All of lore.kernel.org
 help / color / mirror / Atom feed
From: Lin Zhemin <ljm@ljm.qqjane.net>
To: lartc@vger.kernel.org
Subject: [LARTC] Question: TEQL / Equal Cost setup: failed
Date: Tue, 03 Apr 2001 02:33:11 +0000	[thread overview]
Message-ID: <marc-lartc-98626523828589@msgid-missing> (raw)

Hi, all.

I've tried to set up load balance using both of TEQL and nexthop
but in vain. Please tell me where should I correct my setup/notions.
Thanks a lot. I use a version 2.2.16-22 kernel (from Redhat 7.0
precompiled).


Connectivity:
======---------
| PC1   |(211.23.43.88)   (211.23.43.81)  /28 network
|   eth0|---------------- ADSL Router 1 ------- Same ISP but
|       |(211.75.156.252) (211.75.156.249)/29 network
|   eth1|---------------- ADSL Router 2 ------- different gateway
|       |
|   eth2|(intranet)
---------


Using TEQL:
=====[root@clust3 /root]# tc qdisc
qdisc teql0 8003: dev eth0
qdisc teql0 8004: dev eth1

[root@clust3 /root]# ip addr
1: lo: <LOOPBACK,UP> mtu 3924 qdisc noqueue
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 brd 127.255.255.255 scope host lo
2: eth0: <BROADCAST,MULTICAST,UP> mtu 1500 qdisc teql0 qlen 100
    link/ether 00:c0:f6:b1:68:31 brd ff:ff:ff:ff:ff:ff
    inet 211.23.43.88/28 brd 211.23.43.95 scope global eth0
3: eth1: <BROADCAST,MULTICAST,UP> mtu 1500 qdisc teql0 qlen 100
    link/ether 00:e0:4c:39:74:de brd ff:ff:ff:ff:ff:ff
    inet 211.75.156.252/29 brd 211.75.156.255 scope global eth1
4: eth2: <BROADCAST,MULTICAST,UP> mtu 1500 qdisc pfifo_fast qlen 100
    link/ether 00:c0:f6:b1:57:e1 brd ff:ff:ff:ff:ff:ff
    inet 192.168.0.224/24 brd 192.168.0.255 scope global eth2
5: teql0: <BROADCAST,NOARP,UP> mtu 1500 qdisc pfifo_fast qlen 100
    link/generic
    inet 211.75.156.252/29 scope global teql0
    inet 211.23.43.88/28 scope global teql0

[root@clust3 /root]# ip route
211.75.156.248/29 dev teql0  proto kernel  scope link  src
211.75.156.252
211.23.43.80/28 dev teql0  proto kernel  scope link  src 211.23.43.88
192.168.0.0/24 dev eth2  proto kernel  scope link  src 192.168.0.224
127.0.0.0/8 dev lo  scope link
default dev teql0  scope link

[root@clust3 /root]# ping 211.23.43.81
PING 211.23.43.81 (211.23.43.81) from 211.23.43.88 : 56(84) bytes of data.
64 bytes from gw0 (211.23.43.81): icmp_seq=0 ttl%4 time=1.055 msec

[root@clust3 /root]# ping 211.75.156.249
PING 211.75.156.249 (211.75.156.249) from 211.75.156.252 : 56(84) bytes of data.
64 bytes from gw1 (211.75.156.249): icmp_seq=0 ttld time=1.149 msec

[root@clust3 /root]# ping 211.23.16.116  (some internet address)
PING 211.23.16.116 (211.23.16.116) from 211.75.156.252 : 56(84) bytes of data.
From lh1 (211.75.156.252): Destination Host Unreachable

[root@clust3 /root]# traceroute -n 211.23.16.116
traceroute to 211.23.16.116 (211.23.16.116), 30 hops max, 38 byte packets
 1  211.23.43.88  3009.191 ms !H  2999.928 ms !H  3009.885 ms !H




Using TEQL but more tricky
============= (other parameters stayed the same)
[root@clust3 /root]# ip route
211.75.156.248/29 dev teql0  proto kernel  scope link  src
211.75.156.252
211.23.43.80/28 dev teql0  proto kernel  scope link  src 211.23.43.88
192.168.0.0/24 dev eth2  proto kernel  scope link  src 192.168.0.224
127.0.0.0/8 dev lo  scope link
default equalize
        nexthop via 211.23.43.81  dev teql0 weight 1
        nexthop via 211.75.156.249  dev teql0 weight 1

[root@clust3 /root]# ping ljm
PING ljm (211.23.16.116) from 211.23.43.88 : 56(84) bytes of data.
  (no response here)
--- ljm ping statistics ---
26 packets transmitted, 0 packets received, 100% packet loss

[root@clust3 /root]# traceroute -n ljm
traceroute to ljm (211.23.16.116), 30 hops max, 38 byte packets
 1  211.23.43.81  1.200 ms  1.106 ms  1.097 ms
 2  10.23.43.254  52.294 ms  50.076 ms  49.577 ms
 3  168.95.225.94  42.060 ms  39.402 ms  47.919 ms
 4  168.95.225.69  87.703 ms  49.975 ms  49.951 ms
 5  211.23.16.116  90.643 ms  101.042 ms  137.353 ms

(Test at 211.23.16.116)
ljm:~$ ping 211.75.156.252
PING 211.75.156.252 (211.75.156.252) from 211.23.16.116 : 56(84) bytes of data.
64 bytes from 211.75.156.252: icmp_seq=0 ttl%1 timeX3.872 msec

ljm:~$ ping 211.23.43.88
PING 211.23.43.88 (211.23.43.88) from 211.23.16.116 : 56(84) bytes of data.

