CISCO UBR IOS Caveats

[Sephiroth]

Have a look at this Cisco document:

http://www.cisco.com/en/US/docs/cable/cmts/release/notes/12_2sc/122sc_cmts_ub10k_caveats.html

This is more in the realms of the expertise held by BBings and xo, but for those of you wondering why VM can't fix their problems, look at this list of major software faults in the Cisco UBRs.

I don't know if this model of UBR is relevant to VM cable broadband - I suspect it is; but if these models are deployed with the listed IOS and the events concerned are occurring, many of the faults reported in the forums could be explained.

In that case, VM could simply be waiting on Cisco's fix.

This could be a red herring on my part - we'll soon know when this is pounced on!

[Welshchris]

Which fault are u reffuring to of the fix they are waiting for?

[Sephiroth]

Read the document and name your fault.

There were too many to name. Those of us who sort of know this stuff and those interpreting their event logs will see symptoms listed in the Cisco document as potentially relevant. It's a heck of a lot of combinatorial possibilities which is what you are likely to get when millions of CMs are connecting to hundreds of UBRs.

[Ignitionnet]

Talk about opening the floodgates :rofl:

[Welshchris]

i know the CISCO UBR i am on atm has a known issue where sometimes the model of UBR will stop people accessing websites but u can get onto anything else such as email, msn etc.

The only reason i know this is about a week ago when all the rubbish hit the fan and problems occurred all over the country i was still getting issues for 2 days after Swansea was apparently fixed and Second line support told me "there is a known issue with the CISCO UBR KIT UR CONNECTED ON THAT CAN CAUSE THIS FAULT!" 24hrs later it sorted itself out.

Its the Bloody Motorolla BSR faults that they need to sort out with the Ready light issue.

In Swansea Motorolla put one code out and didnt solve it and now they have to do a second one by December the 18th in the hope to cure it.

[Sephiroth]

@Welshchris

Yeah - but the 50Meg customers are a fraction of the problem and this thread isn't about the Motorola BSR and its problems.

[Welshchris]

Seph what im trying to say is, i think they will be putting their efforts into sorting the motorlla out over the CISCO as in the long term they will be doing away with the old kit and putting everyone onto the newer kit.

[KingDaveRa]

Didn't I read that they're pushing towards a Cisco infrastructure at the headends? Knowing how Cisco's stuff works, I wouldn't mind betting that means Cisco UBRs too...

Not saying Cisco kit is bad (I like it), but I've seen so many silly bugs in VPN, NAC, SCEs, Routers, Switches, WAPs, Firewalls, VoIP kit, to name but a few!

One thing Cisco DO suck at is writing client software for the desktop. Most of it is a shonkey bag of sh1te.

[Ignitionnet]

I don't know if this model of UBR is relevant to VM cable broadband - I suspect it is

Yep - a number of CMTS are relevant. The original 50M trials were done on the Cisco 10k and they were the main 50M CMTS before the BSR deployment. There are a fair few about the place and no doubt more to come.

EDIT: Mercifully the vast majority of these are not 'show stoppers' or are not relevant to the VM service or even are management issues not operational ones. No DOCSIS 2.0 upstreams removes further issues.

It looks like a horrendous list but like every other list of this type no operator will come even close to hitting all the caveats.

[xocemp]

Want to start with Malloc Failure and work up BB? :p:

#shut
#no shut

:D

Seph what im trying to say is, i think they will be putting their efforts into sorting the motorlla out over the CISCO as in the long term they will be doing away with the old kit and putting everyone onto the newer kit.

What makes you think all the new kit is Motorola?
That's a rather absurd statement since there is a grater amount of 10k chassis on the overlay than there is BSR/Moto.