Thanks, guys!
On Fri, Dec 20, 2013 at 6:06 PM, James Xu <[email protected]> wrote: > Yeah, it is a bug. can open an Pull Request Kang? > > On 2013年12月21日, at 上午2:20, 肖康(Kang Xiao) <[email protected]> wrote: > > It seems that PR #547 is modified before it's merged to upstream. @nathan, > is it a bug? > > > https://github.com/apache/incubator-storm/commits/master/storm-core/src/clj/backtype/storm/timer.clj > > > On Sat, Dec 21, 2013 at 2:08 AM, 肖康(Kang Xiao) <[email protected]>wrote: > >> Yes, it's truncated. The code is as follows. >> >> (* 1000 (long delay-secs))) >> >> https://github.com/apache/incubator-storm/blob/master/storm-core/src/clj/backtype/storm/timer.clj#L85 >> >> >> On Fri, Dec 20, 2013 at 9:11 AM, Antonio Verardi <[email protected]>wrote: >> >>> Hi, >>> >>> I have already asked this question in the dev mailing list, but no one >>> answered. Maybe I'll be more lucky here. Sorry for the spam. >>> >>> I noticed that Storm 0.9.0 allows to set tick tuple frequencies at the >>> millisecond level: >>> https://github.com/nathanmarz/storm/pull/547 >>> >>> However, I tried to use the feature, but, even though the config seems >>> to show the floating point value I specified (3.9 for example), the tick >>> tuple seems to be generated every truncate(float_value) (in this case, >>> every 3 seconds). >>> >>> Does anyone have experienced the same issue? Should I open an issue on >>> Apache JIRA? >>> >> >> >> >> -- >> Best Regards! >> >> 肖康(Kang Xiao,<[email protected]>) >> Distributed Software Engineer >> > > > > -- > Best Regards! > > 肖康(Kang Xiao,<[email protected]>) > Distributed Software Engineer > > >
