Easy way to implement ppp multilink in Central PRI / Remote BRI

[roesch4alc]

Hi all,

currently we´re running a backup conecpt, where the central router dials into the branches, when their dsl connection went down. Its working in general. The connection in the branches is a BRI Interface, so actually they will be connected with 64k in the backup case. We would like them to be connected with 128k. The central connection is 2x E1. So were working with the serial0/0:15 and serial1/0:15 interfaces.

To get ppp multilink working, I configuried the "ppp multilink" and also "ppp multilink links minimum 2" command on interface Serial0/0:15 on the central site and on interface BRI0 on the branch router. I also got the information, that the routers tried to bring up the second link:

Oct 13 13:32:59.740 MEST: Vi1: MLP: Bundle has 1/2 desired links, requesting another

Output show ppp multi:

Virtual-Access1, bundle name is routername
  Bundle up for 00:10:03
  Dialer interface is Serial1/0:15
  0 lost fragments, 0 reordered, 0 unassigned
  0 discarded, 0 lost received, 3/255 load
  0x0 received sequence, 0x0 sent sequence
  Member links: 1 (max not set, min 2)
    Se1/0:30, since 00:10:03, no frags rcvd

We are using following configuration:

Central.Router:

interface Serial0/0:15
 bandwidth 64
 ip address 10.100.17.1 255.255.255.0
 ip access-group 102 in
 no ip unreachables
 no ip proxy-arp
 ip accounting output-packets
 encapsulation ppp
 ip summary-address eigrp 1 0.0.0.0 0.0.0.0 10
 ip summary-address eigrp 2 0.0.0.0 0.0.0.0 10

dialer map ip 10.100.17.39 name routername broadcast 123
 dialer idle-timeout 310 dialer-group 1
 isdn switch-type primary-net5
 isdn caller 123
 isdn caller 456
 no peer default ip address
 no fair-queue
 no cdp enable
 ppp authentication chap

Branch Router

interface BRI0
 ip address 10.100.17.39 255.255.255.0
 no ip redirects
 no ip unreachables
 no ip proxy-arp
 ip flow ingress
 encapsulation ppp
 dialer idle-timeout 72000
 dialer enable-timeout 1
 dialer map ip 10.100.17.1 name rtr-backup broadcast 123
 dialer hold-queue 100 timeout 30
 isdn switch-type basic-net3
 isdn point-to-point-setup
 isdn caller 123
 no cdp enable
 ppp authentication chap

Traffic flow is created with ip sla or rtr entries.

Does somebody know, whats wrong?

Best Regards

[trfinkenstadt]

Hi,

It's been quite a while since i've configured dialup connectivity.  Is there a requirement for the second channel to only come up when bandwidth utilization reaches a certain threshold?  Could that be the issue?  Do your edge sites utilize more than 64kbits when on the backup?

[roesch4alc]

Hi,

as you can see in the config above I´ve also tried to use the command "dialer load-threshold" I also used it with a value of 1. But in my case the central site wasn´t able to bring up the second channel. This log from the central site also says, that it tried to bring it up: Vi1: MLP: Bundle has 1/2 desired links, requesting another

Do your edge sites utilize more than 64kbits when on the backup?

-->As I wrote, I´m not able to get the second channel running, it´s only 64K.

I even took a look at this, but that seemed to be much to complex for me...

http://www.cisco.com/c/en/us/td/docs/ios/qos/configuration/guide/15_1/qos_15_1_book/mlppp_over_serial.html

I must state, that for a moment I had two BRI channels running, but I lost my ip connectivity to the branch in that moment...And after that I´ve undone my settings.