[nsp] Problems with half bridge PVC

[Dan Armstrong]

I see the command "mtu 1500" in your config.

We half bridge like madmen on a GSR, and the config is _almost_ the same, but " ip
mtu 1500" instead, don't know if that makes a difference.


Alastair Johnson wrote:

> Hi,
>
> Since migration from one POP to another last week - which
> included a switch to another router - I've been having some
> problems with a half bridge ATM PVC.
>
> Basically, for the most part, traffic flows fine across the
> circuit, including 1500b (and larger) pings.
>
> However, I can't telnet to the device on the remote site -
> I get to the cisco Username: prompt and can't type anything.
> I can however press return and get a new username prompt.
>
> Enabling telnet line mode does work around this problem
> so I can still get into the remote device.
>
> Simultaneously, customers on the other end of this circuit
> have problems sending emails with attachments, or browsing
> SSL websites.
>
> config:
>
> cr01.akl1 - 7507, 12.0(26)S, ATM side
>
> interface ATM4/0/0.266 multipoint
>  description Router: CR01.HLZ1 via 2M TCL ATM (IVC407-298)
>  mtu 1500
>  bandwidth 2048
>  ip address 202.89.32.161 255.255.255.252
>  no ip directed-broadcast
>  no ip proxy-arp
>  ip route-cache same-interface
>  ip ospf network broadcast
>  ip policy route-map netguardian
>  atm pvc 266 0 266 aal5snap bridge 2261 2261 oam 30
>  no atm enable-ilmi-trap
> end
>
> cr01.hlz1: cisco 2620, IOS 12.2(16)
>
> !
> interface FastEthernet0/0.266
>  description Router: 2Mbps PVC to CR01.AKL1 via TelstraCLEAR IPExpress (IVC407298)
>  bandwidth 2048
>  encapsulation dot1Q 266
>  ip address 202.89.32.162 255.255.255.252
>  ip wccp web-cache redirect out
>  ip ospf network broadcast
> end
>
> The telco delievrs it as a 2261k ABR (originally - they've tried
> switching it to vbr-nrt to see if that helped.. it didn't) PVC
> to cr01.akl1.  The Ethernet side is delivered as a 10M/Full duplex
> (duplex confirmed with telco) circuit, tagged on our switch as
> 266 and passed via the dot1q trunk to the 2620.
>
> Telco has confirmed the rate-shaping is 2261k, and they've
> confirmed and added the shaping on their switch at the Ethernet
> side.
>
> This all used to work when it was connected to a different 7507
> using 12.2(15)T with RBE.  Obviously RBE isn't available in 12.0S
> so we can't use that.
>
> Anyone have any ideas?  I'm completely stumped, as is the telco.
> The problem is that when we shifted from one router to the other,
> they also rebuilt the PVC from one ATM bearer to another, so
> they're not sure if *they* broke something or if it's us.
>
> The 7507 has another circuit that is identical in all ways except
> it is 10Mb, which has no problems.
>
> any help would be greatly appreciated.
>
> aj
>
> --
> Network Operations              ||      noc. +64.9.915.1825
> Maxnet                          ||      cell. +64.21.639.706
> _______________________________________________
> cisco-nsp mailing list  cisco-nsp at puck.nether.net
> https://puck.nether.net/mailman/listinfo/cisco-nsp
> archive at http://puck.nether.net/pipermail/cisco-nsp/