back to article Telstra's confession to DNS-messin' explains broadband borkage

Australia's dominant carrier, Telstra, has revealed that a botched fix for a DNS service is the reason for a week-long outage that has taken some of its broadband customers offline. The company has explained that a seemingly-innocuous software update to Domain Name Servers (DNS) caused some of them fail and Telstra's DNS …

  1. glen waverley
    Pint

    ABC radio maybe read El Reg?

    Was listening to ABC Radio flagship current affairs program PM on fri nite. After fairly serious item re outage, quotes from senior Telstra manager apologising, free modems on their way to subscribers, reasonable sounding explanation of cause of outage, all in all a line and length ABC report, reporter wraps up with this, as best as I can remember it.

    "This is what is called a total inability to support usual processing, with a suitable acronym, (pause) TITSUP."

    Good one Lucy Carter.

  2. Phil Kingston

    Might be a good year to be in the change management training/simulation/lab-building/kit-supplying/report-writing sector.

  3. Tannin

    Why were Telstra modems and routers resetting themselves at all? Surely this is not something that a real router (as opposed to those strange, firmware-hacked things Telstra uses) should be doing without a human's say-so?

    Edit: I've been experiencing very strange intermittent DNS problems this last week or two on my (brand new) Telstra NBN connection. Possibly this is connected. (Not a new connection as such, it's just been switched to Telstra from Internode. When it actually works, by the way, it's a lot faster. iiNet's backhaul arrangements have never recovered from their ill-advised free Netfix blunder. Slow as toffee every evening peak. Telstra is vastly faster. But it has DNS problems. Should I muck about looking for a workaround? Or just assume that they will figure it out eventually and do nothing? Doing nothing is always tempting.)

  4. mrrockvagas

    Something does not add up

    This goes someway to explain the problems that have been seen in the past week, but not all.

    I run a MSP business in Queensland and have had multiple customers affected by this outage (including one my remote offices). None of these adsl connections use (and never will) Telstra supplied modems.

    Spending 3 hours on hold with Telstra on Tuesday only highlighted the extent of the problem and when they finally agreed to send out a technician, he was a much in the dark as the rest of the population.

    The common factor that was happening will all connections was a large amount of packet loss. As times up to 40%. The technician who visited onsite confirmed that this is what he was seeing on other jobs he was attending as well. Packet loss cannot be atributed ot dns lokup failures or bad telstra modems.

  5. rsole

    Telstra DNS issues explanation does not add up

    Like others the Telstra explanation does not add up. I have been experiencing DNS resolution issues at home for around two weeks. Generally trying again a few minutes later, the DNS resolves, but sometimes this does not happen until half an hour later. This has been very frustrating.

    It seems to be mainly happening around 6-9 pm AEST and happens on all devices. It seemed better last night but I found it still happened occasionally. I am not on the NBN but on Cable and have suspected this may have something to do with network or server congestion.

    Can someone please explain how this issue caused modems and routers to reset - what is that all about? Maybe they should just admit they are still working on the problem and will update us with an explanation when they know what the problems are.

    In the recent mobile network issues the Telstra recompense was completely useless and frankly quite patronising.

  6. mark 177
    FAIL

    So many Telstra customers!

    "About 370,000 thousand customers copped that outage"

    That's impressive, given the population of Australia is only about "22,000 thousand"

  7. Trixr

    Nice "explanation", Telstra

    Surely there was more than one DNS server configured, with each modem configured with a primary and a secondary? And surely they should have planned to update the secondaries and verifying those before updating the primaries?

    Also, no matter now many modem restarts it does (why would it recycle just due to bad DNS?), why would they "brick"?

    Definitely more to this story.

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon

Other stories you might like