BT Openreach VDSL MCT Certification - CPE Pass List

[jawicks]

The following devices have passed the OpenReach MCT testing process or have been given a pass status based on their similarity to other models which have already achieved a pass (indicated with "=" in the pass-date column below). This table will be updated as more devices are passed through the MCT certification process.

Device P/N	IOS Ver	DSL Chipset	Chipset H/W	Chipset F/W	Pass Date	Comment
C887VA-K9	15.4(2)T1	Broadcom	BCM6368	39m_B_38h3_24h	Oct 2014	Baseline 887 unit
C897VA-K9	15.4(2)T1	Broadcom	BCM6368	39m_B_38h3_24h	Oct 2014	Baseline 897 unit
CISCO887VA-K9	=	Broadcom	BCM6368	39m_B_38h3_24h	=	original generation 887
CISCO887VA-M-K9	=	Broadcom	BCM6368	39m_B_38h3_24h	=	original generation 887 (Annex-M)
CISCO887VA-SEC-K9	=	Broadcom	BCM6368	39m_B_38h3_24h	=	original 887 with IPsec s/w bundle
C887VAM-K9	=	Broadcom	BCM6368	39m_B_38h3_24h	=	baseline 887VA (with Annex-M)
C887VA-W-E-K9	=	Broadcom	BCM6368	39m_B_38h3_24h	=	baseline 887VA with integrated WiFi
C887VAM-W-E-K9	=	Broadcom	BCM6368	39m_B_38h3_24h	=	baseline 887VA (Annex-M) + WiFi
C887VA-V-K9	=	Broadcom	BCM6368	39m_B_38h3_24h	=	baseline 887VA with integrated FXS POTS
C887VA-V-W-E-K9	=	Broadcom	BCM6368	39m_B_38h3_24h	=	C887VA-V-K9 + WiFi
C887VA-CUBE-K9	=	Broadcom	BCM6368	39m_B_38h3_24h	=	C887VA with CUBE s/w license bundle
C887VAG+7-K9	=	Broadcom	BCM6368	39m_B_38h3_24h	=	C887VA with integrated 3G
C887VAMG+7-K9	=	Broadcom	BCM6368	39m_B_38h3_24h	=	C887VAM-K9 with integrated 3G
C887VAGW+7-E-K9	=	Broadcom	BCM6368	39m_B_38h3_24h	=	C887VAMG+7-K9 with integrated WiFi
C887VA-WD-E-K9	=	Broadcom	BCM6368	39m_B_38h3_24h	=	C887VA with dual-band WiFi
C887VAG-4G-GA-K9	=	Broadcom	BCM6368	39m_B_38h3_24h	=	C887VA with integrated 3G/4G
C897VA-M-K9	=	Broadcom	BCM6368	39m_B_38h3_24h	=	baseline C897VA (Annex-M)
C897VAW-E-K9	=	Broadcom	BCM6368	39m_B_38h3_24h	=	C897VA-K9 with integrated WiFi
C897VAM-W-E-K9	=	Broadcom	BCM6368	39m_B_38h3_24h	=	C897VA-M-K9 with integrated WiFi
C897VAG-LTE-GA-k9	=	Broadcom	BCM6368	39m_B_38h3_24h	=	C897VA-K9 with integrated 3G/4G
C897VAMG-LTE-GA-K9	=	Broadcom	BCM6368	39m_B_38h3_24h	=	C897VA-M-K9 with integrated 3G/4G
NIM-VAB-A	XE3.14.0	Broadcom	BCM63168	39p1	July 2015	NIM Module for ISR4xxx - conditional pass, NIM fails OAM loopback testing which may impact on-line diagnostics.
EHWIC-VA-DSL-A	15.4(2)T1	Broadcom	BCM6368	39m_B_38h3_24h	Sept 2015	EHWIC plugin for ISR-G2 routers; c1900, c2900, c3900, etc.
EHWIC-VA-DSL-A	15.6(3)M	Broadcom	BCM6368	A2pv6C039t	Jan 2016	Embedded FW and SDK Change with 15.6(3)M requires version 39t and it is not backward compatible with previous versions.
EHWIC-VA-DSL-M	15.4(2)T1	Broadcom	BCM6368	39m_B_38h3_24h	Sept 2015	EHWIC plugin for ISR-G2 routers; c1900, c2900, c3900, etc.
C1113-8P	16.9	Broadcom	BCM63138	A2pvfbH043n	Nov 2018	VDSL only. G.FAST is going through certification.
C927-4P	15.8	Broadcom	BCM63168	A2pv6F039x8	June 2019
C1117-4P	16.9	Broadcom	BCM63168	A2pvbF039t.d26d	June 2019	IOS-XE versions 16.6-16.10
C1117-4P	16.11	Broadcom	BCM63168	A2pv6F039x8	June 2019	IOS-XE version 16.11 onwards
C1117	17.4	Broadcom	A2pv6F039x8.d26d	4.14L.04	November 2021
C1127	17.9.3	Broadcom	A2pv6F039x8.d26d	4.14L.04	July 2023

