On Jan 26, 2:32 pm, Ben Metz <
ben.m...@t8webware.com> wrote:
> Yes. I am concerned. One answer to the problem is persistent mailboxes,
> which Jonas is working on right now. You'll have your choice of persistent
> back-ends(basic Journal(file-system), BeanStalk, Redis, Mongo, ZooKeeper,
> etc...). It remains to be seen if that will be in OS version or
> commercial.
>
> Event Sourcing, *can be helpful* for this problem as well...
> That being said, the problem makes you think hard about architecture and how
> to bring down(and bring back up) certain parts of your system.
>
>
>
>
>
>
>
> On Wed, Jan 26, 2011 at 1:19 PM, Zach Cox <
zcox...@gmail.com> wrote:
> > Wow those sbt plugins are new to me and look great, thanks for passing
> > along!
>
> > One other related question: when you "restart processes" are you
> > concerned at all about losing messages that were in actor mailboxes in
> > ram before the restart?
>
> > On Wed, Jan 26, 2011 at 12:03 PM, Ben Metz <
ben.m...@t8webware.com> wrote:
> > > Thx. Really enjoyed your Lift prez as well... I might have caught the
> > Lift
> > > bug.
> > > We are currently "deploying fat jars" :) So no container. Wade put this
> > > exact question to Coda Hale (Yammer) and Alex Payne(BankSimple/Twitter)
> > > about deploying Scala when we were at OSCON last year. They both said
> > just
> > > "deploy fat jars, and restart processes".