LVS
lvs-users
Google
 
Web LinuxVirtualServer.org

Re: ftp problem

To: Joseph Mack <mack.joseph@xxxxxxx>
Subject: Re: ftp problem
Cc: lvs-users@xxxxxxxxxxxxxxxxxxxxxx
From: Stefan Peter <stefan@xxxxxxxxxxxxxx>
Date: Thu, 25 Jul 2002 15:51:02 +0200 (CEST)
Hi All,

> You found the evidence yourself. If you schedule 20 and 21, then they
> will be regarded as separate services and can be sent to different 
> realservers.
> If you just schedule 21, then the connection to port 20 will be provided
> by ip_vs_ftp (for NAT) or by persistence for DR or Tun.

I see. I have a next question (in fact many more questions) regarding the
previous LVS arrangement I have described in my previous mail.
I tested my cluster using testlvs code and I found something interesting:

1. I have two distributor a master and a backup one. When I was spoofing
the cluster with large number of sources and large number of packets the
two distributors were alternating their functions at aprox. 5-10 seconds.
First the master distributor produced large kernel load (even my shell on
that machine has gone away), then the backup took its role, it got staffed
as did the master in a couple of seconds, but by this time the master
had come back again. Can I accept this as normal behavior?

2. The measurement values were also interesting: on the distributor
machine I measured 26000 packets/sec, as long as on the RSs I measured
3000 packets/sec. I have 4 RS, and I cannot figure out where the packets
have been lost? 26000!=4*3000. On top of all, on the client I measured
even larger speed. I looked into the switches, but I found no packet loss
or error.
(I used TCP spoof with srcnum 500000 and packets 10^8.)
(Machines are IBM330s, 512MBs of RAM, 1GHz Pentiums inside, RedHat 7.3
with 2.4.19-prexx kernel, the network is 100Mbit swiched network.)

ifconfig -a eth0 tells that:

eth0      Link encap:Ethernet  HWaddr 00:09:6B:58:56:51
          ...
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:421318133 errors:0 dropped:0 overruns:0 frame:0
          TX packets:296960432 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:100
          RX bytes:1242409506 (1184.8 Mb)  TX bytes:89641936 (85.4 Mb)

This may give explanation to the packet loss, at least in my
interpretation. Question again: is it normal operation? (Though spoofing
is abnormal.) How bottlenecks can be determined?

3. An off-topic question:

How many VIPs can be configured. On what property it depends?

Regards,
Peter



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