Sorry, I took that mainly from your own words:
I got it. My bad. The fact that English is not my native tongue clearly shows. What I meant by "dead end" is that no new features are being planned/added to the HAMMER 1 or even possible due to B-trees design limitations. But this is also true for example for FFS on OpenBSD. HAMMER1 bugs are fixed regularly when found. Check out for yourself
http://gitweb.dragonflybsd.org/dragonfly.git
Is v2 supposed to replace the original or is v1 being kept as a complete, stable file system? How much effort is being put into ongoing support for v1 now the devs have moved their attention onto a completely new version?
I am not sure that anybody knows what future holds for HAMMER 1 once HAMMER 2 get released. HAMMER 2 is meant to be complete, stable separate file system. It has well defined list of objectives.
http://leaf.dragonflybsd.org/mailarchive/users/2012-02/msg00020.html
Project is significantly behind original schedule so I don't want to think about HAMMER 2.
I recall vividly that Pawel Jakub Dawidek publicly expressed his doubts that Matt Dillan can pull something like HAMMER 1 and write a file system on his own without corporate backing. He did and I hope he will do it one more time with HAMMER 2. Matt and me are people of about the same age and definitely closer to the end of our lives than to the beggining. I learned C programming using his C compiler on Amiga. He was 16 and I was 16. I really want him to pull this one
I may have taken this thread off topic slightly. My original concern was just that there were a lot of unfounded plus points for HAMMER ("amazing performance"/"beats the pants of everything",etc). Half the ZFS down points were also unfounded or irrelevant and peoples attempting to right them were either asked for solid proof* or outright called liars.
*Fair enough but I don't see much proof for a lot of the HAMMER claims.
I am by no stretch of imagination serious expert on file systems and I doubt you will find such people hanging here. I started a tread out of my personal frustration for the lack of any compassion between these two file systems and my hope that other users like myself who have been exposed to both file systems will fill in missing pieces. This forum is probably the only place where you can find people who have been exposed to both file systems.
I have no idea why people took so hard for example my statement that ZFS needs lots of RAM. I am having hard time to see why people have to defend ZFS so hard or their defend their choice to use ZFS. The coolest thing about HAMMER and DF is that it is labor of love. Matt got rich during dot com boom. He and handful of other like minded guys are hacking on that in their spare time. As a curios person I was always fascinated what they were doing and try using their labor of love at work. It didn't go quite the way I wanted first time but I am sure I will try again. Since my kids have to eat every day regardless whether Monit works or doesn't on DF I use enterprise tested technology ZFS+FreeBSD at my workplace. ZFS is pretty good you know. It is better than playing with XFS and mdadm.
DF guys are not selling anything, they are not trying to compete with Oracle or with FreeBSD for that matter. People who think that FreeBSD and ZFS are the best thing after slice of bread should ignore this thread. People who work for large server farms probably should ignore this thread as well. People who are too serious forum posts probably should ignore it as well.
That leaves me with a targeted audience of like minded geeks, who suffered cabin fever like myself hopefully from different reasons than mine (I had terrible Bronchitis over past 10 days with prevented me from skiing with my children over the holidays).
