Hello,
This seems odd and very high. The ghc compilation was taking its
time, so I thought to look at the stats. I saw that from last
Fri 6PM, named was running with 95% + CPU!!!? I killed and
restarted, and the results below are similar.
ps -augx > named-cpu-hog
Attachment:
named-cpu-hog
Description: Binary data
bash-4.2$ uname -a
NetBSD peano.jtcl.org 6.0_STABLE NetBSD 6.0_STABLE (Peano4 -r netbsd-6
-A -D 10/18/2012 (today) auich*) #0: Thu Oct 18 07:10:04 CDT 2012
root%peano.jtcl.org@localhost:/usr/obj/sys/arch/i386/compile/Peano4 i386
bash-4.2$ uptime
4:28PM up 2 days, 22:58, 2 users, load averages: 1.35, 1.49, 1.48
bash-4.2$
2 users Load 2.15 1.81 1.61 Mon Nov 5
16:33:21
Proc:r d s Csw Trp Sys Int Sof Flt PAGING SWAPPING
1 17 136763 378013 32 106 in out in out
ops
79.4% Sy 20.6% Us 0.0% Ni 0.0% In 0.0% Id pages
| | | | | | | | | | |
========================================>>>>>>>>>> forks
fkppw
memory totals (in kB) 132 Interrupts fksvm
real virtual free 100 cpu0 timer pwait
Active 684376 688664 555136 ioapic0 pin 9 relck
All 970244 974532 6842780 ioapic0 pin 1 rlkok
32 ioapic0 pin 12 noram
Namei Sys-cache Proc-cache ioapic0 pin 16 ndcpy
Calls hits % hits % ioapic0 pin 14 fltcp
43 43 100 ioapic0 pin 15 zfod
ioapic0 pin 7 cow
Disks: sd0 wd0 cd0 256 fmin
seeks 341 ftarg
xfers itarg
bytes 1550 wired
%busy pdfre
pdscn
I can rebuild from new cvs pull when I get to it, but I have no
set of tools built up for analyzing kernel and bind, named
issues, and where to dive in. From the 7/25/2012
DNS security warning, I had followed the directions and added
external/bind/dist post 7/25 in an extra step with cvs. I did
not with the 6.0_STABLE announcement and cvs pull. Suggestions?
I can live with the load until I get the new ghc built and
functioning for the job I need/want it. Thanks in advance.
John R. Towler
jtowler%soncom.com@localhost