I may install it tonight,,, after another "miller"
-----Original Message-----
From: REVES,JOSEPH (HP-ColSprings,ex1) [mailto:joe_reves@hp.com]
Sent: Monday, July 16, 2001 6:54 PM
To: 'traffic-management@inmon.com'
Subject: RE: Announce: Traffic Server 2.0.21
This is so damn cool! I haven't seen this representation running in
*years*.
We have a question right now about the HSG -> HP traffic: how do we
visualize it? Can we filter what's displayed in the circles view? (you knew
that was coming, huh? ;-)
There's still a few folks around who will really appreciate this, and it's
fun to see the rest of the reactions to this "novel" representation. Very
nice, Neil. Thanks!
Very cool enterprise management features, as well.
joer
Joseph Reves
Hewlett-Packard
Enterprise Network Organization
Tel: +1 719 351 9621
ECard: https://ecardfile.com/id/jreves <https://ecardfile.com/id/jreves>
-----Original Message-----
From: Neil McKee [mailto:neil_mckee@inmon.com]
Sent: Monday, July 16, 2001 4:04 PM
To: traffic-management@inmon.com
Subject: Announce: Traffic Server 2.0.21
Traffic Server version 2.0.21
is now available for download.
Changes include:
* Circles visualization. Under the menu option Monitor->Traffic
click the top left box to see the top interfaces by frames, then change the
chart option at the bottom from "segment counters" to "circles". This is a
visualization of the top 100 traffic flows in the last minute, show in a
spatial arrangement rather than on a time-line, with the most common
protocols represented by different colors.
* Some tips for driving this tool are:
* draw box to zoom in.
* click right button to zoom out again.
* click on a node or flow to label it.
* double-click on a node (or its label) to open a host report.
* double-click on a link (or its label) to see that flow over time.
* double-click on a protocol at the top to select only that protocol.
* double-click on the "other" protocol, to add a filter and see the
next 5 protocols.
* The options at the bottom right are:
* subnet - cluster by subnet, using subnets discovered from InMon
sFlow or Cisco NetFlow data sources.
* agent - cluster by switch agents, using end-node location
information discovered by the Traffic Server.
* zone - cluster by zone, using the [zone] definitions from the server
configuration file.
* zone and subnet - cluster by zone and subnet, using the [zone] and
[subnet] definitions from the server configuration file.
* Multi-site Monitor. Under the new menu option Multi-site->Monitor
you will now be able to see an integrated status map with a row of status
buttons for every monitored site in your enterprise. This enterprise-wide
view allows the minute-by-minute status of every port in your enterprise to
be reflected on one web page. Clicking to drill down will take you to the
relevant page on the Traffic Server for that site.
* Multi-site Configuration. The server configuration file can now use
#include <URL> statements to import configuration from a web page stored
somewhere else. For example, if your organization has a number of monitored
sites, you can define a "master" configuration file for the enterprise which
they all use. This master configuration file would define the [enterprise]
section, set the default thresholds and sampling rates, and then use a
series of include statements to import the [site] sections from each remote
site. This way each site can be administered separately, but as long
as each [site] section is accessible via a URL, then the other traffic
servers can take advantage of it too.
* TraceTargets. The traceroute engine can now be seeded with a set of
default targets that it should run periodic trace tests to all the time
(even if those remote subnets do not communicate heavily with your site).
This is configured by setting
TraceTargets=<ip1>/<mask1>,<ip2>/<mask2>,...,<ipN>/<maskN> somewhere in the
[site] section of the configuration file (next to the TraceEnabled=YES
setting). The <ip> settings here should be fully specified IP addresses of
reliable hosts in that remote subnet. The <mask> settings can either be the
number of bits in the IP subnet prefix, or a fully specified mask. For
example, both "TraceTargets=216.167.121.121/24" and
"TraceTargets=216.167.121.121/255.255.255.0" are valid.
-----------
Neil McKee mailto:Neil_McKee@InMon.com <mailto:Neil_McKee@InMon.com>
InMon Corp. //www.InMon.com <//www.InMon.com>
This archive was generated by hypermail 2b29 : 07/16/01 EDT