Hi,
using ntop 3.0 on debian sarge together with PF_RING3 and NAPI (e1000)
with Kernel 2.6.
ntop is running great,
With the exception that rrd graphs have some holes, depending on the
hardware (P4 3.2G or Xeon, 2GByte RAM) and depending on the monitored
traffic.
Snort/barnyard is also running on the same hardware
I think, that I have a problem with the disk I/O. Output of vmstat and
iostat see below.
To try to fix the problem
- I did use option -m for local network.
- The rrds are updated only every 15min instead 5min.
- I also used a filter in the rrd settings for the local network.
This helped for some probes, but not for all.
Has anybody further suggestions how to tune ntop with rrd?
Some suggestion how to find the bottleneck?
Many thanks, Roger
procs -----------memory---------- ---swap-- -----io---- --system-- ----cpu----
r b swpd free buff cache si so bi bo in cs us sy id wa
0 1 22308 7464 79776 513404 0 0 23 27 3 17 6 3 60 31
Linux 2.6.8 10/05/2005
avg-cpu: %user %nice %sys %iowait %idle
6.37 0.00 2.69 31.20 59.74
Device: tps Blk_read/s Blk_wrtn/s Blk_read Blk_wrtn
hda 1.33 2.98 20.07 2048954 13796240
hdb 140.04 2338.92 1087.38 1607466130 747324496
avg-cpu: %user %nice %sys %iowait %idle
6.97 0.00 4.48 7.46 81.09
Device: tps Blk_read/s Blk_wrtn/s Blk_read Blk_wrtn
hda 0.00 0.00 0.00 0 0
hdb 25.74 205.94 0.00 208 0
avg-cpu: %user %nice %sys %iowait %idle
7.04 0.00 4.52 4.52 83.92
Device: tps Blk_read/s Blk_wrtn/s Blk_read Blk_wrtn
hda 5.05 0.00 169.70 0 168
hdb 21.21 169.70 0.00 168 0
_______________________________________________
Ntop mailing list
[email protected]
http://listgateway.unipi.it/mailman/listinfo/ntop
