On Wed, Apr 8, 2015 at 4:50 PM, Matthew Dillon <[email protected]> wrote:
> I'll try to keep the document more up-to-date. I added a section at > the top showing its current status. Good progress is being made and, > in fact, just using it as a normal one-off filesystem works pretty > well, > with some caveats. But it is not ready for prime time yet, and it > won't > be until I am happy with basic multi-master operation and have working > high-level network protocols for remote access with cache coherency. > > Also, I am still making minor adjustments to the media format every > week, and anyone trying to use it seriously now will break pretty > quickly. > > There are lots of things on the roadmap, the basic roadmap for my > work right now is: > > - error handling > - synchronization to out-of-sync masters and slaves (w/local > storage) > - multi-master with local storage > > Once I have it working well with local storage I will start on the > higher-level network protocols which themselves are quite complex, > particularly when dealing with cache coherency issues. > > There are also going to be some significant personal interruptions > this year that will slow the work down. > > -- > > In anycase, I don't think the openbsd dev actually took any sort of > serious potshot at H2, the slide show was mostly talking about > OpenBSD's own forward-development issues. DragonFly's forward > development issues are in very good shape, mostly owing to the > hard work of DFly developers on major subsystems over the last decade > and very strong porting work in other areas that leverages stuff from > linux, FreeBSD, OpenBSD, and NetBSD. > > I like to think of H2 as the capper, so I am going to do it right. > > -Matt > Matthew Dillon > <[email protected]> > > Are you using a formal testing system like kyua , Jenkins used in FreeBSD ? Thank you very much . Mehmet Erol Sanliturk
