Resultados 1 a 4 de 4
  1. #1
    WHT-BR Top Member
    Data de Ingresso
    Dec 2010
    Posts
    18,423

    Exclamation [EN] Atom C2000: falha irrecuperável após 18 meses de uso

    The failures start appearing after a unit has been in use for around 18 months.

    Once the processor fails, "the system will stop functioning, will not boot, and is not recoverable"

    Juha Saarinen
    Feb 7 2017


    A serious flaw with Intel's Atom C2000 product family can cause processors to fail completely, rendering the devices they power inoperable after just 18 months of operation.

    The low-power Atom C2000 Silvermont processor range was introduced three years ago, and is found in popular network switches and routers, microservers, and network accessible storage systems.

    Kit vendor Cisco has issued an advisory for the problem, noting the failures start appearing after a unit has been in use for around 18 months.

    Once the processor fails, "the system will stop functioning, will not boot, and is not recoverable".

    Cisco optical networking, routing, security, and switching gear - including the ASA and ISA3000 family - are affected.

    In its C2000 product family specification document [pdf], Intel states that the LPC clock outputs could stop working. If that happens, systems with the faulty processors will no longer be able to boot up.

    The LPC bus was introduced by Intel in 2002. It is four bits wide, runs at 33MHz, and connects relatively low speed parts of computer systems such as boot read only memory, and older devices such PS/2 keyboards and mice, to the processor.

    Intel's chief financial officer Bob Swan said the Atom C2000 processors experienced "slightly higher expected failure rates under certain use and time constraints" in the company's latest earnings call with analysts this year.

    Swan did not say how many processor failures Intel has recorded and how much rectifying the issue will cost the company, but admitted it would weigh on Intel's data centre group margins this year.

    He said Intel has the problem "relatively well-bounded with a minor design fix that we're working with our clients to resolve".

    The Atom C2000 chip is used in products from several other large hardware makers, including Dell, HP, NEC, Qanta, Supermicro, and Synology, but only Cisco so far is warning customers about the potential of processor failure.

    https://www.itnews.com.au/news/dying...uipment-450238

  2. #2
    WHT-BR Top Member
    Data de Ingresso
    Dec 2010
    Posts
    18,423

    Exclamation Online.net: Component erratum potentially affecting Dedibox SC/XC and Scaleway C2/VC1

    Edouard Bonlieu
    9th of February 2017

    At Online and Scaleway we deliver robust and powerful servers. Our goal is to offer servers at the best price without compromise in quality and by providing an excellent user experience.

    Three years ago, we started designing our own server internally in the Online labs. Everything from the PCB to the air flow is designed and created internally, to provide extremely affordable and high performance servers. The hardware is built in our factory localized in France, following a high quality process certified ISO 9001 and ISO 14001.

    We've been recently alerted by one of our suppliers of an erratum concerning a component used in some of our servers (Dedibox SC 2016, Dedibox SC 2015, Dedibox XC 2016, Dedibox XC 2015, Scaleway C2S, C2M, C2L, VC1S, VC1M, VC1L). This impacts the component by reducing its lifetime at an accelerated rate. As of now we're not seeing any occurrences for this erratum in our datacenters across 67855 servers deployed.

    From the information we get from our supplier, this component can fail under certain use cases and time constraints, causing the server to stop working and not able to boot anymore.

    If the server hardware fails in any way, our users shouldn't be impacted too much as our value proposition consists in replacing all faulty hardware. We're taking this situation very seriously and are working with this supplier on a strategy to move forward in the most transparent way.

    FAQ

    What is the source of this erratum?

    Online stands by its manufacturing quality standard and customer advocacy for more than 17 years that contributes to the company's reputation. We'd like to be more open but this erratum is related to a component protected by NDAs, and whose communication is deliberately restricted by the manufacturer. We are not in a position to communicate as we would like. We apologize about that.

    What are your solutions to address this issue?

    Our electronics laboratory is currently evaluating the impact related to this erratum. We will do our best to maintain the level of service and keep our high quality requirements.

    Did you already observe any occurrences for this erratum?

    We are running this component for 3 years and have a significant amount of servers in production using it. Our statistics, including the oldest servers, do not show significant rates of failure but we remain extremely vigilant.

