Thanks John. That makes more sense.
John Kalucki wrote:
The wording wasn't accurate and has since been changed. They're going
to be poking around with network and monitoring traffic, and there's
always a very small chance that something will go wrong, a link will
drop, and it'll take a few seconds for the network topology to
re-converge. Usually we hardly take note of, or can even detect, a
partition, due to the low convergency latency. However, when we're
running the backlinks so hot, anything could happen. They're just
trying to communicate the somewhat elevated risk of this measurement
that is only useful during a peak event and is necessary to add
appropriate capacity and balance load.
On Wed, Jun 16, 2010 at 6:44 AM, BJ Weschke <bwesc...@btwtech.com> wrote:
A "planned" maintenance from 9-10:30A EDT and during one of the World Cup
games? Wow. I hope it's really important.
Taylor Singletary wrote:
A little late notice, but just wanted to make sure you've all seen what
was posted on the Twitter status blog a bit earlier:
We’ll be working with our network provider to perform some tests
and maintenance. During this time you can expect a high rate of
June 16th @ 6-7:30 AM PDT
Twitter.com and* Api.twitter.com <http://Api.twitter.com>*
*Might be a bit bumpy this morning, but intentionally so.*
Developer Advocate, Twitter