[tomgrace1]

Can somebody confirm the 867VAE has also passed this? I believe the chipset is the same as the 887VA.

[jawicks]

Hi Tom, the devices with part-numbers beginning C867VAE do not have the same DSL chipset as those which are on the list of devices equivalent to c887VA-K9 and c897VA-K9 and so are not certified for use.  The CISCO867VAE-K9 does have the same chipset but I have found these units are pretty under-powered for use on higher-speed DSL lines and therefore I am not currently planning to push for these models to be certified by OpenReach.

Certifying the newer ISR hardware models shipping plus preparing for those new ones in the pipeline are my priority right at the moment.

Rgds, Jim.W.

[tomgrace1]

Jim

Many thanks for your prompt response on this, it is very helpful. We tend to really only look at the 867VAE-K9 for up to 40Mb/s services, i appreciate they may be limited for the higher speeds. If these are unlikely to be approved anyway then i suppose they are not really an option on native VDSL services moving forward.

What is the best resource i can use to keep up to date with the approved models, is it here?

Regards

Tom

[gareth.bryan]

Hi Jim,

In many situations Openreach install cabs in areas where traditional DSL services are unable to attain even 1-2mbps,

so whilst you may consider them underpowered they still firmly have a place in this market and I urge you to reconsider submitting them for MCT testing.

From personal experience the 6368 is great SoC and should have no real world issues hitting 38mbps/8k pps.

Really (depending on if/how smp and broadcom hw acceleration was implemented on the platform) much more.

Not such an issue on the other 8xx/ehwic ranges as the 6368 is just bridged (was surprised to see the modems still run linux though!)

Many thanks,

[jawicks]

Hi Tom, for the BT specific certifications then yes I am aiming to keep this site as the central point for updates when new kit undergoes certification or previous devices require changes such as Firmware updates.

As for the c867 then seeing as there has been a couple of requests now for this to undergo testing then I will try to get this onto the schedule, though as I mentioned above my priority is with the newer hardware SKUs. The MCT test-slots from OpenReach are in pretty short supply therefore c867 will be submitted on a "best-efforts basis" if I am able to wrestle another testing window off OpenReach AND I don't have any backlog of other devices which need to go through the process.

So in summary, I will try to get c867 through, but due to processes involved, don't hold you breath

Rgds, Jim.W.

[jawicks]

Hi Gareth, I just replied to one of the other contributors (I'm not sure what view you have of all comments/replies so apologies for repetition) to say that based upon the requests for c867 then I will try to get the device through but it won't be at the expense of delaying the newer hardware SKU's which I have already pencilled in MCT dates for off OpenReach therefore I would be expecting 6months until I can get it scheduled into the front end of the process, assuming a pass first time.  If it's a lot of engineering work required to align the c867 with what we already have certified (c887, c897, etc) then we might need to think again.

For the throughput, etc of the 6368 then the Broadcom chipset will have no issues pushing the required bandwidth (works fine on c887VA and c897VA with that chip installed), but the actual grunt of the processor of the c867 is where the bottle-neck would potentially lie if deploying with services layered on top of the basic DSL (IPsec, etc).

Rgds, Jim.W.

[gareth.bryan]

Thanks for the speedy reply Jim.

(Yep I could see your reply to Tom above)

Really appreciate you trying to get it on board, it really is a great little router that fits into our portfolio very nicely so would be glad to see it on the list!

I see that the dsl firmware "blob" for the 867vae is inline with it's bigger brothers so fingers crossed.

