[Bug 275063] kernel using 100% CPU in arc_prune
- Reply: bugzilla-noreply_a_freebsd.org: "[Bug 275063] kernel using 100% CPU in arc_prune"
- Reply: bugzilla-noreply_a_freebsd.org: "[Bug 275063] kernel using 100% CPU in arc_prune"
- Reply: bugzilla-noreply_a_freebsd.org: "[Bug 275063] kernel using 100% CPU in arc_prune"
- Reply: bugzilla-noreply_a_freebsd.org: "[Bug 275063] kernel using 100% CPU in arc_prune"
- Reply: bugzilla-noreply_a_freebsd.org: "[Bug 275063] kernel using 100% CPU in arc_prune"
- Reply: bugzilla-noreply_a_freebsd.org: "[Bug 275063] kernel using 100% CPU in arc_prune"
- Reply: bugzilla-noreply_a_freebsd.org: "[Bug 275063] kernel using 100% CPU in arc_prune"
- Reply: bugzilla-noreply_a_freebsd.org: "[Bug 275063] kernel using 100% CPU in arc_prune"
- Reply: bugzilla-noreply_a_freebsd.org: "[Bug 275063] kernel using 100% CPU in arc_prune"
- Reply: bugzilla-noreply_a_freebsd.org: "[Bug 275063] kernel using 100% CPU in arc_prune"
- Reply: bugzilla-noreply_a_freebsd.org: "[Bug 275063] kernel using 100% CPU in arc_prune"
- Reply: bugzilla-noreply_a_freebsd.org: "[Bug 275063] kernel using 100% CPU in arc_prune"
- Reply: bugzilla-noreply_a_freebsd.org: "[Bug 275063] kernel using 100% CPU in arc_prune"
- Reply: bugzilla-noreply_a_freebsd.org: "[Bug 275063] kernel using 100% CPU in arc_prune"
- Reply: bugzilla-noreply_a_freebsd.org: "[Bug 275063] kernel using 100% CPU in arc_prune"
- Reply: bugzilla-noreply_a_freebsd.org: "[Bug 275063] kernel using 100% CPU in arc_prune"
- Go to: [ bottom of page ] [ top of archives ] [ this month ]
Date: Mon, 13 Nov 2023 21:10:33 UTC
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=275063
Bug ID: 275063
Summary: kernel using 100% CPU in arc_prune
Product: Base System
Version: Unspecified
Hardware: Any
OS: Any
Status: New
Severity: Affects Only Me
Priority: ---
Component: kern
Assignee: bugs@FreeBSD.org
Reporter: d8zNeCFG@aon.at
Scenario:
- FreeBSD stable/14 at e4fb49e867ae70412205d1aade0b76fae158b551
- Doing an rsync from UFS to ZFS, where both are on the same physical disks
- The scenario always worked flawlessly on releng/12.4
- 4-core CPU
- Running top with H and S keys pressed (show system processes, show threads)
Result:
- 200% CPU time is used: 100% by rsync, another 100% by something labeled
"arc_prune" in top
- top snapshot:
last pid: 4132; load averages: 2.04, 1.92, 1.69
up 0+13:28:31 22:08:35
369 threads: 7 running, 342 sleeping, 20 waiting
CPU: 0.0% user, 0.0% nice, 49.9% system, 0.0% interrupt, 50.1% idle
Mem: 64M Active, 142M Inact, 222M Laundry, 2632M Wired, 276M Buf, 12G Free
ARC: 661M Total, 153M MFU, 394M MRU, 16M Anon, 14M Header, 68M Other
380M Compressed, 941M Uncompressed, 2.48:1 Ratio
Swap: 20G Total, 20G Free
PID USERNAME PRI NICE SIZE RES STATE C TIME WCPU COMMAND
0 root -8 - 0B 1376K CPU1 1 28:14 100.00%
kernel{arc_prune}
1491 root 132 0 26M 11M CPU2 2 21:04 96.99% rsync
10 root 187 ki31 0B 64K CPU0 0 765:33 83.53% idle{idle:
cpu0}
10 root 187 ki31 0B 64K RUN 2 763:29 68.92% idle{idle:
cpu2}
10 root 187 ki31 0B 64K RUN 3 762:42 47.56% idle{idle:
cpu3}
20 root -16 - 0B 16K vlruwt 3 0:43 3.07% vnlru
2 root -60 - 0B 64K WAIT 0 0:54 0.08% clock{clock
(0)}
4132 root 20 0 14M 4188K CPU3 3 0:00 0.06% top
11 root -64 - 0B 256K WAIT 0 0:44 0.01% intr{irq29:
ahci1}
4 root -16 - 0B 48K - 3 0:40 0.01% cam{doneq0}
2824 root 20 0 25M 13M select 0 0:00 0.01% xterm
17 root -16 - 0B 48K psleep 0 0:15 0.01%
pagedaemon{dom0}
779 root 20 0 33M 6976K select 3 0:05 0.01% mountd
698 root 20 0 13M 1840K select 0 0:04 0.01% powerd
16 root -16 - 0B 16K - 0 0:04 0.01%
rand_harvestq
2876 root 20 0 16M 4696K select 0 0:01 0.01% xload
0 root -64 - 0B 1376K - 3 0:03 0.00%
kernel{fw0_taskq}
11 root -64 - 0B 256K WAIT 2 0:53 0.00% intr{irq30:
re0}
695 root 20 0 19M 6028K select 0 0:03 0.00% ntpd
861 root 20 0 63M 24M select 3 0:01 0.00% httpd
21 root 16 - 0B 16K syncer 0 0:03 0.00% syncer
19 root 20 - 0B 144K sdflus 3 0:01 0.00% bufdaemon{/
worker}
15 root -8 - 0B 1104K mmp->m 2 0:01 0.00%
zfskern{mmp_thread_enter}
15 root -8 - 0B 1104K t->zth 0 0:07 0.00%
zfskern{arc_reap}
...
Expected result:
- The system should not slow considerably by spending a full core on
"arc_prune"
-- Martin
--
You are receiving this mail because:
You are the assignee for the bug.