--- 211.23.43.88 ping statistics ---
6 packets transmitted, 0 packets received, 100% packet loss

ljm:~$ traceroute -n 211.23.43.88
traceroute to 211.23.43.88 (211.23.43.88), 30 hops max, 40 byte packets
 1  211.23.16.254  63.665 ms  49.588 ms  50.368 ms
 2  168.95.225.62  52.108 ms  45.403 ms  46.223 ms
 3  168.95.225.36  52.886 ms  49.627 ms  50.394 ms
 4  * * * 

ljm:~$ traceroute -n 211.75.156.252
traceroute to 211.75.156.252 (211.75.156.252), 30 hops max, 40 byte packets
 1  211.23.16.254  50.059 ms  66.848 ms  49.462 ms
 2  168.95.225.62  46.122 ms  46.13 ms  46.496 ms
 3  168.95.225.34  66.211 ms  49.667 ms  50.291 ms
 4  211.75.156.249  90.048 ms  96.67 ms  140.575 ms
 5  211.75.156.252  84.296 ms  103.313 ms  81.974 ms



Equal Cost Without TEQL (To be sure, teql0 is dettached and set down)
===========[root@clust3 /root]# ip addr
1: lo: <LOOPBACK,UP> mtu 3924 qdisc noqueue
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 brd 127.255.255.255 scope host lo
2: eth0: <BROADCAST,MULTICAST,UP> mtu 1500 qdisc pfifo_fast qlen 100
    link/ether 00:c0:f6:b1:68:31 brd ff:ff:ff:ff:ff:ff
    inet 211.23.43.88/28 brd 211.23.43.95 scope global eth0
3: eth1: <BROADCAST,MULTICAST,UP> mtu 1500 qdisc pfifo_fast qlen 100
    link/ether 00:e0:4c:39:74:de brd ff:ff:ff:ff:ff:ff
    inet 211.75.156.252/29 brd 211.75.156.255 scope global eth1
4: eth2: <BROADCAST,MULTICAST,UP> mtu 1500 qdisc pfifo_fast qlen 100
    link/ether 00:c0:f6:b1:57:e1 brd ff:ff:ff:ff:ff:ff
    inet 192.168.0.224/24 brd 192.168.0.255 scope global eth2
5: teql0: <BROADCAST,NOARP> mtu 1500 qdisc pfifo_fast qlen 100
    link/generic

[root@clust3 /root]# ip route
211.75.156.248/29 dev eth1  scope link  src 211.75.156.252
211.23.43.80/28 dev eth0  scope link  src 211.23.43.88
192.168.0.0/24 dev eth2  proto kernel  scope link  src 192.168.0.224
127.0.0.0/8 dev lo  scope link
default equalize
        nexthop via 211.23.43.81  dev eth0 weight 1
        nexthop via 211.75.156.249  dev eth1 weight 1

[root@clust3 /root]# ping gw0
PING gw0 (211.23.43.81) from 211.23.43.88 : 56(84) bytes of data.
64 bytes from gw0 (211.23.43.81): icmp_seq=0 ttl%4 time=2.360 msec

[root@clust3 /root]# ping gw1
PING gw1 (211.75.156.249) from 211.75.156.252 : 56(84) bytes of data.
64 bytes from gw1 (211.75.156.249): icmp_seq=0 ttld time=1.848 msec

[root@clust3 /root]# ping -R ljm
PING ljm (211.23.16.116) from 211.23.43.88 : 56(124) bytes of data.
64 bytes from ljm (211.23.16.116): icmp_seq=0 ttl%1 time\x108.360 msec RR:     lh0 (211.23.43.88)
        0.0.0.0     <---- ??????
        tp-b-c6r6.router.hinet.net (168.95.225.94)
        ljm (211.23.16.116)
        ljm (211.23.16.116)
        tp-b-c6r5.router.hinet.net (168.95.225.62)
        gw0 (211.23.43.81)
        lh0 (211.23.43.88)

64 bytes from ljm (211.23.16.116): icmp_seq=1 ttl%1 time\x104.320 msec
(same route)

[root@clust3 /root]# traceroute -n ljm
traceroute to ljm (211.23.16.116), 30 hops max, 38 byte packets
 1  211.23.43.81  1.195 ms  1.101 ms  1.094 ms
 2  10.23.43.254  45.981 ms  49.400 ms  55.687 ms
 3  168.95.225.94  43.254 ms  39.596 ms  40.620 ms
 4  168.95.225.69  46.018 ms  49.208 ms  66.728 ms
 5  211.23.16.116  96.789 ms  99.233 ms  92.604 ms

(Test at 211.23.16.116)

1. ping 211.23.43.88  : replies only by eth0, inspected using tcpdump
2. ping 211.75.156.252: replies only by eth1, inspected using tcpdump
3. wget http://211.23.43.88/large.iso
   Traffic goes merely through eth0, so that load balance is not
   realised.


If you've read here, thanks a lot!
I'm totally confused by TEQL and Equal Cost. I've read all pages that
google could find by keyword 'TEQL', 2.4 Advanced Routing HOW-TO, and
LRP Load Balance HOW-TO. There are indeed a grate lacune of document
in this category...


-- 
原鄉人的血液沒有回到原鄉，是不會停止沸騰的。(鍾理和)
Fingerprint20 = CE32 D237 02C0 FE31 FEA9  B858 DE8F AE2D D810 F2D9

_______________________________________________
LARTC mailing list / LARTC@mailman.ds9a.nl
http://mailman.ds9a.nl/mailman/listinfo/lartc HOWTO: http://ds9a.nl/2.4Routing/

                 reply	other threads:[~2001-04-03  2:33 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=marc-lartc-98626523828589@msgid-missing \
    --to=ljm@ljm.qqjane.net \
    --cc=lartc@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body. 
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.