Configuring Two Serial Lines on Cisco 2610

[Augurer]

I'm attempting to enable a second serial line on our Cisco 2610, and am having difficulties allowing both the original T1 and the new T1 to work simultaneously.

I am unsure if the issue is with the configuration of the OpenBSD server, or if it is the router itself.

What do I need to do on the router, other than configuring the interface, to allow both Serial lines to feed the same Ethernet interface?

We currently have a static route of 0.0.0.0 0.0.0.0 Serial0/0 defined.

 

[JOAMON]

If both T1 lines are from the same provider I would contact them and see if they support bonding them into one 3meg line using multilink ppp.

 

[Augurer]

The T1 lines are from different providers, and aren't even close to being on the same sub-net.

 

[firecodex]

I don't think it would be possible to connect both to the same interface if the two T1's are from separate providers AND connect to separate subnets. You might try using subinterfaces, but I think you'll have to add another interface card (or in the case of the 2610, upgrade to a bigger router).

It seems that you're doing this because you've run out of interfaces, but if the lines connect to two different subnets then it doesn't seem that it would work, besides being sensible. The 2610 is end of life anyway, do yourself a favor and upgrade to a 2801 or 2811 and gain the support & features in addition to expansion capabilities. The T1 cards from your 2610 should work in the 2801/2811 as well, if you really want to go on the cheap.


------------------
Firecodex
CCNA/MCSA/MCP/A+/Net+

 

[Augurer]

I'm not really concerned with access the second T1 line from behind the router. All I want to do is allow packets coming into the router from the ethernet interface to be sent out both serial lines in a balanced fashion.

I'm believe the second T1 line is configured appropriately (the line and the protocol are up), however I'm not sure how to setup routing rules on the router to allow it to utilize both lines.

Currently, as I mentioned above, there is a single static route defined on the router as follows:

ip route 0.0.0.0 0.0.0.0 Serial0/0

I tried adding an identical route for Serial0/1, but traffic seems to stop flowing through the ethernet interface when I do that.

 

[plshlpme]

do you use bgp or another routing protocol?
is your internal addressing routable? or are you using nat?

and will both isps allow you to route your internal addressing if they are routable?

 

[Augurer]

I believe the only protocol being used is the static route I mentioned above.

If I run show ip bgp, show ip rip, show ip egp, etc all I receive is a blank line.

What do you mean by "internal addressing routable"? We have a router connected to an openBSD server using NAT. The openBSD server routes some of our external addresses to internal hosts. I do not believe either ISP has knowledge of our internal network.

 

[3wsparky]

just add another static route as above

0.0.0.0 0.0.0.0 Serial0/0 1
0.0.0.0 0.0.0.0 Serial0/1 1

the one at the end is a prefered route this will do a very basic form of round robin altho i think ip cef might cause this to prefer the first learnt route, any ideas guys ?

you can do a :no ip cef
but that means every packet will be examed to check for the best route which can be processor intensive

if you have other cisco boxes then i would suggest setting up a routing protocol so they can share routing infomation.

something like

router eigrp 1
network 172.30.*.*
network 192.168.1.*

hope this helps
Terry

 

[Augurer]

I tried running

ip route 0.0.0.0 0.0.0.0 Serial0/1

and external access is stopped completely when I do that. Is that not the proper way to add another static route?

 

[plshlpme]

youll have to post more of your config to ge more help...
i think there are to many unknowns still.

 

[3wsparky]

your missing the Priority from the end of the statement

the one after the interface number */* 1 <is a cost of the path

by adding that for both routes should round robin

 

[Augurer]

Here's the current config:

Building configuration...

