Hi Matt, Could you also let us know how you increment tweet numbers. It is not constantly linearly incremented from 1, is it. That would mean there are over 2 billion tweets in the system, which isn't.
On Sat, Jun 13, 2009 at 1:52 AM, Matt Sanford <[email protected]> wrote: > > Hi there, > > That is indeed what I meant. We are planning to skip some ids to force > the 2^32 change during business hours. Twitter itself should be fine but I > originally announced this to the list so people could make sure they'll also > be fine. There is no change to the format of responses and the number will > continue to grow upward. This was just fair warning that you might have used > the Rails default definition (or some other method) that relies on signed > 32-bit integers. > The 'decide to do this' part is deciding to do this now by skipping ids > rather than let it occur naturally 12 hours from now when people have been > up for 24-hours and might not be at their best. Let's not allow the > 'insulting' vagueness devolve into insulting tone, please. We're working on > co-ordinating internally to do this at 21:00 GMT but like all things > involving groups of people we may run a little late. Sometime after 21:00 > GMT this is still planned. We'll update @twitterapi when the exact time > comes. > > Thanks; > – Matt Sanford / @mzsanford > Twitter Dev > > On Jun 12, 2009, at 1:06 PM, Stuart wrote: > > >> 2009/6/12 J. Adam Moore <[email protected]>: >> >>> >>> So do I just allocate as many bits as I can in my database to the id >>> field and hope that it doesn't ever run out? I'm confused why you just >>> announced that. Okay, so an overflow is happening. Is that your fault? >>> Is this fixable on your end, my end. Is this just for people who are >>> using 32-bit signed ints to store ids? Decide to do what? Roll it over >>> like an odometer or increase the field size? Forgive me for being an >>> idiot, but 'decide to do this' is just about vague enough to be >>> insulting. I was happily under the assumption that this problem was >>> considered long, long ago when the field size was initially chosen by >>> who I also assumed to be smart people. >>> >> >> I read it as "we're considering skipping a bunch of IDs so we hit the >> limit during today rather than sometime over the weekend. That way >> there will be people at Twitter able to react to support issues that >> might arise. >> >> As for what developers should do I think it's pretty obvious. If >> you're using a signed 32-bit integer to store tweet IDs you need to >> change that ASAP because judgement day is coming!!! >> >> -Stuart >> >> -- >> http://stut.net/projects/twitter >> >> On Jun 12, 10:23 am, Matt Sanford <[email protected]> wrote: >>> >>>> Hi all, >>>> >>>> The overflow of the 32-bit signed integer value for status ids >>>> (a.k.a "The Twitpocalypse" [1]) is fast approaching. The current >>>> estimate is around tomorrow at around 11am GMT, or 3:00am Pacific time >>>> in the case of Twitter. There is some discussion internally about >>>> accelerating things so we'll be in the office and able to cope. Nobody >>>> is their freshest at 3:00am, not to mention it would be nice to not >>>> have apps broken throughout the weekend if one-person developer teams >>>> don't notice. No decision has been made yet but I wanted to get >>>> something out to you all so you know what's going on in the event we >>>> decide to do this. >>>> >>>> Thanks; >>>> – Matt Sanford / @mzsanford >>>> Twitter Dev >>>> >>>> [1] -http://www.twitpocalypse.com/ >>>> >>> >>> > -- Regards, Lakshman becomingguru.com lakshmanprasad.com
