[1/2OT]Multi-tier client-server apps monitorin g w/Nagios integration?!?

[stefmit]

The statements below are correct. For further discussion, let's take a very 
popular example: Oracle applications. This environment is made usually out of 
(the simplest way) a web/forms based server, to which the clients connect, 
with a database server "behind" it. At first appearance, it may look like 
something easily achievebale (the measurements, that is): one would just have 
to place a sniffer at each point of the multi-tier path, then consolidate the 
info. BUT (at least as far as my experience goes) the client-web/forms 
communication (which could be easily identified by simple methods such as the 
socket pair (IP and TCP port)) is not that easily tracked in the backend, 
simply because the web/forms <--> database communication takes place on 
behalf of ALL clients connecting to the from end, with no easy way to 
segregate.

How would I do it? The only way I could see this being feasible would be to 
gain the knowledge (from Oracle) on what identifies a specific session 
(attention: even a username is not good, as the web/forms <--> database may 
use its own appmanager to talk on behalf of all front-end clients, thus no 
segregation possible this way), INSIDE the  payload of a TCP packet, and have 
the sniffer (perhaps a sort of rule written for Snort - or at least this is 
what I was planning to do) filter for those specific entries, in order to 
distinguish in all that traffic what belongs to whom. 

Even if the above would be feasible (and - let's say - perhaps even without 
Oracle's support, as someone may trace long enough such traffic as to 
identify common patterns, and thus extract the identification for 
sessions?!?), the challenge still remains to consolidate this info with the 
one coming from the client <--> web/forms server trace. And this is what I 
was asking to begin with: who - among our peers on this list - has been able 
to achieve this using Nagios (and - perhaps - some plugins)?

There was one response earlier this week from someone who stated having done 
just that. Would this person be kind enough to share (if not specific 
plugins, if they would be considered "unsharable" with our community), at 
least the fundamental concepts having allowed them to achieve what I 
described above? 

Thx,
Stef


On Wednesday 18 June 2003 07:25 pm, Stanley Hopcroft wrote:
> Dear Ladies and Gentlemen,
>
> With respect, I think the orginal question was to _analyse_ end-user
> (total or measured) response times across a multi-tier application by
> accounting for the processing time consumed by components in each tier
> eg
>
> Average Total_response time for transaction = 25 seconds
>
> Avarage times in components
>
> % in DB         13%
> % in WAN        5%
> % in LAN        0.05%
> % in app server 45%
> % in web server 5%
>
> sort of analysis.



-------------------------------------------------------
This SF.Net email is sponsored by: INetU
Attention Web Developers & Consultants: Become An INetU Hosting Partner.
Refer Dedicated Servers. We Manage Them. You Get 10% Monthly Commission!
INetU Dedicated Managed Hosting http://www.inetu.net/partner/index.php
_______________________________________________
Nagios-users mailing list
Nagios-users at lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nagios-users
::: Please include Nagios version, plugin version (-v) and OS when reporting any issue. 
::: Messages without supporting info will risk being sent to /dev/null