|
Hello,
On Mon, 14 May 2001, Joseph Mack wrote:
> according to the ipvsadm man page, for "lc" scheduling, the
> new connections are assigned according to the number of
> "active connections". Is this the same as "ActConn" in the
> output of ipvsadm?
The formula is: ActConn * K + InActConn
where K can be 32 ot 50, I don't remember the last used value.
So, it is not only the active conns, this will break UDP.
> I've been running the polygraph simple.pg test over the weekend
> using rr scheduling on what (AFAIK) are 4 identical real-servers
> in a VS-NAT LVS. There are no ActConn and a large number of
> InActConn. Presumably the client makes a new connection for each
> request. Over a period of time the number of InActConn (and
> load average) on two of the real-servers drops to about 2% of the
> other two real-servers. Changing the scheduling to lc doesn't change
> this. I don't know why two of the real-servers are not getting
> their share of requests with rr, but I would have expected that
> whatever the problem was, then changing to "lc" would fix it.
The implicit persistence of TCP connection reuse can cause
such side effects even for RR. Is this TCP SYN test?
When the setup includes small number of hosts and the used rate is
big enough to reuse the client's port, the LVS detects existing
connections and new connections are not created. This is the reason
you can see some of the rs not to be used at all, even for such method
as RR.
> (I will have to look further to see if these real-servers
> are in fact identical)
For LC they are identical.
> If the number of "active connections" used to determine the
> scheduling is "ActConn", then for services which don't
> maintain connections, the scheduler won't have much information,
> just "0" for all real-servers?
>
> Joe
>
> --
> Joseph Mack PhD, Senior Systems Engineer, Lockheed Martin
> contractor to the National Environmental Supercomputer Center,
> mailto:mack.joseph@xxxxxxx ph# 919-541-0007, RTP, NC, USA
Regards
--
Julian Anastasov <ja@xxxxxx>
|