Current configuration:
!
version 12.0
service timestamps debug uptime
service timestamps log uptime
no service password-encryption
!
hostname boygeorge
!
<PASSWORD INFO REMOVED>
!
ip subnet-zero
!
partition flash 2 8 8
!
gateway
!
interface Ethernet0/0
ip address QWEST.EXTERNAL.ADDY.HERE 255.255.255.224 secondary
ip address ATT.EXTERNAL.ADDY.HERE 255.255.255.224
no ip directed-broadcast
no ip mroute-cache
full-duplex
!
interface Serial0/0
description AT&T
ip address ATT.SERIAL.ADDY.HERE 255.255.255.252
no ip directed-broadcast
encapsulation ppp
no ip mroute-cache
no keepalive
service-module t1 timeslots 1-24
!
interface Serial0/1
description QWEST
ip address QEST.SERIAL.ADDY.HERE 255.255.255.252
no ip directed-broadcast
no keepalive
service-module t1 timeslots 1-24
!
ip classless
ip route 0.0.0.0 0.0.0.0 Serial0/0
no ip http server
!
<SNMP INFO REMOVED>
!
<LINE INFO REMOVED>
!
no scheduler allocate
end

As for the priority, I thought it was defaulted to 1, since it's a static route?

 

[LarryTheCucumber]

enable cef(Cisco Express forwarding) on the router.

http://www.cisco.com/univercd/cc/td/doc/product/software/ios121/121cgcr/switch_c/xcprt2/xcdcef.htm

I have the configuration you are discussing.

We send/receive to specific internal addresses via the
first T1 and use the other T1 for web.

We NAT to the two sub-nets at the firewall.

Work great.

 

[plshlpme]

so in this case though... are both T1's for internet access? or are they just links between two offices?

 

[Augurer]

In my case, both T1s are intended for the internet. I've tried enabling CEF with the 0.0.0.0 0.0.0.0 static route, and I know data is being sent over the second serial interface, but data is not being received on it.

 

[LarryTheCucumber]

Here is the main part of my config
!
version 12.2

ip subnet-zero
no ip source-route
ip cef
!
!
!
!
controller T1 0/0
framing esf
clock source line primary
linecode b8zs
channel-group 0 timeslots 1-24 speed 64
!
controller T1 0/1
framing esf
clock source internal
linecode b8zs
channel-group 0 timeslots 1-24 speed 64
!
!
!
interface FastEthernet0/0
ip address XXX.XXX.XXX.XXX 255.255.255.224
speed auto
full-duplex
!
interface Serial0/0:0
no ip address
encapsulation frame-relay
!
interface Serial0/0:0.1 point-to-point
ip address xxx.xxx.xxx.xxx 255.255.255.252
frame-relay interface-dlci 16
!
interface Serial0/1:0
no ip address
ip access-group 150 in
encapsulation frame-relay
!
interface Serial0/1:0.16 point-to-point
ip address xxx.xxx.xxx.xxx 255.255.255.252
frame-relay interface-dlci 16
!
ip classless
ip route 0.0.0.0 0.0.0.0 (Serial0/0:0.1 IP Address)
ip route 0.0.0.0 0.0.0.0 (Serial0/1:0.16 IP Address)
ip route (Serial0/1:0.16 IP subnet) 255.255.255.240 F/W IP


I then use NAT at the F/W to send/receive across the two lines.

 

[LarryTheCucumber]

Here are the stats on my interfaces

Serial0/1:0 is up, line protocol is up
Hardware is GT96K Serial
MTU 1500 bytes, BW 1536 Kbit, DLY 20000 usec,
reliability 255/255, txload 40/255, rxload 127/255
Encapsulation FRAME-RELAY, loopback not set
Keepalive set (10 sec)
LMI enq sent 1079720, LMI stat recvd 1079711, LMI upd recvd 0, DTE LMI up
LMI enq recvd 0, LMI stat sent 0, LMI upd sent 0
LMI DLCI 1023 LMI type is CISCO frame relay DTE
Broadcast queue 0/64, broadcasts sent/dropped 180024/0, interface broadcasts 95
Last input 00:00:02, output 00:00:02, output hang never
Last clearing of "show interface" counters 17w5d
Input queue: 0/75/0/0 (size/max/drops/flushes); Total output drops: 152247
Queueing strategy: weighted fair
Output queue: 0/1000/64/152247 (size/max total/threshold/drops)
Conversations 0/124/256 (active/max active/max total)
Reserved Conversations 0/0 (allocated/max allocated)
Available Bandwidth 1152 kilobits/sec
5 minute input rate 767000 bits/sec, 107 packets/sec
5 minute output rate 241000 bits/sec, 82 packets/sec
429852621 packets input, 62950396 bytes, 0 no buffer
Received 0 broadcasts, 0 runts, 1754 giants, 0 throttles
0 input errors, 0 CRC, 0 frame, 0 overrun, 0 ignored, 0 abort
386740555 packets output, 1236119345 bytes, 0 underruns
0 output errors, 0 collisions, 0 interface resets
0 output buffer failures, 0 output buffers swapped out
0 carrier transitions
Timeslot(s) Used:1-24, SCC: 4, Transmitter delay is 0 flags


