 166679 6388 8669 2 27 41 31 0
1 1 7024704 146832 1924 6466112 68409 135974 78778 136382 6356 8203 1 19 42 38 0
2 0 7736084 181116 1928 6476672 56372 144775 64527 145122 8093 6853 1 16 41 43 0
0 7 8381188 197892 432 6456988 74326 159450 87829 160040 8327 8744 4 23 45 28 0
1 0 3944136 3240020 1400 6488888 9170 67 14278 74 2112 4084 22 3 69 6 0
1 0 3926636 2819400 2060 6521132 3596 0 7028 0 1848 3374 25 3 71 1 0

So you can see when it kicks in. I have made the following modifications and still have the problem:

/etc/sysctl.conf:
vm.swappiness=10

/etc/rc.local:
# Fix khugepaged issues (20150706)
echo never > /sys/kernel/mm/transparent_hugepage/defrag
echo 0 > /sys/kernel/mm/transparent_hugepage/khugepaged/defrag

# Another fix for khugepaged issues (20151210)
echo madvise > /sys/kernel/mm/transparent_hugepage/enabled

Revision history for this message

mm (mtl-0) wrote on 2015-12-12:

#10

After upgrading to Xubuntu 15.10 on 2 home server boxes I am experiencing a similar problem: kswapd0 uses 99,9% of CPU. The only way to make it stop is to issue the swapoff command. When I try to issue swapon again it automatically produces 99,9% of CPU load again. (Even if the swap isn’t getting filled)

Any clue about that? It’s really annoying.

Revision history for this message

mm (mtl-0) wrote on 2015-12-12:

#11

Update: Even with no swap (partition/file) attached the kswapd0 is generating 100% cpu load.

Report a bug

This report contains Public information

Everyone can see this information.

Duplicate of bug #1518457 Remove

You are

Subscribing...

Edit bug mail

Other bug subscribers

Bug attachments

Add attachment

Remote bug watches

Bug watches keep track of this bug in other bug trackers.

Ubuntulinux package

kswapd takes 100% CPU although there is no swap

Bug Description

Other bug subscribers

Bug attachments

Remote bug watches

Ubuntu
linux package