I had noticed that the 867vae was using the 6368 cpu natively rather than running it standalone and bridging to the platform (as is with 8xx and ehwic). I would expect dual 400mhz viper cores *should* get it close to the 887va for control plane stuff, with regard to nat/crypto/acl/small packet acceleration that would all come down to how/if Cisco managed to integrate with the broadcom "cmf" voodoo.

Anywho, I digress - thanks again and fingers crossed!

[r.pennington]

Hi James, do you have the tested reference configurations for the 897VA-K9 and the 1921 with the EHWIC-VA-DSL-A ?

[jawicks]

Sorry Richard, my bad, I have been meaning to post something on here for a while now and never gotten around to it.  I will try to post the full config document this week if I get a moment.

The most important is to reference the correct Firmware version under the DSL controller configuration if it is not included in the embedded image in IOS version.  Then a few tweaks to the default modem settings as shown below.

controller VDSL 0

   operating mode vdsl2

   training log filename flash:897VA.log

   firmware filename flash:VA_A_39m_B_38h3_24h.bin

   modem disableV43          <-- be careful, this one is case-sensitive

   sra

Rgds, Jim.W.

[r.pennington]

Great thanks for that, if you have the full configs that would be really useful, I'll check back next week.

[jawicks]

draft version of config document has been posted at the link below, all comments / suggestions for changes welcome.

Connecting Cisco ISR routers to BT GEA

[jawicks]

Note:- Cisco IOS 15.6(3)M and later requires Firmware version A2pv6C039t as a minimum to work properly due to some changes in the interfaces between IOS and the Modem Firmware. The wording in the release notes at the link below are not clear on this point (blind copied from earlier Firmware versions) and so I have requested the tech-writers to clarify that in the docs.

Release Notes for VDSL CPE PHY Release A2pv6C039t - Cisco

The 39t Firmware version is progressing through testing in BT OpenReach conformance labs and should exit by end of September so I will update the support matrix assuming we get a green-light off the test teams.

[jawicks]

Chaps, the new C867VAE devices will be tested via the BT-OpenReach MCT process which are the cut-price lower spec variants of our CPE.

[jawicks]

Note. The ISR800 and ISR G2 which run the IOS release 15M/T is integrated with the DSL Firmware 35J for DSL Annex(es) A/M and 35J for Annex B/J. With the IOS release 15.6(3)M, we plan to change the embedded BRCM Firmware in IOS to 39Tfor Annex A/M and 38r1 for Annex B/J. This change will affect only the 880/890 and ISR G2+EHWIC SKUs which carry the BRCM6368 chip-set. More importantly, we also plan to change the current SDK version used for bootloader and embedded linux. This is the SDK used to compile the currently available FW. This SDK being a very old version,. Hence we plan to upgrade the SDK for the bootloader and embedded linux to the latest version 4.14L.04A . This will need customers to us either the embedded 39T FW version or later, compiled with the newer SDK if they plan to migrate IOS 15.6(3)M version or later. Customers who are on the existing IOS release 15.6(2)T version or earlier ( for ex. 15.4(3)M) can continue to use the earlier FW version(s) available.

The Firmware release notes are available at the link below

Release Notes for VDSL CPE PHY Release A2pv6C039t - Cisco

Previous version release notes

Release Notes for VDSL CPE PHY Release A2pv6C039m - Cisco

Q) Will this change impact Annex B/J systems like (C886VA-K9 or C896VA-K9)?

A) Yes. This change will be effective for both Annex A/M and Annex B/J based systems.

Q) Will this change impact ADSL/VDSL NIM for ISR4000 series

A) No. This change is only effective for the BRCM6368 based ADSL/VDSL interface or EHWIC on the 800 series/ISR G2. The ADSL/VDSL NIM has BCM63168 chipset

Q) Is it possible to run the existing FW that we use in customer environment with the IOS 15.6(3)M release?

A) No. The existing FW available on CCO has been compiled with the older SDK version 4.02L.03. For the FW to work with 15.6(3)M IOS release we will need to release a newer FW image on CCO compiled with the current SDK 4.14L.04A .

Q) Is it possible to use the 39T FW with the older IOS releases like 15.5(3)M release?

A) Yes. This FW is posted on CCO and available as an selectable option on CCW ordering tool.

[jmakepeac]

Hi

Has the NIM-VAB-M been approved by Openreach? This would let us use the same modem everywhere; for ADSL and VDSL and have the option of order Annex M on ADSL2+.

Also, is there likely to be any movement on the caveat for the NIM-VAB-A?

John