memory use, bidirectional behavior
MN
m.newton at stanford.edu
Fri Aug 24 18:58:52 EDT 2007
Hi - The figures were for the non-smallmemory. For small memory, the
memory stays at about 75M (with both directions seen) or 4M (with
one direction seen).
With argus-3.0.0.smallmemory.rc.2.tar.gz, the anomalous behavior where
seeing 1 direction pegs the load average at ~1, with < 1% to argus
vs seeing both directions has a lower load average, all of which
is attributed to Argus
remains.
This is with no flags or .X files - just straight off the server.
- mike
ps: In more detail, with the 'top's being sorted by CPU utilization:
Seeing one direction:
top - 15:46:42 up 2 days, 38 min, 1 user, load average: 0.99, 1.03, 1.32
Tasks: 71 total, 1 running, 70 sleeping, 0 stopped, 0 zombie
Cpu0 : 0.0%us, 0.0%sy, 0.0%ni, 91.4%id, 0.0%wa, 1.3%hi, 7.3%si, 0.0%st
Cpu1 : 0.0%us, 0.0%sy, 0.0%ni,100.0%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st
Cpu2 : 0.0%us, 0.0%sy, 0.0%ni, 92.7%id, 0.0%wa, 2.7%hi, 4.7%si, 0.0%st
Cpu3 : 0.0%us, 0.0%sy, 0.0%ni,100.0%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st
Mem: 2075440k total, 120908k used, 1954532k free, 1800k buffers
Swap: 7815612k total, 8432k used, 7807180k free, 87960k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
27540 root 18 0 4136 1648 508 R 0 0.1 0:00.00 argus
Load average=.99, but no one's using antyhing, versus..
Seeing both directions, after a while:
top - 15:55:06 up 2 days, 46 min, 1 user, load average: 0.78, 0.78, 1.06
Tasks: 71 total, 2 running, 69 sleeping, 0 stopped, 0 zombie
Cpu0 : 9.3%us, 15.3%sy, 0.0%ni, 52.3%id, 0.0%wa, 3.3%hi, 19.7%si, 0.0%st
Cpu1 : 0.0%us, 0.3%sy, 0.0%ni, 99.7%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st
Cpu2 : 11.0%us, 17.0%sy, 0.0%ni, 46.3%id, 0.0%wa, 4.0%hi, 21.7%si, 0.0%st
Cpu3 : 0.0%us, 0.0%sy, 0.0%ni,100.0%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st
Mem: 2075440k total, 404764k used, 1670676k free, 3028k buffers
Swap: 7815612k total, 8072k used, 7807540k free, 284680k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
27540 root 15 0 87160 82m 508 R 72 4.1 4:50.62 argus
Argus=~= 72%, load average =~ .78 ... which seems more reasonable. The
difference could be just the way that the kernel attributes cycles to
processes.
> Message: 6
> Date: Fri, 24 Aug 2007 13:46:54 -0700
> From: Peter Van Epp <vanepp at sfu.ca>
> Subject: Re: [ARGUS] memory use, bidirectional behavior
> To: argus-info at lists.andrew.cmu.edu
> Message-ID: <20070824204654.GC13707 at sfu.ca>
> Content-Type: text/plain; charset=us-ascii
>
> On Fri, Aug 24, 2007 at 01:16:57PM -0700, MN wrote:
> >
> > A few data points that might possibly help with the memory leaks and
> > which also show a strange behavior regarding CPU utilization as well:
> >
> > I've been testing Argus on some new donated boxes with 10G cards (266MHz
> > PCI-X buses, x86/32bit), but with only 1G interfaces currently in use.
> > The two directions of the currently tapped link show up on eth0 and eth1.
> >
> <snip>
>
> This seems odd! Which argus are us using
> /dev/argus-3.0/argus-3.0.0.tar.gz or
> /dev/argus-3.0/argus-3.0.0.smallmemory.rc.2
>
> with /dev/argus-3.0/argus-3.0.0.smallmemory.rc.2 on an IBM quad core Power5
> 64 bit PPC box with dual Intel gig nics argus-3.0.0.smallmemory.rc.2 and
> SUSE 10.2 with the pf-ring kernel mods installed has been running fine since
> last night on dual interfaces (as Carter noted the -U flag and I think -J
> for jitter aren't implemented yet) and apparantly putting out good data.
>
> root 23980 5.1 5.3 278904 211660 ? SLl Aug23 54:06 argus -JR -P 560 -i eth0 -i eth1 -U 512 -m -F /scratch/argus.conf
> vanepp 25602 0.0 0.0 3132 832 pts/0 S+ 13:36 0:00 grep argus
>
>
> top - 13:42:31 up 2 days, 5:13, 1 user, load average: 0.07, 0.07, 0.08
> Tasks: 78 total, 1 running, 77 sleeping, 0 stopped, 0 zombie
> Cpu(s): 2.8%us, 1.0%sy, 0.0%ni, 94.2%id, 0.0%wa, 0.2%hi, 1.8%si, 0.0%st
> Mem: 3936076k total, 1502456k used, 2433620k free, 408036k buffers
> Swap: 6202496k total, 0k used, 6202496k free, 361340k cached
>
> PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
> 23980 root 15 0 304m 206m 1424 S 7 5.4 54:33.90 argus
> 25605 vanepp 15 0 2720 1188 948 R 0 0.0 0:00.01 top
>
> I was seeing bad things on argus-3.0.0 before this but the two
> smallmemory releases seem to have fixed the bloat.
>
> Peter Van Epp / Operations and Technical Support
> Simon Fraser University, Burnaby, B.C. Canada
More information about the argus
mailing list