sh int s0/0:0
Serial0/0:0 is up, line protocol is up
Hardware is GT96K Serial
MTU 1500 bytes, BW 1536 Kbit, DLY 20000 usec,
reliability 255/255, txload 42/255, rxload 19/255
Encapsulation FRAME-RELAY, loopback not set
Keepalive set (10 sec)
LMI enq sent 1079724, LMI stat recvd 1079723, LMI upd recvd 0, DTE LMI up
LMI enq recvd 0, LMI stat sent 0, LMI upd sent 0
LMI DLCI 1023 LMI type is CISCO frame relay DTE
Broadcast queue 0/64, broadcasts sent/dropped 180015/0, interface broadcasts 85
Last input 00:00:00, output 00:00:00, output hang never
Last clearing of "show interface" counters 17w5d
Input queue: 0/75/0/0 (size/max/drops/flushes); Total output drops: 85582
Queueing strategy: weighted fair
Output queue: 0/1000/64/85582 (size/max total/threshold/drops)
Conversations 0/70/256 (active/max active/max total)
Reserved Conversations 0/0 (allocated/max allocated)
Available Bandwidth 1152 kilobits/sec
5 minute input rate 119000 bits/sec, 44 packets/sec
5 minute output rate 253000 bits/sec, 63 packets/sec
381110768 packets input, 3749467971 bytes, 0 no buffer
Received 0 broadcasts, 0 runts, 1 giants, 0 throttles
0 input errors, 0 CRC, 0 frame, 0 overrun, 0 ignored, 0 abort
395863075 packets output, 2005505202 bytes, 0 underruns
0 output errors, 0 collisions, 0 interface resets
0 output buffer failures, 0 output buffers swapped out
0 carrier transitions
Timeslot(s) Used:1-24, SCC: 0, Transmitter delay is 0 flags

 

[plshlpme]

the way i see this is...
you hav a /30 going to both providers

your bsd server is doign nat for your internal clients
now what address is it natting to?

i think your passing traffic to your secondary link with an ip address that it will not route for you. or if your lucky and it is routing it to the world for you the advertised route from the world is going back in your other interface.

do you follow what im saying?

unless you are running bgp with both providers and they both advertise your address space equally i think your going to have asymetrical problems.

to do what your are trying to do i think youll need to implement nat on your router and overload out both interfaces with a routemap...

 

[Augurer]

We have two /27 addresses, but otherwise your synopsis is correct. The address the internal server is NAT-ing is just the AT&T one.

I did call QWEST in an effort to determine what the problem is, and they tell me the protocol is down. However, my router tells me both line and protocol are up, and it indicates that traffic is being sent equally out both lines.

I'm relatively certain that QWEST is not routing my traffic back to my AT&T line, however, as when I setup the route for the QWEST serial line (with CEF & by packet load sharing), internet access is slowed considerably with occasional timeouts.

 

[plshlpme]

i believe you would be better to use per destination load sharing... it is not a true load share.. but better for the data streams..

what you need to know though is if quest will route your at&t address space.. because as i said before if your natting onto at&t address space and then routing it out the qwest interface i suspect the traffic would be dropped.