    As of today, we do not know if we will be impacted by this component erratum.


    https://blog.online.net/2017/02/09/c...aleway-c2-vc1/

  3. #3
    WHT-BR Top Member
    Data de Ingresso
    Dec 2010
    Posts
    18,423

    Intel's Atom C2000 chips are bricking products – and it's not just Cisco hit

    Thomas Claburn
    6 Feb 2017

    Intel's Atom C2000 processor family has a fault that effectively bricks devices, costing the company a significant amount of money to correct. But the semiconductor giant won't disclose precisely how many chips are affected nor which products are at risk.

    On its Q4 2016 earnings call earlier this month, chief financial officer Robert Swan said a product issue limited profitability during the quarter, forcing the biz to set aside a pot of cash to deal with the problem.

    "We were observing a product quality issue in the fourth quarter with slightly higher expected failure rates under certain use and time constraints, and we established a reserve to deal with that," he said. "We think we have it relatively well-bounded with a minor design fix that we're working with our clients to resolve."

    Coincidentally, Cisco last week issued an advisory warning that several of its routing, optical networking, security and switch products sold prior to November 16, 2016 contain a faulty clock component that is likely to fail at an accelerated rate after 18 months of operation.

    Cisco at the time declined to name the supplier of that component. When asked on Monday whether Intel supplied the faulty electronics, a Cisco spokesperson told that The Register that the networking giant does not intend to publicly name the supplier.

    Intel indicated in a January 2017 revision of its Atom C2000 family documentation that the chip line contains a clock flaw. Errata note AVR.54, titled "System May Experience Inability to Boot or May Cease Operation," explains that the Atom C2000 Low Pin Count bus clock outputs (LPC_CLKOUT0 and LPC_CLKOUT1) may stop functioning. Permanently.

    An Intel spokesperson in an email to The Register characterized the issue as "a degradation of a circuit element under high use conditions at a rate higher than Intel’s quality goals after multiple years of service."

    "If the LPC clock(s) stop functioning the system will no longer be able to boot," Intel's documentation explains.

    This consequence is precisely what Cisco says may happen to its devices given enough time. "Once the component has failed, the system will stop functioning, will not boot, and is not recoverable," Cisco's advisory states.

    The Register asked Intel whether it could confirm that Cisco's advisory could be attributed to an Intel component. Intel said it could not confirm or deny whether its chip issue is the one affecting Cisco gear, citing a policy of not commenting on customers. We note that the affected Cisco ASA 55xx products use Intel's Atom C2000 system-on-chips at least.
    Radio silence

    We asked Intel to provide specific details about when it began and stopped shipping Intel Atom C2000 processors with faulty clock outputs. Intel declined to comment. The official errata says the B0 stepping of C2xxx Atoms are vulnerable to failure, and these parts began shipping in 2013. The specific SKUs are:

    C2308, C2338, C2350, C2358, C2508, C2518, C2530, C2538, C2550, C2558, C2718, C2730, C2738, C2750, and C2758.

    We asked Intel how many affected Atom C2000 chips have been shipped and how much fixing the issue will cost the company. Intel declined to comment.

    Intel did, however, provide some insight on how the Atom C2000 flaw might be addressed. "A board level workaround exists for the existing production stepping of the product which resolves the issue," a company spokesperson said in an email. "Additionally, Intel will implement and validate a minor silicon fix in a new product stepping that resolves this issue."

    Many other technology vendors make products with Intel Atom C2000 processors, including Dell and Synology. The Register pinged Dell via email, and it was not immediately available for comment.

    People with Synology DS1815+ storage boxes have been reporting complete hardware failures; the DS1815+ is powered by an Intel Atom C2538.

    Other vendors using Atom C2000 chips include Asrock, Aaeon, HP, Infortrend, Lanner, NEC, Newisys, Netgate, Netgear, Quanta, Supermicro, and ZNYX Networks. The chipset is aimed at networking devices, storage systems, and microserver workloads. If you know of any affected or failed gear, please let us know.

    According to this Intel data-sheet [PDF], LPC_CLKOUT0 and LPC_CLKOUT1 are driven by the processor to provide essential timing signals to hardware on the board, including the boot ROM. If these signals stop ticking, the rest of the electronics stops, too.

    Updated to add at 14:29 UTC, February 7

    Synology told The Reg it was "investigating this issue together with Intel," saying it would "provide more detailed information in the week as it becomes available."

