Unknown HZ value
i saw that my mail hub decided to start showing this now
too.. first it was my web server. seems to be somewhat
related to uptime. my webserver started it a few months ago
(it has about 390 days uptime) and my mailserver just recently
(it has about 249 days uptime)
seems to be a kernel bug but wanted to confirm ...i read
a thread on the kernel list but only saw discussion on what
could of been a bug. both systems running 2.2.19
now that i think about it it is probably the cause of
the inverse results ive been getting for uptime, cpu load
and memory usage from mrtg from yet another server that
recently passed the 220-230 day uptime. a couple weeks ago
the mrtg graphs for a mailserver completly inversed out
of nowhere(nothing changed on the system).
very odd! here is the error:
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
Linux mail 2.2.19 #1 SMP Wed Mar 28 10:23:03 PST 2001 i686 unknown
admin@mail-ca2:~$ uptime
Unknown HZ value! (3) Assume 100.
1:33pm up 256 days, 18:02, 1 user, load average: 0.06, 0.03, 0.01
on the kernel mailing list i saw it seemed to be specifically
related to SMP systems and all 3 that are affected are dual proc.
yep that is the cause! i just checked my mrtg graphs for the mail
hub and now they are inversed. the thread mentioned something
about idle time looping..and it seems it did, from procinfo
on the mail server:
user : 9d 3:18:56.01 0.0% page in : 54302702 disk 1:
13365391r114547578wnice : 1d 7:31:10.76 0.0% page out:324831048
disk 2: 8805788r47393718w
system: 19d 9:21:42.79 0.0% swap in : 109345
idle : 483d 15:57:35.24 0.0% swap out: 12434
uptime: 256d 18:04:42.39 context :1562566116
damnit!
maybe time to post to the kernel list ..
never thought there would be a downside to uptime :) !@#
nate
Reply to: