26 Reply by jvalenzuela

  • jvalenzuela
  • Initech programmer
  • Offline
  • From: Orlando, FL
  • Registered: 2007-10-10
  • Posts: 223

Re: CM-1 not present

Ooo, since you have some ex-Cirrus folk, perhaps I can ask for some information that may help with the problems experienced in this thread.

Given the extreme flexibility available with advanced Cobranet configuration dialogs, I've never seen any documentation giving hard performance limits with regard to network traffic. For example I know I can transmit 4 8 channel bundles, I also know that I cannot transmit 16 2 channel bundles(or maybe I can, I've never tried). But I don't know anything about the middle area of that curve. It would be EXTREMELY useful to have some information like the maximum bytes/sec and frames/sec that a CM1/CM2 can tx/rx. Such information would leave little guess work trying to figure out if I've overloaded such and such a node.

27 Reply by Mike Dirks

  • Mike Dirks
  • Mike Dirks
  • Program Engineer
  • Offline
  • From: Satellite Beach, FL.
  • Registered: 2008-03-25
  • Posts: 11

Re: CM-1 not present

Wow, I should get on here more often.
We took another step the other day after I came home from vacation to a dozen messages about the system crashing again. It was reported that they let the in house guys get on the system to make a change. The system started up with the CM-1 error again and kept crashing until they loaded my back-up file 6 days later. The site has no idea what they changed but I am going to go out on limb and say it was cobranet smile ..... Instead of re-installing the the back-up system as planned we decided it would be better to add another NION to the system and re-distribute the cobranet. As Josh explained earlier, it wasn't doing any good to have the back-up anyway.
As far as the eithernet thing goes, It did help. We ran from 4-09-08 until 5-16-08 with no errors after separated the networks. 5-16-08 was when the inhouse guys made their change and it all went haywire.

I will try to get to some more of the questions in here later but I have to go for now. Thanks for all the posts on this. It has really been a help and I am learning alot.

Nice to meet you Steve.

Mike
Abacus Control Solutions

28 Reply by cobraguy

  • cobraguy
  • cobraguy
  • Peavey CobraNet Guru
  • Offline
  • Registered: 2008-05-28
  • Posts: 79

Re: CM-1 not present

Hi,

In response to Jason's last post about how to determine if a CM-1's resources are oversubscribed or not:
There are many things that can load down a CM-1, including other traffic on the net that the device must look at even though it doesn't care about it.
Cirrus developed a Python script a while ago that can be used to model a CM-1 configuration and then let you know if it the device's resources are oversubscribed. They have agreed to allow us to re-distribute this script in executable form, not the source. So a compiled executable of this script is now available.  The .ZIP file with all the necessary files as well as a .PDF describing the care and feeding of the program are available here:

http://downloads.peavey.com/mm/index.cf … e%20Device

Please note that this is for CM-1 only and also note that the Rev F boards and pre Rev F boards model a little differently. Read the PDF before using.

Nihilism is best done by professionals

29 Reply by cobraguy

  • cobraguy
  • cobraguy
  • Peavey CobraNet Guru
  • Offline
  • Registered: 2008-05-28
  • Posts: 79

Re: CM-1 not present

Mike,

Are you using any NIONs in XDAB connected clusters??
If so then how so?

Nihilism is best done by professionals

30 Reply by jvalenzuela

  • jvalenzuela
  • Initech programmer
  • Offline
  • From: Orlando, FL
  • Registered: 2007-10-10
  • Posts: 223

Re: CM-1 not present

Thanks for the link to the CM-1 cycle utility. That is exactly what I am looking for, what would be the chances of seeing similar tools for other types of Cobranet interfaces?

I have a quick question regarding the [ip packets] section. Are 'dropped packets' IP frames received by the CM-1 but not destined for it? If so, I would assume then that 'copied packets' are IP frames destined for the CM-1 and then copied to the host device? In the 'dropped packets' figure, should I also include non-IP frames received and ignored by the CM-1, STP for example. Are these quantities frames per second?

Finally are the CM-1 modules in the Nion products Rev F?

Last edited by jvalenzuela (2008-06-05 10:55:08)

31 Reply by Fergy

  • Fergy
  • MM Tech Support
  • Offline
  • From: The Home Office--Meridian
  • Registered: 2007-08-15
  • Posts: 334

Re: CM-1 not present

Rev F for the near future.  We haven't seen any of the new rev.

Make it intuitive, never leave them guessing.