[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

[WORKED-AROUND] Re: Unknown HZ value



<quote who="nate">

> mail:~# uptime
> Unknown HZ value! (0) Assume 100.
>  1:31pm  up 249 days,  2:39,  2 users,  load average: 0.10, 0.03,
>  0.01

did some more research and it seems it is a bug in the kernel,
but procps detects it and spits out the message. the procps
in woody claims to have a fix that disables this behavior

from the changelog:(2.0.7-6)
"A warning when libproc cannot calculate the HZ value (probably due
to a kernel bug) has been supressed by default because it broke
people's scripts unnecessarily."

however, the warning is not disabled by default. i compiled
it on a potato box and installed it and still had the error.
so i grepped the source and deleted the 1 line of code that
spit out the error, recompiled, reinstalled and its working
now! woohoo!!@. the line was proc/sysinfo.c line 165:

proc/sysinfo.c:165:    fprintf(stderr, "Unknown HZ value! (%ld) Assume
%ld.\n", h, Hertz);
i say worked around instead of solved since its still seems
its a kernel bug that goes back a long ways(i saw a post
from ben collins when he was running 2.2.6 or something
and had the same error)

i saw some discussion for a procps HZ patch on a redhat
mailing list with similar info, just mentioned a way
to turn off the warning but it was still considered
a bug in the kernel.

maybe my mrtg scripts will inverse again to show
proper data once again! i hope ..

nate






Reply to: