A packet trace isn't going to tell us anything that we don't already know. The API is occasionally underwater between about 13:30 and 18:30 UTC on weekdays and we're working on fixing this.
On Tue, May 11, 2010 at 7:51 AM, Dewald Pretorius <dpr...@gmail.com> wrote: > Glen, > > My system makes thousands of outgoing calls per minute, and a lot of > them are going to non-Twitter destinations. Even if it wouldn't mess > with my app, I'm not sending a TCP dump/log anywhere. There's nothing > inappropriate in there, but calls, destinations, and volume of calls > that my system makes are proprietary business information. > > On May 11, 11:34 am, glenn gillen <gl...@rubypond.com> wrote: >> On May 11, 3:21 pm, Dewald Pretorius <dpr...@gmail.com> wrote: >> >> > As well (I know it has been discussed elsewhere), I am consistently >> > seeing API response times in excess of 3 seconds per call. It is >> > actually rare to see one that takes less than 2 seconds. This is on >> > connections from Dallas. But, even using twitter.com in the web >> > browser from Canada is also painfully slow. >> >> Dewald, >> >> I can't provide any constructive feedback to your problem, but I was >> wondering if you had any input on this message I posted earlier >> today:http://groups.google.com/group/twitter-development-talk/msg/27ea7a163... >> >> Also, having had to play the tech support role for a while in a >> previous life, is running tcpdump on a live server for 30-60 seconds >> really going to mess with your app in any noticeable way? The majority >> of the time when people ask for that kind of information it's not to >> satisfy their sadomasochistic desires to trawl through network dumps, >> it's to make their life easier while trying to identify/recreate and >> fix your problem. >> >> -- >> Glenn Gillenhttp://glenngillen.com/ >
