On Thu Aug 16, 2012, Thomas Fjellstrom wrote:
> On Thu Aug 16, 2012, Chris Wilson wrote:
> > Hi Thomas,
> > 
> > On Thu, 16 Aug 2012, Thomas Fjellstrom wrote:
> > > I have a 6501-50 and installed Debian Sid onto a 32G mSATA SSD.
> > > 
> > > I haven't done extensive testing but performance seems to be good
> > > network wise. However when I run htop it shows a near constant 100%
> > > cpu use most of which is tagged as kernel use but it isn't assigned to
> > > any specific kernel threads.
> > > 
> > > Has anyone witnessed anything like this on their 6501's?
> > 
> > That looks like a bug in htop. I can see the same thing on Squeeze and a
> > custom 3.2 kernel, except that it's not accounted to kernel, it's not
> > 
> > accounted at all:
> >    1  [******************************************************100.0%]
> > 
> > Tasks: 93 total, 1 running
> > 
> >    2  [##***                                                   5.6%]
> > 
> > Load average: 0.43 0.65 0.74
> > 
> >    Mem[|||||||||||||||||||||############******************306/882MB]
> > 
> > Uptime: 2 days, 22:07:03
> > 
> >    Swp[                                                       0/0MB]
> >    
> >    PID USER     PRI  NI  VIRT   RES   SHR S CPU% MEM%   TIME+  Command
> >   
> >   7764 root      20   0  2552  1256   964 R  9.0  0.1  0:00.42 htop
> >   
> >   2423 root      20   0 24380 15096 14452 S  3.0  1.7  2h46:45 pmacctd:
> >    Core Process [default]
> >   
> >   2462 proxy     20   0  180M  155M  2256 S  1.0 17.6 57:16.83 (squid)
> >   -YC
> >   
> >    -f /etc/squid3/squid.conf
> > 
> > 11329 dnsmasq   20   0  5312  1596   708 S  1.0  0.2  2:19.03
> > 
> >    /usr/sbin/dnsmasq -x /var/run/dnsmasq/dnsmasq.pid -u dnsmasq -r
> >    /var/run/dnsmas
> >   
> >   2464 root      20   0 21280 11556 10776 S  0.0  1.3 19:24.03 pmacctd:
> >   IMT
> >   
> >    Plugin [default]
> >   
> >   2591 tomcat6   20   0  263M 39792  2020 S  0.0  4.4 14:21.65
> >   
> >    /usr/lib/jvm/java-6-openjdk/bin/java
> >    -Djava.util.logging.config.file=/var/lib/t
> >   
> >   3531 mysql     20   0  133M 29092  1948 S  0.0  3.2  1:55.54
> >   
> >    /usr/sbin/mysqld --basedir=/usr --datadir=/media/hdb1/mysql
> >    --user=mysql --pid-
> > 
> > While top shows that the system is mostly idle:
> > 
> > top - 11:43:34 up 2 days, 22:05,  2 users,  load average: 0.49, 0.71,
> > 0.76 Tasks: 115 total,   1 running, 113 sleeping,   0 stopped,   1
> > zombie Cpu0  :  0.0%us,  1.8%sy,  0.0%ni, 90.1%id,  0.0%wa,  0.0%hi, 
> > 8.1%si, 0.0%st
> > Cpu1  :  1.0%us,  1.0%sy,  0.0%ni, 98.0%id,  0.0%wa,  0.0%hi,  0.0%si,
> > 0.0%st
> > Mem:    903632k total,   884628k used,    19004k free,   176692k buffers
> > Swap:        0k total,        0k used,        0k free,   393492k cached
> > 
> >    PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND
> >   
> >   2423 root      20   0 24380  14m  14m S    6  1.7 166:41.26 pmacctd
> >   7614 root      20   0  2504 1164  896 R    1  0.1   0:01.99 top
> >   1014 nobody    20   0  6168 3716 1744 S    1  0.4  31:17.17 openvpn
> >   
> >    719 root      20   0     0    0    0 S    0  0.0   0:31.93 flush-254:1
> >   
> >   2462 proxy     20   0  180m 155m 2256 S    0 17.6  57:16.04 squid3
> > 
> > htop might be a bit broken.
> > 
> > Cheers, Chris.
> 
> Possibly. But I find it strange that it only happens on the soekris. It
> behaves fine on every other machine I have (couple amd quadcores, intel
> xeon quadcore, core2duo laptop etc), including a Raspberry Pi.

So I've been having networking problems with the soekris. I tried changing to 
the -486 debian kernel, then I tried a 2.6.32 kernel that's linked to on the 
wiki and that seems to have solved the weird problems I was having including 
htop showing 100% cpu.

-- 
Thomas Fjellstrom
tho...@fjellstrom.ca
_______________________________________________
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech

Reply via email to