Re: [lvs-users] Fwd: InActConn dilemma

To: " users mailing list." <lvs-users@xxxxxxxxxxxxxxxxxxxxxx>
Subject: Re: [lvs-users] Fwd: InActConn dilemma
From: Graeme Fowler <graeme@xxxxxxxxxxx>
Date: Sun, 14 Feb 2010 20:59:07 +0000
On Sun, 2010-02-14 at 11:58 +0100, Andre Magri wrote:
> That was a great explanation Graeme thank you so much.  Indeed enable
> keepalives was disabled on www1.

Bingo :)

> I have one more question for you though :)

Oh, go on then...

> Wouldn't it be less taxing on the loadbalancer to have keepalives enabled?

No, not really. The cost of session setup/teardown is pretty small
compared to the quantity of data pushed over the connection; also
keeping a session in the active state costs slightly more than having it
inactive so keepalives can appear more costly to both the director and
the webserver itself.

In practice, with a large number of clients from a large number of
networks you'll find that there's a good chance a significant proportion
of the clients will either not request keepalives at all, or be behind
proxies which don't.

> At one point we had a slowdown on the responsiveness of the site and I
> thought at the time that having all those inactive connections were a
> contributing factor.

I would doubt it. You'd have to have a *lot* - literally millions, if
not billions - to get to that state, and you'd have log messages about


Please read the documentation before posting - it's available at: mailing list - lvs-users@xxxxxxxxxxxxxxxxxxxxxx
Send requests to lvs-users-request@xxxxxxxxxxxxxxxxxxxxxx
or go to

<Prev in Thread] Current Thread [Next in Thread>