Light Reading Mobile – Telecom News, Analysis, Events, and Research
Networking the Telecom Industry
Sign up for our Free Telecom Weekly Newsletter
Connect with us
Comments
Current display:       Newest Comments First       Display in Chronological Order
yike
User Ranking
Friday October 19, 2012 1:27:21 AM
no ratings

I would like to have the list of the happended issues for vendors to beat each other ... will that be helpful ?

http://www.lightreading.com/blog.asp?blog_sectionid=414&doc_id=223187

  • Orange France , July 2012: A major Friday evening service outage that affected about 26 million customers followed a software upgrade to Alcatel-Lucent (NYSE: ALU) HLR software carried out by staff from the vendor and the carrier. The upgrade introduced data inconsistencies that led to a rising tide of inconsistent messages being shared by Orange France's subscriber systems, which eventually became overloaded and unable to process new requests. The operator provided details of why and how the outage happened in testimony to the country's National Assembly: France takes these issues seriously.
  • Telenor ASA (Nasdaq: TELN), June 2011: An upgrade to a mobile broadband-related server resulted in the "most extensive breakdown that Telenor has experienced since the mobile network was established in 1993." Signaling traffic between various servers escalated and then "created disturbances" in other network elements, affecting the availability of voice and text message services to 3 million users. (See Telenor Explains Mobile Outage.)
  • Verizon Wireless , April 2011: The U.S. operator suffered problems with its 4G/LTE service that were linked to a Nokia Siemens Networks HSS. Verizon has subsequently suffered a number of interruptions to its 4G services, but the problems appear to have been caused by a different network issue each time. (See Verizon: LTE Is Back and Euronews: April 21.)
  • T-Mobile Deutschland GmbH , April 2009: The German giant's 40 million mobile customers were cut off for about four hours following a HLR system crash. The vendor supplier, NSN, held up its hands and issued an apology, a move not only to be applauded but one that would have helped T-Mobile get to grips with the fallout. (See Server Glitch Crashes T-Mobile Network.)

    The good news is that there isn't a major danger of a service outage every time subscriber database software needs to be updated. "Regular HLR software upgrades happen all the time and they shouldn't be the cause of any problems," says Heavy Reading Senior Analyst Jim Hodges. "The major issues tend to come when there is a hardware upgrade or an operator is moving from a HLR to a HSS, or sometimes when there are a lot of new features in a software upgrade."

    Those sorts of major upgrades are likely to become more commonplace as operators upgrade their packet core and subscriber data systems capabilities to support LTE services, so it's a safe bet that further reports of major mobile service outages will reach us during the next few years.

  • Soupafly
    User Ranking
    Thursday October 18, 2012 8:51:27 AM
    no ratings

    This is indeed an embarassing and serious blow to Ericsson. You can bet your a$$ that AlcaLu, NSN, Huawei, ZTE, etc will all be using this as a PR weapon to beat them with.

    The shambles in NZ with Alu RAN platform (or a "faulty" server as attributed) was a PR disaster for their RAN team & caused goodwill & reputational damage. T

    The fact that this happened twice in a short space of time in a Tier1 operator/account with a piece of equipment that is critical to 3G/LTE networks rubs salt into the wounds.

    If I was a betting man, I would hazard that this was a combination of;

    a) Redback Networks. The new server smacks of their tech.

    b) A software issue/ bug.

    The Ericsson statement, as it stands, is just PR spun nonsense.

    It fails to define (in anyway whatsoever) what might have occured.

    If I was taking a wild guess I would suggest a replication bug that spread across the active/active nodes and lead to a DB corruption. The ripple was not detected quickly enough or propgated quicker than expected and the failback didn't happen, as originally modelled.

    When your dealing with an active network with millions of active users & connections, any fail scenario that involves live data and has a revenue authentication and assurance function is the stuff of nightmares! Believe me!

     

    Michelle Donegan
    User Ranking
    Wednesday October 17, 2012 2:21:48 PM
    no ratings

    I updated the story with a statement from Ericsson.



    The blogs and comments are the opinions only of the writers and do not reflect the views of Light Reading. They are no substitute for your own research and should not be relied upon for trading or any other purpose.