Can a XPEC control superloops that are in diffrent network groups 2

[spivy122]

We have a CS1000m Multigroup(81c). We are currently running RLS 4.5 with Motorola 68040e processors and IGS. We are getting ready to upgrade to RLS 5.0, replace our Motorola Processors with Pentium Processors and replace the IGS with FNF.

Our vendor is telling us that we need to change our IPE shelfs around because we have Superloops from multiple network groups controlled by the same controller card(XPEC).

XPEC

S0 S1 S2 S3 LOC DIS RGTP

01 052 052 052 052 NO 08
02 036 020 004 004 NO 08
03 028 068 084 084 NO 08
05 060 060 044 044 NO 08
07 056 056 056 056 NO 08
08 100 100 104 104 NO 08
09 116 116 120 120 NO 08

Is this a true statement and if so can somebody please provide me a copy of the NTP or bulletin stating that it is required and why?

 

[bigindian65]

I would say yes.
You are pretty much segmented non-blocking, not totallly obviously, medium.

Your shelves are as such ( not you, Nortel, but you are included)
Group 0 shelf 0 0 thru 15
Group 0 shelf 1 16 thru 31
Group 1 shelf 0 32 thru 47
Group 1 shelf 1 48 thru 63

Etc, etc in multiples of 16 per shelf, BY multiples of 4
You get it? You see superloop 36 in group 1 but 4 & 20 in group 0 off same controller? Bad engineering.

Mato' Was'aka

 

[spivy122]

Bigindian thanks for the reply. I understand how to count loops and network groups. I get it...What I don't get is why is it Bad engineering?

My company won't spend money just because somebody said I think it is a bad way of setting it up. They require that we have documentation stating this is the required way or the way we have it configured is no longer supported.

Does anybody have documentation stating that you shouldn't configure the IPE shelfs the way we have?

 

[bigindian65]

sorry about the mini-class there. I am checking w/ a Nortel Engineer and a friend of mine that was a design engineer for them for many years, I'll let you know what they come back with.

Mato' Was'aka

 

[bigindian65]

This is from a current Nortel Solutions Engineer


"He's just creating additional traffic on the buses, I doubt you could find anything on it in the NTPs. But if he is doing an upgrade it would probably be a good time to straighten it out. Also if he runs into any flaky errors in the future, Nortel TAC would probably suggest adjusting it."

Mato' Was'aka

 

[steve64]

Nortel ETAS might even refuse even to work with you until it is corrected. I guess that whoever set it up to begin with didn't have an understanding of groups, shelves etc. Like bigindiana65 says if your doing an upgrade fix it at the same time you do the upgrade.

 

[spivy122]

All thanks for the help on this! Our vendor finally got a reply from Nortel. Our configuration is supported and no changes are needed. This was an issue when upgrading to FNF from IGS but Nortel resolved the issue with the new clock controller NTRB53AA. The product bulletin that addresses this is P-2002-1596-NA.

 

[bigindian65]

Had another Engineer come back with exactly what you found out, sorry I didn't post it.


That was an issue a long time ago but not anymore. I’ll try to double check tomorrow but off the top of my head their giving your friend old information. Also, I am absolutely surprised he’s a multi-group Succession 4.5 with that processor. The 68040e, if memory serves is not supported in this configuration. Minimum they should have the 68060e w/112Mb.

The original issue was with the clocking on the FIJI Network and that was resolved about 4 years ago. Either way I’ll look into it but I’m pretty damn sure it’s a non issue.


Mato' Was'aka