Faulty Intel Atom C2000 cpu, potential premature failure, Sangoma FreePBX 300, PBXact UC 300

hi,

this hit the newswire over the last week but seen no reference with it regard to Sangoma products, or any other PBX for that matter.

This looks like an intel problem and effects all companies using their C2000 atom cpu’s. i.e. it is not Sangoma’s fault. What will differentiate the suppliers is how they deal with it.

Unless the hardware manufacturer has addressed this with a “board level fix” then even the equipment you buy today will have this problem.

For details of the problem google “intel atom bug”. Essentially high failure rate from 18months into the life of the prouct.

Sources include intel’s shareholders quarterly conference call, i.e. profits warning.

Who has acted so far: Cisco, Supermicro, Synology among others…

Intels January 2017 C2000 spec update, table 1, page 10. AVR54
"System May Experience Inability to Boot or May Cease Operation"

This bug effect the revision B0, and the few cpu’s I’ve checked, only the revision B0 exists. i.e if you are using a C2000 atom cpu you have this bug unless it has been fixed by the board manufacturer in the recent months. (i.e. Supermicro say all boards since Jan 2017 are ok)

Check your cpu revision out here to be sure:
http://www.cpu-world.com/CPUs/Atom/TYPE-Atom%20C2000%20series.html
(also check on wikipedia)

Have a call in with my retailer. Awaiting response.

Also a quick chat with a sangoma US sales rep and he was not aware of any issues yet.

I would be really bummed if I bought one of these after this became public. Just really glad held of buying that spare pbx.

Do not be tempted to buy cheap Atom based kit on ebay either :slight_smile:

Thanks.


note this is all my personal opinions and I do not speak for my employer or for any other group.

For other vendor respensors:

From Cisco’s FAQ:
Q: Do you expect these products to fail at 18 months in operation?

Although the issue may occur beginning at 18 months in operation, based on information provided by the supplier, we don’t expect an unusual spike in failures until year three of runtime.

Most cpus’s of the C2000 were launched Q3 2013, so these issues are now only coming to light as systems hit 3yrs of operation. So far no idea what percentage of boards do fail.

A

Was there a question here? Have you contacted your reseller or distributor to see if this has been addressed? If you have bought any of these devices it would seem that is the appropriate route to go with any concerns. This post feels a bit like trolling. It may not be but an account with 3 posts comes around after not posting a while just to slam a product without any constructive feedback or attempt for clarification… When you hear hoof beats in the distance in montana you look for a horse not a zebra…

Our PBX 100 is not based on Intel Atom. Its a Quad Core Celeron not Atom based. Product Management is looking already on the impact this may have on our Atom based PBX 300 platform but we have seen no failures of Motherboards in the past 2 years on this platform and Sangoma has been using it with other Products for 3 plus years now but we will have a official statement up soon on this along with changes we are making to address the issue with Intel.

tony,

the freepbx appliance web page lists C2000 for FreePBX 100

And same info here
http://www.voipsupply.com/sangoma-freepbx-phone-system-100

Ahh ya that is not the case. We use the quad core Celeron as of last year. Someone missed updating that page in marketing

@jfinstrom

A quick google of “intel atom c2000 bug” will confirm the story. You will see I am not trolling.

Believe me, I wish this was not happening and would be very happy if I am wrong about this.

As stated in original post already have a call in with my supplier.

I’m a new owner of FreePBX 300 system. And new to the asterisk world. Appreciate all the help found in the community. Just trying to give back a little. Been happy with talking to sales folks from Sangoma, and wish them nothing but success.

From ServeTheHome website:

  • Cisco: Cisco has been extremely proactive on this. There is a full support page on the issue. They are replacing products for customers under warranty/ support contract as of November 16, 2016. My first tech job was working in a Cisco factory in a department working on refurbished boards. 20 years later I would expect that they will perform platform level fixes to current boards RMA’d.
  • Supermicro: RMA for platform-level workaround available for concerned customers. We also did confirm that Supermicro has implemented the platform level workaround in products shipped from January 2017 onwards.
  • Netgate: Netgate, the company behind pfSense firewalls, just (7 Feb 2017) posted information about a component issue and how they are handling replacements.
  • QCT confirmed switch control plane C2758 CPUs are impacted by a clock signal component issue. For example, in QCT’s Broadcom Trident II switches such as the bare metal switch version of our T3048-LY8 lab switches. Note the ARM variants like we are using are not impacted.
  • Synology: Company is working with a vendor on this issue.
  • ASRock Rack – Not seeing major issues from customers yet. Working with a vendor on this issue. Current course of action is to replace boards under three year warranty.
  • Others: Virtually every vendor uses Avoton/ Rangeley in some capacity. There is a good chance that if you are running a SDN switch with something like Cumulus Linux on it, you have a Rangeley chip inside. These things are everywhere. Most vendors cited NDA and declined to comment.

glad to hear it! Will edit the post title to reflect that.

For the FreePBX 300:
$ cat /proc/cpuinfo
…model name : Intel® Atom™ CPU C2758 @ 2.41GHz

PBXact UC 100 is fine too and should be removed from the title

will do, Thanks.

Great to hear Sangoma is talking to their suppliers.

Cisco are claiming failures occur 3yrs into the products life. (some other reports say after 18 months)

Details are few, but so far not seen any vendor claiming it does not effect their systems because of their board being designed like this, and others that do fail are designed like that. Seems NDA has them all very quiet about it.

Also no reports out there of what percentage of boards do actually fail.

Either way we all benefit from a clear statement.

Fortunately still have 5 months left on our warranty :slight_smile:

FYI this motherboard and CPU combo has been used at Sangoma in production since late 2013 so we have plenty of systems well over 3 years now with no abnormal failure rates.

Anyways I am going to lock this thread and as stated Product Management at Sangoma has been on this for weeks now since we got notice from Intel and suppliers but beyond that I can not comment.

1 Like