    It added: "At this time Synology has not seen any indication that this issue has caused an increase in failure rates for DiskStation or RackStation models equipped with Intel Atom C2000 series processors. However once additional information is available, Synology will post an advisory on this topic.

    "In the meantime it is still safe to continue to use your device, however should you encounter any issues these will be handled via our support teams and all the devices are still covered under the standard warranty provided with all our units. Technical Support can be reached via www.synology.com/ticket."

    Updated to add at 15:36 UTC, February 7

    Synology subsequently asked to have its statement above revised to omit any mention of Intel.

    Updated to add at 18:23 UTC, February 8

    Once again, Synology has been in touch, seemingly now able to use the I word, to say: "Intel has recently notified Synology regarding the issue of the processor’s increased degradation chance of a specific component after heavy, prolonged usage.

    "Synology has not currently seen any indication that this issue has caused an increase in failure rates for DiskStation or RackStation models equipped with Intel Atom C2000 series processors compared to other models manufactured in the same time frame not equipped with the affected processors."

    https://www.theregister.co.uk/2017/0...o_faulty_chip/

  4. #4
    WHT-BR Top Member
    Data de Ingresso
    Dec 2010
    Posts
    18,423

    HPE joins Cisco, Juniper with faulty clock technology problem

    Michael Coone
    Feb 17, 2017

    Hewlett Packard Enterprise is the latest vendor to identify a faulty clocking component of its products that can cause them to crash and not recover.

    HPE joined Cisco and Juniper in identifying the problem, even going so far as to tap the widely-suspected Intel-based clock element as the cause.

    In a statement the company said:

    “To the best of our knowledge, our customers are not experiencing failures due to the Intel C2000 chip, which is deployed on a limited number of our products. We remain committed to assuring the highest quality experience from our solutions and are proactively working with Intel to mitigate any future risk and impact on our customers.”

    Unlike Cisco and Juniper however, HPE did not identify its affected products nor offer any details as to what customers can do with them should the problem occur.

    Cisco said when it first identified the clock problem earlier this month it would “support our customers and partners, Cisco will proactively provide replacement products under warranty or covered by any valid services contract dated as of November 16, 2016, which have this component.”

    Cisco also said it had set aside $125 million to help cover the cost of replacing networking gear impacted by a fatal clock component. Cisco CFO Kelly Kramer told analysts at this week’s second quarter earnings announcement: “We have had an issue from a supplier come out, and we did book a reserve for $125 million, you can see in our [generally accepted accounting principles (GAAP)] results and in the press release, to cover that. We always, and continue to stand by our customers through any situations like this. This is very proactive. This is a failure rate that will happen over time, but we're working with our customers to work through that. So we're not anticipating any impact from that from a top line perspective…Again, we're working very proactively with our customers and in terms of how quickly and where they want to do their replacement. So we're working very, very closely, but as of right now we have not seen and don't anticipate any massive revenue impact from this.”

    Juniper meanwhile said this week that it “is aware of an issue related to a component manufactured by a supplier which impacts a limited set of our product line. We are currently working directly with any impacted customers on a swift solution.”

    Neither Cisco nor Juniper have been willing to pinpoint the killer clock signaling component-maker but the problems coincide with difficulties described by Intel on its Atom C2000 chip that is used by a number of hardware makers.

    Other vendors such as Synology a network attached storage vendor in Taiwan have been identified as using the clock technology as well. Dell has not commented on the issue so far.

    The IDG News Service recently wrote of the Atom’s troubles, reporting in January that Intel added an erratum to the Atom C2000 documentation stating systems with the chip "may experience [an] inability to boot or may cease operation."

    The chip is the last among Intel's line of short-lived low-power Atom chips for servers. It was used in microservers but also networking equipment from companies like Cisco, which has issued an advisory about a product defect related to a component degrading clock signals over time. A clock signal degrade hurts the ability of the chip to carry out tasks. Intel is trying to fix the issue but declined to comment on when it'll deliver an update, the IDG story stated.

    "There's a board level workaround that we are sharing with customers now," an Intel spokesman said in an email to the IDG News Service "Additionally, we are implementing and validating a minor silicon fix in a new product [update]."

    http://www.networkworld.com/article/...y-problem.html

Permissões de Postagem

  • Você não pode iniciar novos tópicos
  • Você não pode enviar respostas
  • Você não pode enviar anexos
  • Você não pode editar suas mensagens
  •