Hi , one question.

I would like to know if  I can use  a directive in pound to do a check url
to backend for check better the backend status, actually i know that pound
check only backend ports, no backend service status with a GET for example.

Is it for new versions? Is it implemented?

Thanks

Emilio,

You seem to understand that you need to use HAport and you're correct. The
HAport directive only checks for the existence of a open or closed port. If the
port is closed, the BackEnd is considered dead. For example, let's say your
configuration file looks like this:

Service
    BackEnd
        Address 192.168.x.x
        Port    80
        HAport  192.168.x.x 8080
    End
End

Every 'Alive' seconds Pound will check for a open port on 192.168.x.x:8080. The
state of the HAport should be controlled by an intelligent application or
script. Your program can do something as simple as GET or it might actually
being doing intelligent business logic to determine if a BackEnd should be
removed. Depending on your skill set and time frame you may want to look at
Python as your 'HAport' control language. It supports socket, url handling and
many other useful modules and classes to quickly make a robust application.

That being said, it is out of scope for Pound to do error handling at the level
you're concerned with. The HAport directive powered by an intelligent
application is the way to go.

Thanks,
Chris


-----Original Message-----
From: Emilio Campos [mailto:emilio.campos.martin(at)gmail.com]
Sent: Thursday, April 08, 2010 5:23 AM
To: pound(at)apsis.ch
Subject: [Pound Mailing List] scripts for check backends status

Hi , one question.

I would like to know if  I can use  a directive in pound to do a check url
to backend for check better the backend status, actually i know that pound
check only backend ports, no backend service status with a GET for example.

Is it for new versions? Is it implemented?

Thanks


--
To unsubscribe send an email with subject unsubscribe to pound(at)apsis.ch.
Please contact roseg(at)apsis.ch for questions.

[...]

Chris
very grateful for the clarification

Thanks!

2010/4/8 Chris Morrow <cmorrow(at)verrus.com>
[...]