(RADIATOR) how the FailureBackOffTime works?

[TRUONG, Vi-thang]

Hi,

For the clause "<AuthBy RADIUS>", we can use the parameter
FailureBackoffTime.
Does anyone use this parameter and can explain me how it works ?
I find this in the document:

6.29.8 FailureBackoffTime
This optional parameter specifies how long a failed remote server will be
removed from
the forwarding list. If no reply is received from a Host (after all the
Retries have
expired), it will be marked as failed for FailureBackoffTime seconds. After
that time has
expired, it will again be eligible for forwarding. The default is 0, which
means that the
host is always regarded as working.
Caution: with most types of load balancing modules, the default of 0 will
mean endless
retransmission of each request until a reply is received.

I want to know what RADIATOR does with this scenario: I use two remote
server ( a primary and a secondary) and the primary does not reply, RADIATOR
remove it from the forwarding list during 2 minutes for example
(FailureBackoffTime=120s) and try the secondary  but the secondary also gets
down during this time.
So RADIATOR remove the secondary from the forwarding list during 2 minutes
also. But what does it do then? there is no remote server in the forwarding
list. Does it try to check if the primary is up now?

Does it also have the same effect when it is using in a clause like <AuthBy
ROUNDROBIN>, <AuthBy VOLUMEBALANCE>, <AuthBy LOADBALANCE>.

Thanks for your help,
Vi-Thang.
===
Archive at http://www.open.com.au/archives/radiator/
Announcements on radiator-announce at open.com.au
To unsubscribe, email 'majordomo at open.com.au' with
'unsubscribe radiator' in the body of the message.